Customer demand forcing move from .NET to J2EE

Discussions

News: Customer demand forcing move from .NET to J2EE

  1. Customer demand forcing move from .NET to J2EE (275 messages)

    San Juan has linked to a piece which discusses scenarios in which companies that are using .NET, have had enough demand that they ported solutions to Java. In one example, Infogate Online Ltd had to port its dotnet application in c-sharp to Java based on the demands of an Asian customer.
    Infogate describes OnDema as a multi-tier distributed Web application that manages the distribution of TV channels, video on demand, pay-per-view and games on demand directly to end users. The application uses ASP.NET and ADO.NET and contains approximately 500,000 lines of C# code. But last year a customer in the Far East requested a J2EE version. Not wanting to be cut off from the Java half of the world, Infogate wanted to move in that direction.
    Read the original article: Global IPTV .NET software provider opens channel to Java

    Read San Juan's posting: Customer demand forcing move from dotnet to J2EE

    Threaded Messages (275)

  2. Not so sure...[ Go to top ]

    With help from Mainsoft consulting services, Fogel was able to move the .NET application to Tomcat 5 running on Sun Solaris 8. As Mainsoft describes the project "less than one percent of the C# code had to be modified to overcome inherent differences between .NET and J2EE code and to integrate with third-party components required by the application."

    I am not a believer of this much automation. How do you use best of breed tools and apis such as spring and hibernate with this automation. This is one shortcut I would avoid. It would decrease the time to convert, but the time to maintain would be ten fold.
  3. Me neither...[ Go to top ]

    Good point Matt - I hadn't thought of it quite that way. I'm guessing that either all of their business logic was embedded in stored procs or their whole DAO layer was hand-rolled, neither of which are desirable in lots and lots of ways.

    For even more reasons I still remain skeptical that there are more than a handful of cases where it's justifiable to convert a fully constructed app written in C#.NET to J2EE or vice-versa. I doubt this is a meaningful trend.
  4. But last year a customer in the Far East requested a J2EE version. Not wanting to be cut off from the Java half of the world, Infogate wanted to move in that direction.
    How does moving "in that direction" translates into moving from .NET to Java? From what is written, they ported .Net app to Java, keeping both codebases for different customers. I can understand a desire to have only one Web server and only one dynamic content controller (servlet or ASP.NET), if an organization already has one of these.
  5. It sill is C#[ Go to top ]

    They are using this product :

    http://www.mainwin.com/products/vmw_j2ee.html

    So using this product, C# code runs on a JVM.
  6. This is all bad ..[ Go to top ]

    The application uses ASP.NET and ADO.NET and contains approximately 500,000 lines of C# code.
    ......
    less than one percent of the C# code had to be modified

    If I have an application in struts + EJB of that magnitude and if I need to port it to JSF/Spring/Tapestry + Hibernate I guess it'll take a lot more effort than what it took them to convert from .NET to JAVA.

    News like this do not present the whole picture. I believe the job of a respected site like TSS is to do indepth analysis and to come out with real facts and figures. Sadly on this account the analysis and facts were seriously lacking.
  7. This is all bad ..[ Go to top ]

    If I have an application in struts + EJB of that magnitude and if I need to port it to JSF/Spring/Tapestry + Hibernate I guess it'll take a lot more effort than what it took them to convert from .NET to JAVA.News like this do not present the whole picture.

    why in the hell you may need to that ,portin from Struts+EJB -> Spring+Hibernate??
    In case you wish to do that lines of code 'n click is not that much compare to .NET migration.
    in former u shiftin Frameworks in later u migratin from one Platform to other!
  8. This is a valid trend[ Go to top ]

    I personally believe this is a valid trend. There seems to be the notion that because of the newness of .NET or simply the facade of invincibility that Microsoft projects (which, nevertheless, has been roundly - and ultimate? - damaged the last few years), that there is no market for switching from .NET to J2EE, or at least moving away from a .NET-only focus.

    The openness of Java tools, the inexpensiveness of its implementations, and the robustness and class of the enterprise class products from IBM, Sun, BEA, ORACLE, and others, means that there is a very large market for J2EE in the consumer and enterprise markets. Companies that try to FORCE .NET on its consumers will not only limit itself to a segment of the market, but might quickly find itself at a disadvantage vis a vis competitors that can provide the same application at a much lower cost (due to the diversity of open Java implmentations).

    It is a sign of the times that I hear so much talk of interop from Microsoft between J2EE and .NET and much less commitment from the Java side.
  9. This is a valid trend[ Go to top ]

    .It is a sign of the times that I hear so much talk of interop from Microsoft between J2EE and .NET and much less commitment from the Java side.

    I think this a very good point, I also was suprised by the number of J2EE/SOAP/Java interop articles over at msdn.microsoft.com - some of them were actually quite good, well presented, informative and made no mention of MS specific technologies!

    Its also interesting that MS appears to be "borrowing" bits from both Java and open source projects. Reading about the latest .NET sdk, I came across the notion of a "lightweight" SOAP server for developers that did not require (Gasp) a web server... just like Apache Axis has been doing for quite some time.

    I also read further that this breakthrough innovation was made possible by a new fangled kernel HTTP (http.sys) server process, pretty much exactly the same as Linux introduced in kernel 2.4 a few years back.

    No doubt technology flows both ways,... indeed in a few years time I ought to be able to sell my java skills to Microsoft shops because .Net and Java will converge even more when Microsoft realises that convergence is the only way it can tap the very large pool of Java Developers :-)

    J.
  10. This is a valid trend[ Go to top ]

    It proves nothing but I am unfortunately seeing the opposite trend (increasing “interest” in / use of .NET) in many of my clients (mostly large financial organisations).

    They quote reasons including better productivity / development cost and simpler integration using one (MS) architecture.

    It’s depressing.
  11. I dunno[ Go to top ]

    I dunno....personally, I've known a couple of corporations (outsourcing software IT shops in my country) that have completely moved from offering a .NET-based framework to a J2EE based one...I kid you not, and I'm not making this up for propaganda reasons (honest!).

    The reason is that they had embraced .NET in the beginning, when the hype was greatest, then realized the market for those skills was not actually there (or let's be generous and say, the market was not "mature"). Thus, the switcheroo.

    I'm a firm believer in the reality of the market, and the really bad thing for .NET was that it rose at a time when IT spending was not expanding at the rapid pace of the dotbomb years. NET will definitely grow, I've never said it won't, but mostly at the expense of shops that were already Microsoft - which explains why MIcrosoft is desperate for interop - they realize that adoption of .NEt will be facilitated if customers know they can interoperate with the much more robust J2EE systems.

    You're probably seeing a turning point here - there's a reason why Microsoft stock has been dead in the water for so long, EVEN after the infusion of billions of dollars from the company's cash hoard. Its cash cows - office, windows - are not growing as fast as before, the next version of windows is long in coming, and the hope of .NET as its salvation is probably shot in the water as well (mixed metaphors - oh my!), given the fact it never lived up to the hype (witness the slowly dying Passport, gutted by the Liberty Alliance), and never generated the rush of developers into its arms that Microsoft hoped for (although mono might change that in a perverse way).



    People always used to say Microsoft won by outlasting its competitors...unfortunately for the company, open source and Java are not single entities that measure their life spans in years, but are a very diverse, competitive, and energized "ecosystem" that will probably be here to stay for the next decade or more.
  12. What I can see ......[ Go to top ]

    It seems Java has already had its heyday. The one for .NET is certainly coming.

    I am a long-time Java programmer and is it sad to see that past two years were kind of 'me-too' for Java. The XML binding, serialization, web services, native-level interoperability, generics, etc. All those things flashed in early 2000 and then got stalled in development for almost five years. When .NET kicked the collective ass of Sun, IBM and BEA we have seen some progress. But innovative solutions like Spring, HiveMind, Hibernate and Tapestry still remain ignored, if not actively fought by Sun.

    The trend I can see is that Java is slowly moving from a leading platform/language position into the Yet Another one. It has its pluses and minuses so companies will go back and forth. But the herd move to Java might be a story of distant past.

    Personally, I see the fact that insurance and banking business widely adopts Java at this time, as a kind of confirmation. Those companies always stay on the trailing side of adoption curve due to their conservative technology policies. What Java replaces there is OS/2, mainframe and proprietary client/server solutions.

    As a firm believer in market reality, I am learning .NET not to stay with outdated skills ....
  13. Well, it's always a good thing to be learning new things ;-)

    HOWEVER, I'm gonna point out that Java is not exactly past its heydey...in fact, Java's growth right now is MUCH MUCH faster than .NET, and it is growing in a much larger array of niches.

    How is this so?

    If you read Russell Beattie, you'll note his infatuation with mobile devices...well, Java right now is the leading contender in this very fast growing area. It is in the majority of new cellphones, for example, and has increased 100% in one year and is now deployed in 750 MILLION smartcards around the world, helping such organizations as the USA army and entire nations like Taiwan, whose national ID cards are Java-based. .NET or any MS technology is not even close in this newest, most RAD and INNOVATIVE area. In fact, the limited access of these in such a wide arena is what ultimately will doom it to just another ho-hum desktop/server language/platform like C/C++/VB

    I'll even put forward a thought about Java on the server side/desktop...

    I'll postulate for you that the health of a platform can also be described by the activity (competition==innovation!) surrounding the CORE supporters (ie. Sun) of that platform. In this sense, the frenetic pace of competition and innovation going on in Javaland is an accurate gauge of the health and future direction of the platform.

    You may think having so many competing choices is bad - as opposed to the "dictatorship" of a single monopoly - but there is a reason why NATURE itself uses competition to optimize species, and why very competitive capitalistic systems have won out against overly centralized and cooperative communistic systems.

    In this sense, I LOVE the arguments, fight, competition in javaland - I love the fact there are so many choices, and that people are "fighting" against each other to try to get THEIR framework or tool or system as the best in town! I LOVE that there is so much to BILE about (which explains Hani's popularity!)

    remember the treadmill arguments of many Microsoft-centered programmers, who are always trying to keep up with the new tech thrown at them and the deliberate "killing off" of the tech that they spent so much time learning...

    I've been a java programmer since 1995/96, and I'll be a java programmer (albeit in managerial positions) long after most of .NET is consigned to the dustbins of IT history, just as VB/DNA/J++/etc and all the other MSFT technologies fell after their short life spans.

    And guess what? The programs i write in java today (and wrote in 1996), will still probably run and remain useful!
  14. the end of the browser[ Go to top ]

    First you have to have the language in place. To release .NET was a major undertaking, more expensive than sending man to moon according to Bill Gates. Then you decide what do with it.

    So with everything in place, what are MS now going to do with it? Whatever it is, that is what you should brace yourself against. Notice that MS did not copy J2EE.

    It is no secret, XAML is the final icing on the cake that will make .NET the major enterprise platform for the next 20 years of so (just as long as C/C++ ruled). And don't argue that it lies a long time in the future. The time to learn C#/XAML is now, start to build up your libraries and routines if you want to stay "hot".

    It will not mean the end of the browser for the great public mainstream, but it certainly will mean the end of the browser in enterprise applications area. It is futile to think that cellphones and handhelds will stand against this avalanche.

    http://theserverside.com/news/thread.tss?thread_id=30730#150569

    Regards
    Rolf Tollerud
  15. the end of the browser[ Go to top ]

    Rolf... like for the apache thread, ur sayings are ridiculous. You just say things and ask us not to doubt of it.
    MS did copy J2EE like J2EE copied other stuff... everybody learns from another one !

    Peoples i know using dotnet are knowing heavily using nhibernate and nant .. and they plant to see spring.. so please, stop.

    And I don't agree with you, both platforms will survive, together... along. That's life.

    I think J2EE will receive quite lots of help from Linux as dotnet don't run on linux ( please, mono is 18 months late on microsoft and I don't know anybody using it in production except novell ).
  16. the end of the browser[ Go to top ]

    First you have to have the language in place. To release .NET was a major undertaking, more expensive than sending man to moon according to Bill Gates. Then you decide what do with it.

    Send Bill Gates to the moon?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  17. it is contagious - it is the future

    I would love to discuss with you but unfortunately it is impossible, you being too emotional. For all the others that don't see the significance of XAML and discern that Microsoft is trying to obsolete the browser - just wait and see.

    I worked in this business in 20 year, but never have I had some much fun and rewards before as with this new technology.
  18. it is contagious - it is the future[ Go to top ]

    ... Microsoft is trying to obsolete the browser - ...

    I agree that MS is trying to obselete the browser. At least for applications. And I am happy that they are. What a pain. But MS is doing it for survival reasons not because it is better.

    As for Cameron being emotional - yes, laughter is an emotion.
  19. Too late[ Go to top ]

    ROTFLOL! I think you were born a little too late...

    A few points:

    (1) Unlike Java, which is pushed by a lot of corporations and organizations (ie., it has many stakeholders), .NET is for all practical purposes pushed strongly by one corporation - Microsoft.
     
    (2) major corporations too have their "lifespan" as the dominant force in an industry, and Microsoft's time as THE dominant force in software has already passed, so I'm not too sure about that "20 years from now" prediction you just made, especially sice Microsoft hasa history of pulling the rug from under its developers every often or so in order to rush out "the next big thing".

    Why do i say Microsoft's time has passed? Because when all software represented was the desktop (for the mass consumer and corporations), where it has a monopoly, Microsoft was untouchable. Now that software means not just the desktop, Microsoft has become just another force in the software environment, albeit one of the biggest "forces".

    This was why I was looking with interest at Microsoft's attempts to extend its dominance BEYOND the desktop (the "caged tiger" hypothesis)...unfortunately for it, it has not been able to monopolize any other major area, and I believe this is the crux whereupon everything turns (if Microsoft had managed to dominate these other software arenas as it has the desktop, then believe me, I would be sacrificing my firstborn to the Microsoft "God" already):

    1. On the server, Linux/J2EE/Open source LAMP has been thwarthing its reach for a monopoly, and in fact gaining ground on any Microsoft solution, inc ASP.

    2. On the PDA, Palm, a minute company with resources only a tny fraction of Microsoft's has held off Microsoft for years (even though commentators in the 1990s had said Windows would overwhelm it within months of Windows CE's introduction - so much for "experts"). Now, the PDA market is being marginalized by other mobile smart devices, although Java-based "PDAs" wih specialized functionality like Blackberry have exploded.

    3. On smartcards, Microsoft twice tried to dominate the market in the 1990s, but was roundly beaten by Java both times, with Java now running on 750 million cards worldwide, powering id cards used by entire nations and organizations like the US army.

    4. On mobile phones, microsoft was again thwarted in its attempt to dominate the market, and here again Java shines.

    Etc, etc...
  20. Too late[ Go to top ]

    San Juan,
       .NET may be pushed by only M$, but take a look at what's happening on the Java side -

    (a) Corporations are now in a dilemma with EJB 3.0 - if they implement a solution today, they will be supported as legacy in no time.

    (b) J2EE is meant for very large scale enterprise apps. I'm not sure people understand that and want to use it for your regular size app (90%+ of apps built fall in this category). For simple 2-tier apps, the combination of jsp/servlet/struts doesn't help with productivity compared to asp.net

    (c) If j2ee/jsp/servlet is more mature, how come M$ has a vibrant and lucrative component market for asp.net? Few things compare to the offerings of infragistics/component one/component source on the java side for web development.
  21. Too late[ Go to top ]

    San Juan,   .NET may be pushed by only M$, but take a look at what's happening on the Java side -(a) Corporations are now in a dilemma with EJB 3.0 - if they implement a solution today, they will be supported as legacy in no time.

    You do realise that EJB3.0 is backwardly compatible?
    (b) J2EE is meant for very large scale enterprise apps. I'm not sure people understand that and want to use it for your regular size app (90%+ of apps built fall in this category).

    Yes. We use it because it works very well. Considering the huge number of Tomcat deployments, you seem to have a minority opinion here.
    For simple 2-tier apps, the combination of jsp/servlet/struts doesn't help with productivity compared to asp.net(c)

    Coding speed is not everything.
    If j2ee/jsp/servlet is more mature, how come M$ has a vibrant and lucrative component market for asp.net?

    How is the component market for asp.net in any way related to the maturity of j2ee? The component aspect is very recent in j2ee, with the arrival of JSF, and the use of components is a very minor part of the infrastructure of a serious web application.
    Few things compare to the offerings of infragistics/component one/component source on the java side for web development.

    Who wants one source for everything? The Java side encourages freedom of choice and innovation.

    Also, all this is of no consequence if I am one of the majority who want to deploy my app on a non-Microsoft platform.
  22. Too late[ Go to top ]

    and what does it mean "serious web application" ?

    Dmitry
    http://www.servletsuite.com
  23. Before you flame me, check who said it:
    http://news.com.com/jboss+eyes+front-end+technologies/2100-1007_3-5487360.html?part=

    Another nice quote from the article:
    Now that .Net has come along, which is easy to code with, much is needed to simplify Java

    And you know what I think? Only with clear assessments like this is that a product can really start enhancing. Perhaps this is one of the reasons that JBoss came to the forefront of J2EE servers.
  24. JBOSS[ Go to top ]

    "Perhaps this is one of the reasons that JBoss came to the forefront of J2EE servers."

    Perhaps it's because it's cheap? LOL. I'm gonna want to have a ringside seat when Apache comes out swinging...

    I have nothing against open source actually - it's causing havoc with microsoft because it's basically using the same weapon against microsoft that microsoft used against other companies (sell very inexpensive products to chip away at the market leader, or take the market completely away by giving stuff for free)
  25. Before you flame me, check who said it:http://news.com.com/jboss+eyes+front-end+technologies/2100-1007_3-5487360.html?part=Another nice quote from the article:
    Now that .Net has come along, which is easy to code with, much is needed to simplify Java
    And you know what I think? Only with clear assessments like this is that a product can really start enhancing. Perhaps this is one of the reasons that JBoss came to the forefront of J2EE servers.

    They are leaving out some pretty good Java UI frameworks when they say doing Java web UIs are tough. This article is more about JBoss trying to get service contracts based on the industry Java FUD. (not blaming JBoss for trying to capitalize on it).

    I am and have been using VS.Net daily. It is not that easy to "code" with. Some things are pretty easy. But usually they don't involve coding.
  26. Too late[ Go to top ]

    -(a) Corporations are now in a dilemma with EJB 3.0 - if they implement a solution today, they will be supported as legacy in no time.
    So will the .Net solutions. New version of SQL Server coming. New version of Windows coming. New version of Exchange coming. New version of .Net coming. New persistance mechanism coming. What ever will replace .Net is coming. ...
    (b) J2EE is meant for very large scale enterprise apps.
    Is it? Maybe what the big vendors are pushing is. But some much in J2EE(Java) isn't. What J2EE does provide is the ability to grow and change.
     I'm not sure people understand that and want to use it for your regular size app (90%+ of apps built fall in this category).
    Nothing wrong with that. I suggest you check out Spring and Echo and SwingOnWeb and ...
     For simple 2-tier apps, the combination of jsp/servlet/struts doesn't help with productivity compared to asp.net

    Agreed. But what happens when the app grows? With Java, I can upgrade all my tools/apis as needed. (Tomcat >> JBoss >> Weblogic) (Exo >> Websphere Portal) (HSQL >> Firebird >> Oracle) BTW, these are examples not saying any is better or picked any for any other reason than to be fair in mentioning.

    Also, there are many tools/APIs that can equal or exceed the initial productivity of ASP.Net. Echo and etc.
    (c) If j2ee/jsp/servlet is more mature, how come M$ has a vibrant and lucrative component market for asp.net? Few things compare to the offerings of infragistics/component one/component source on the java side for web development.

    How many databound table controls do you need? Of the ASP.Net projects I've done, not a one used 3rd party components.

    And who really cares about ASP.Net and the fabulous 3rd party web controls? According to our resident prophet (Rolf) MS is killing the browser (as we know it). :)
  27. Too late[ Go to top ]

    San Juan,   .NET may be pushed by only M$, but take a look at what's happening on the Java side -(a) Corporations are now in a dilemma with EJB 3.0 - if they implement a solution today, they will be supported as legacy in no time.(b) J2EE is meant for very large scale enterprise apps. I'm not sure people understand that and want to use it for your regular size app (90%+ of apps built fall in this category). For simple 2-tier apps, the combination of jsp/servlet/struts doesn't help with productivity compared to asp.net(c) If j2ee/jsp/servlet is more mature, how come M$ has a vibrant and lucrative component market for asp.net? Few things compare to the offerings of infragistics/component one/component source on the java side for web development.

    Sure, J2EE is complex and designed to support large-scale applications, but it's also easy to simplify J2EE for mid-scale appliations, e.g. lokk at what Spring is doing. I think your point about ASP.NET is valid, and I think JSF was designed with some of this in mind.
  28. Complex?[ Go to top ]

    Here is a blog I just posted (it be of no use to newbies)
    http://www.sandrasf.com/kiss
    Sure, J2EE is complex ...

    Parts of it are complex, and they are being redesigned.... or people are using open source alternatives. J2EE is a set of choices, some people make bad ones.
    designed to support large-scale applications

    Anything complex can't do large scale applications. It would run slow and would fall from it's own weight. (That is what I called "Marketing oriented architecture") Else large applications converting are to LAMP - a real sucess story that adds value, it's very effective. (I posted link to PHP apps in the thread on this page on Sun's performance paper)


    .V
  29. Why is there nobody that keeps a track record? It is fun to discuss and intellectual rewarding to try to make predictions but it is a little boring that nobody keep track.

    Here is just a little pick of the issues in the past.

    1) I was right that windows would out compete OS2.
    2) That IE was a better product then Netscape Navigator 3x-4x.
    3) The J2EE was a slow and cumbersome technology, heavenly overused.
    4) The .NET would get traction and be a major competitor to Java
    5) That cheap Intel-processors in most cases is a better choice that Sun hardware.
    6) That Microsoft would get the upper hand in the handheld market.

    and many more, some that is not decided yet, for example that MS will the market leader of High-End mobile phones.

    "(1) Unlike Java, which is pushed by a lot of corporations and organizations"

    Do you mean IBM, Sun, BEA, and ORACLE?
    Of the four, two of them will probably not exist in a couple of years, and Oracle is about to be transformed to be a consultant company, just like IBM. Only IBM (one of the most evil companies that ever existed) is a real threat to Microsoft.

    "Microsoft's time has passed? Because when all software represented was the desktop (for the mass consumer and corporations), where it has a monopoly, Microsoft was untouchable"

    You forget that the so called monopoly was won in fair fight and competition with OS2, Apple, Unix etc.. At the time nobody had expected or predicted the Web including Tim Berners-Lee et all. Sun was desperately trying to sell Unix workstations, a business idea that had failed, and was saved in last minute by the miracle of the Web. Certainly no by any effort or brilliance from Sun.

    But who is going to save you next time? Another miracle? Don't hold your breath. That MS survived the web and now is stronger than ever is another miracle and a tribute to Microsoft, than know how to turn the whole company around from one week to the next. And when they adapt quickly to save themselves they are accused of "pulling the rug from under its developers every often or so in order to rush out the next big thing." You can never win, can you?

    Thoroughly debunked by the competition and then suddenly easy success, not really earned.. That is the history behind Sun and Java and a key to understanding of the aggressive physiology behind the Java advocates. I leave the rest to the intelligent reader endowed with reason, logic, common sense and an understanding of the human nature.

    "commentators in the 1990s had said Windows would overwhelm it (Palm)"

    Talk about revisionist history!, at the contrary everybody laughed at Microsoft and said they had no chance against Palm and that "Windows CE was a joke".
    Fine joke it was! ;) And now when MS has out competed Palm and everybody you are saying that the PDA market is finished!

    "Java is big in the smart card market"
    Fine. Finally something that Java can do! After all, that was what Java originally was made for!
     
    So I hope you understand me when I say that I am a little tired, even if in the discussion after discussions my predictions comes out true - in the same time everything is forgotten and a new discussion is started.

    The ability to deceive themselves is great in the Java world.

    Regards
    Rolf Tollerud
  30. Loony Oracles[ Go to top ]

    Rolf, you're a total loon ;-)

    Firstly, those "predictions" you allegedly made are so generalized and obvious they're not that much of a prediction set.

    For example, NO ONE disputed .NET would be big and gain traction simply because MSFT bet EVERYTHING on it - the question was actually, would it simply cannibalize older Microsoft tech, or would it overwhelm other tech out there (plus, would it stop the move of VB shops to Java - about 30% of which are moving accg to one stat) - well, you do know what's happening, right? Where's passport now? Stopped in its tracks by the lIberty Alliance. Where are the VB shops now? Still moving, probably at the same rate, to java, especially those angry at the Microsoft "treadmill". Is there any dent at all to Java and PHP and others? Not that I can see - Java is still the top skill preferred for jobs - not even J2EE (which is most affected by .NET) really feels that much heat (witness the recent move of the NY Stock Exchange to Java, or eBay's move to J2EE).

    Also, I've always laughed whenever people point out Netscape as an example of how "unbeatable" microsoft is....by bundling IE FOR FREE with every windows PC out there, Microsoft basically killed its competitor by removing the market itself. Ironically, now Microsoft itself is being killed the same way by Open source and all the great open source Java tools out there.

    With regards to your argument about the stakeholders of Java, aren't you missing a LOT of them? The fact that you focused on simply the big commercial J2EE app servers shows just how much you DON'T know about Javaland. What about corporations like Nokia, Motorola, Schlumberger and others? What about all the open source organizations out there that have large Java commitments, including the biggest one of all - Apache?

    With regards to your Palm argument, I suggest reading the books (say, something like "Piloting Palm", which records the history of Palm)...when Microsoft introduced windows ce, nearly all the pundits said the unstoppable MIcrosoft juggernaut would quickly make mincemeat of Palm, which was a very small underfinanced company about a FRACTION the size of Microsoft. You know this is what happens everytime Microsoft decides to enter a market - everyone jumps the gun and thinks MSFT will roll over any competition. ROTFLOL.

    But don't you worry Rolf - your Microsoft God will be here for a very long time still - it just won't be as overdominant and influential as it was when mass consumers only had the monopolized desktop to work with. Java and open source will make sure of that.
  31. too much fun[ Go to top ]

    You make it sound like you're the only one who has ever doubted the useful of product X. Back when XML was first coming out, I said to my friends, "I doubt this thing will be useful or do anything remotely interesting." Boy was I ever wrong. On a daily basis, just about 3/4 of the things I use or work in involve XML or a XML-based langauge.

    Fact is, many people have said the same exact things you're claiming to have predicted. No one can predict the future. It's kinda like the psychics on the psychics friends network. If they really could predict the future, they wouldn't be answering phones and pretending to have some special power :)

    I'll keep on using and learning both technologies. When both of them are replaced by the next big movement, I'll learn that.
  32. too much fun[ Go to top ]

    I just hate it when people don’t think for themselves. Especially when there is a mobbing situation, as is the case with Microsoft. Often here in TSS there is talk about MS HYPE or FUD. But the Java/Unix camp generates 1000 times more HYPE or FUD; you don't have to be a Socrates to notice that.

    Most of this people are young and believe in what they are saying and in their good cause. But the biggest atrocities in the world are done in good faith, "in a good cause". That makes it extra revolting.

    For 3(4?) years ago Microsoft did not have the .NET, most people used Windows 98 and the whole situation looked like that MS was going to loose. Now the situation is entirely different, but the Java camp still lives in the past. "This year is going to be the year of Linux at the desktop" is not going to happen.

    I don't care of the unwashed masses, or even people like Cameron that only is interested in protecting his investment. But you Peter, seems like a reasonable man to me, an intelligent person that can think for himself, so I find it extra strange that you are preaching this J2EE stuff. "Academic Crap", as Miguel de Icaza puts it.

    Regards
    Rolf Tollerud
  33. based on experience[ Go to top ]

    I can only speak from experience. Having tried to build transaction heavy applications with both ASP, Java (J2EE/J2SE) and .NET, I have to say it can be done with just about every platform. The difference is the amount of work. For example, 2 years ago I tried to convince some people we needed a reliable ORM solution for the project, but rather than look at third party solutions, 3/4 of the team argued against it. After I left that project, those individuals made a 180 degree turn. This was after rolling their own primitive data access solution and going through the pains of learning how not to build one. I tried to find the right balance, but unfortunately only a few people would read the articles I sent out and think the problem through.

    My experience has been in a very narrow area the last few years, so much of this doesn't apply to small/medium applications. Although I wish architectural decisions were based on actual requirements, my experience is that often those in charge have no clue what the requirements are. So what ends up happening is those making the decisions "over estimate" their needs. In doing so, they are their own worse enemy.

    I see a lot of this, but by the time i get in there, the platform is set in stone. Therefore I focus on is figuring how to make "it" work. what ever "it" happens to be.
  34. it is the complexity that is relevant[ Go to top ]

    OK, just one last word.

    "My experience has been in a very narrow area the last few years, so much of this doesn't apply to small/medium applications"

    IMO when you use the description small/medium/large the characterization should refer to the complexity of the application, not the load. The question of the load is irrelevant. To think that an app is complex only because it handles a large number of transactions is naive. The most obvious example is the SAP that often does not have more than 100-200 users, and even less.

    It is better to use the terms trivial/medium/complex. As I said before, a large number of transactions can very simple be achieved by the simple remedy of stateless server.

    Regards
    Rolf Tollerud
  35. too much fun[ Go to top ]

    I just hate it when people don’t think for themselves.

    Then you won't like the Microsoft crowd at all. :0

    Not that Java is not with its non-thinkers. But MS technologies is all about letting MS do the thinking for you.
  36. too much fun[ Go to top ]

    ... or even people like Cameron that only is interested in protecting his investment.
    Hmmm. I suggest you read this - http://www.tangosol.com/coherence-uses-j.jsp
  37. Pot kettle[ Go to top ]

    "I just hate it when people don’t think for themselves."

    Kettle, pot. Pot, kettle.

    Rolf:

    You seem to be living in some paranoic reality, with hints of the twilight zone to boot. When exactly did it seem like MSFT was going to "lose" with dotnet?...in fact, the first few years of hype were the worse, with C-sharp being heralded as the "Java-killer", the editor in chief of JDJ (alan) and our fearless leader, Rick Ross, sounding like two frightened chickens in the fox-filled hen coop, crying about the demise of Java in a few years and the unstoppable rise of .NET...oh no! oh my! i is scared o' the big bad wolf!

    Now that .NET is happy just to cannibalize the older Microsoft technology, even my wife (who uses Microsoft tech everyday - she's a VB/access developer) wonders what happened to it after all the hype.
  38. more soviet revisionist history[ Go to top ]

    Very well San Juan,

    Why don't you provide me with some links about the unstoppable rise of .NET for two or 3 years ago? And when you are at it, some links about the unstoppable Microsoft juggernaut that would quickly make mincemeat of Palm?

    In the meanwhile I can provide you with hundreds of links where MS was thoroughly bashed and maked fun of.. How many do you want? 100? 200? 300?

    To deny that MS is unanimously bashed all over the world! What reality are you living in?

    Regards
    Rolf Tollerud
  39. more soviet revisionist history[ Go to top ]

    Is Microsoft the only right solution? Is it heralded to be the 'future' of all computing? Is that what you really want, Rolf?

    Are we at great peril if we do not use XAML and .NET? Are we risking to be wiped out if we do not? That is your prediction?

    Microsoft-only world would be very very bleak. We would be using IE6 in year 2050 if it were for Microsoft to decide.

    But let's see. What if XAML proves to be a dud? What if it never flies, and legacy applications linger around as the constant for the next 30 years? What if Windows 98 still has a major user base 20 years from now? And maybe 50% of all desktops are still the 'old' ones after 10 years, that do not run XAML. What if people never upgraded? Well, there will be great many XAML .NET applications, but maybe 10% of all applications in active use ever will be those made with XAML.

    Who would trust using such a technology as their only outlet to the Internet? The realization here is that, nothing like that will ever work because it is not an open standard. Microsoft hopes that the Windows desktop penetration will make it a defacto standard but any such hopes are simplistic and utterly naive. It is a matter of choice - would you let Microsoft control your future in the Internet, as it is becoming ever more important for many areas of business? Hell no!

    XAML and .NET will provide niche solutions in this arena and dominance will never take place. Do you want to be a niche player? Then only use XAML. Raggety HTML will wipe the floors with XAML.
  40. Burritos and brain barfs[ Go to top ]

    Rolf:

    You're eating too many burritos again.

    What does "making fun of MS" have anything to do with my points?...OF COURSE people will always make fun of MS...that's not the point.

    Firstly, just open up any JDJ or Javapro article or editorial from a couple years or so back (or don't you even read the journals?), or just look back on the Javalobby editorials of Ross and you'll see the absurd panic that was gripping Java proponents at the time - culminating of course in Alan's now classic "Java will be dead in 5 years" (due to .NET) jab at the Java psyche.

    Secondly, go to your local bookstore or library, and pull out any book on Palm's history...I suggest Piloting Palm, which has some nice references to the abject dejection of Palm and its supporters, the proclamations of the technorati on the doom of Palm in 1997 or so during the intro of Windows CE.

    if you really wanna go to muck-level, then just go to the newsgroups of years back and note how people predicted (again, and again) the demise of palm, or how C-sharp awhile back was supposed to be THE java-killer (how far have the mighty fallen? There's absolutely no talk today of how C-sharp will kill off java)

    When Microsoft enters a market, the PREVAILING opinion of most of the press and the general populace is that it will steamroller all competition, simply because of its overwhelming resources and monopoly. Perhaps even your burrito-infused brain can grasp this concept?
  41. I am tired of discussing with the immature, emotional, ignorant and uncultivated. I give you a challenge. Read the Gallic War by Julius Caesar and we can talk again. There is a place there that is relevant to the matter of Microsoft vs Sun/IBM/Oracle and the rest. Your task is to identify the place,

    "As an author, Caesar is one of only two statesmen/authors from the late Roman Republic whose works have survived (Cicero being the other.) Along with his Civil War commentaries, his work is a priceless look into the politics, culture, warfare, and personalities of that period. Few, if any, political or military leaders in history have had an impact on humanity as great and far reaching as Caesar. His conquests and statesmanship still resonate today in both our political and social institutions. It is a work that everyone should read at least once in their life to better understand the world they live in today."

    Good luck
    Rolf Tollerud
  42. Why Is Rolf Still Here?[ Go to top ]

    Rolf,

    Hmmm.. Judging from your past and present posts on the server side (including this one), The 'big-three' themes in your postings seem to be the following:

    1] .NET is good (it certainly is a viable alternative to J2EE and I wouldn't mind getting a few systems under my belt for diverification reasons).

    2] MS is powerful, innovative, and shrewd (general consensus is that you are correct).

    3] The people on the server side that rebut your posts are generally beneath you. (Whatever floats your boat)

    My question is: If you don't think the Java platform is viable in the long term, and if you think that developers posting on here are "immature, emotional, ignorant and uncultivated", then why do you persist on posting day after day, thread after thread? Why not pick up shop and spend your time and energy posting on a .NET site (time is money).

    This is a Java site after all.

    J2EE forever... :)

    -Colin
  43. I am tired of discussing with the immature, emotional, ignorant and uncultivated. I give you a challenge. Read the Gallic War by Julius Caesar and we can talk again.

    Well who is the ignorant and emotional here??

    you have been a MS.NET cheerleader on this site and you claimed that MS.NET will kill Java. But this did not happen and WILL not happen, and you are not flexible enough to admit your own misconceptions.

    stop that junk please,
    i would like to be able to read TSS without your trolly messages
  44. Et tu Brutus?[ Go to top ]

    Rolf:

    (1) Why is it I never get a straight argument from you, but points that veer wildly off course?

    (2) Caesar? How about Deng Xiao Ping? His policies and actions affected a hell of a lot more people than Caesar ever did during his times. Here's something more recent for your consumption:

    Take a look at any history text book of recent years.

    Take a good look at the free-wheeling, ultra-competitive world of the capitalistic societies versus the heavy, centralized economies.

    Sure, the overly-centralized entities could marshall overwhelming forces against specific tasks at any one time simply because of the benefits of coordination (dotnet anyone?), but it would always suffer in the end because over-centralization dampens innovation.

    I would say actually that the greatest thing centralized MS has going for it right now are the poor slobs doing Mono, who seem to believe mono actually poses a threat to MSFT, when in fact they are playing right into the company's plans. If they actually believe Mono will revive Linux on the general desktop in countries like the USA, or that we would get droves of Windows developers suddenly realizing Linux is for them simply because of Mono, then they might be in for a rude shock. Microsoft's minions are playing Mono for all the propaganda points it can get.
  45. go back further[ Go to top ]

    How about read the I-Ching, Reg vedas, Mhabharata, Sun Tzu art of war, Epic of Gilgamesh, Ramayana or several dozen other texts which have survived the centuries. Or better yet, go read up on Heizenberg uncertainty principle. I don't see any need to call others emotional or ignorant.

    Fact is, everyone is emotional! You'd have to be a robot or dead to not have emotions or be emotional. All things follow a natural cycle and die eventually. I think it's useful to keep that in mind when we talk about technology, since what ever is popular today will not stay that way. We started out with client server and moved away. Now there's several trends, including going back to client/server. There is no be all or end all technologies. Only mistakes believing there is. Statements like Java will kill MS or .NET will kill java is rather pointless. Can we focus on examples backed up by first hand experience? I think that makes the debate more interesting and educational for everyone. Quoting someone else should be used as supporting evidence, not the primary argument.
  46. I am tired of discussing with the immature, emotional, ignorant and uncultivated.

    I love it when people describe themselves without knowing it. "Pot, Kettle on line two"

    I have a question: Why on earth don’t you just go away then? Try is slowly. See if you can not post on TTS, Java lobby etc until April. Then see if you can make it to the end of the year.
  47. Rolf and google[ Go to top ]

    eheh i had fun today too, I searched on google about rolf to see a bit his posts :)

    Atfer browsing some posts, I think rolf is either:
    - Working for MS
    - Addict to fud
    - both

    An example :
    http://lists.gnu.org/archive/html/gnu-misc-discuss/2004-05/msg00622.html
    The mail was "Linux is shit" :)
    He is not only against J2EE but also again linux and other stuffs...

    A wonderful list of stupid things... quite impressive:
    - Linux is a 30 year old incorrectly designed OS that should have been "Microkernel".
    - Linux-administrator take it as an insult that they should support an ordinary user
    - The Win X system/Open Office crash every 10 min

    Thx to google, I was able to see quite lots of funny sentences he said.. Too bad he forgot some of his other predictions
  48. Warning: Some sarcasm ahead.
    Why is there nobody that keeps a track record? It is fun to discuss and intellectual rewarding to try to make predictions but it is a little boring that nobody keep track.

    Here is just a little pick of the issues in the past.

    1) I was right that windows would out compete OS2.
    2) That IE was a better product then Netscape Navigator 3x-4x.
    3) The J2EE was a slow and cumbersome technology, heavenly overused.
    4) The .NET would get traction and be a major competitor to Java
    5) That cheap Intel-processors in most cases is a better choice that Sun hardware.
    6) That Microsoft would get the upper hand in the handheld market.and many more, some that is not decided yet, for example that MS will the market leader of High-End mobile phones.

    Rolf, I know how you feel. It is the same thing for me: I am forever bothered that the world does not bow down in front of my immense genius. It is very hard to understand how a world full of dimwits can not realize our (that's you and me, Rolf) innate superiority, and worship us as deities.

    As a co-genius of yours, I too keep a track record of predictions that I have made. Here are just a few:

    1) That Microsoft Bob would not redefine computing, despite the overwhelming predictions of the press, the many millions of dollars poured into it (more than the moon landing, I hear), and despite even changing Windows APIs to support it because Bill was [this word was removed to comply with FCC rules] the project manager.

    2) That Microsoft project Blackbird would not kill off the HTML and HTTP standards, despite the overwhelming predictions of the press, the many millions of dollars poured into it (more than the moon landing, I hear), and despite even changing Windows APIs to support it.

    3) That instead of being a "must have" for every online commerce and community site, and a way for Microsoft to skim money off the top of every online transaction ever, Microsoft Passport would be a complete failure, despite the overwhelming predictions of the press, the many millions of dollars poured into it (more than the moon landing, I hear), and Microsoft buying and bribing several high-volume online sites just to convert them to using Passport.

    4) That instead of being on every phone and set top box, Windows would have less than 1% penetration while Java would be steadily closing on 100% penetration, despite the overwhelming predictions of the press, the many millions of dollars poured into it (more than the moon landing, I hear), and Microsoft investing big chunks of dough in cable and wireless companies (most of which seem to end up suing them in return.)

    5) That ActiveX would be one of the least-used plug-in standards for web pages, and would be viewed as a virus instead of an enabling technology, despite the overwhelming predictions of the press, the many millions of dollars poured into it (more than the moon landing, I hear), and Microsoft bribing various web sites to include Active-X, to provide free access only to IE users, and to explicitly include HTML features that a certain other browser did not support.

    6) That Rolf would not actually stop posting here after he promised to.

    See what I mean? I've never been wrong. Every prediction that I later claim to have made earlier is right!

    Peace,

    Cameron Purdy
    Prophets -r- Us: Registered Swamis
    Coherence: Using grid computing to predict the future
  49. Bob, Blackbird?

    I've also heared that MS has not sold so many MS-mousepads as they had hoped for. You forgot that!

    hi hi

    With respect
    Rolf Tollerud
  50. 5) That ActiveX would be one of the least-used plug-in standards for web pages, and would be viewed as a virus instead of an enabling technology, despite the overwhelming predictions of the press, the many millions of dollars poured into it (more than the moon landing, I hear), and Microsoft bribing various web sites to include Active-X, to provide free access only to IE users, and to explicitly include HTML features that a certain other browser did not support.

    I believe the IE Flash plugin is also an ActiveX control.
    Coherence: Using grid computing to predict the future

    I knew the day would come when just couldn't resist anymore to take a very close look at Coherence ;-)
  51. Thank you, Alexander

    Tomorrow is the 10 of January. Exactly two years since the famous "TheServerSide Calls for Real World J2EE Project Stories"

    Floyd Marinescu:
    "we're sitting on a lot of great stories that we're trying to choose from"
    "we're sitting on a lot of great stories that we're trying to choose from"

    I am still waiting...

    I do think one story was actually published though. ;)

    Happy New Year
    Rolf Tollerud
  52. Moving to a new neigborhood?[ Go to top ]

    hmmmm...rolf? now that this jawfest is over, shouldn't you be movin' over to theserverside.net instead of continuin to jaw over here?...i heard they need some comments.....

    hmmmm...and here's a real biggie that .NET lost to J2EE:
    http://www.sun.com/service/about/success/recent/ebay_5.html

    ...and i heard the NYSE just moved to J2EE as well, as did one of the world's biggest online casinos (from asp (asp.net?) to J2EE for security reasons!) after their windows boxes got hacked a couple times.
  53. still waiting..[ Go to top ]

    the world shortest list..

    Woody Allen used to joke about the world shortest book, "Jewish sport achievements through the times". He is a Jew himself, so he can do that. However, he didn't knew about the book "J2EE success stories". ;)

    Don't you have any better example than the three year old case of eBay that took the decision before .NET was available? That made their most stupid decision ever? That is in the claws of IBM that consider a one billion contract a small contract? That has 100+ developers working with the project year after year?

    You can do better, I am sure.

    Regards
    Rolf Tollerud
  54. J2EE powers the world[ Go to top ]

    wait a minute here...are you actually disputing that J2EE is being used in many enterprises? are you actually that deluded and demented? (am i actually asking a troll who spends his valuable life hours trolling away at a technology he hates these rational questions?)

    The NYSE (or don't you know what that is?) story actually occured just a few weeks ago, and the online casino moved to J2EE from Microsoft about 1 month ago. Looks pretty recent to me, eh?

    Just looking at ONE single J2EE app server reveals how dominant J2EE is in enterprises:

    http://www-306.ibm.com/software/success/cssdb.nsf/softwareL2VW?OpenView&Count=30&RestrictToCategory=software_WebSphereApplicationServer

    Rolf:

    BEA has about 30,000 customers worldwide, and IBM's Websphere powers:

    65% of the Fortune 500 companies
    80% of the top US healthcare companies
    75% of commercial banks worldwide
    90% of the top commercial banks in the US
    67% of the world's largest banks use IBM messaging servers
    15 of the top Wall Street brokerage firms
    7 of the 8 largest US telecommunications companies

    And to think that JBOSS supposedly is deployed in more places than these two!

    Not to mention the upcoming Apache J2EE app server...Rolf, please, please go away!

    I guess now you're gonna go off again on your usual non sequiturs, eh?
  55. Also, Rolf, your contention about eBay is not valid...

    In fact, Microsoft had been touting eBay as THE best case study for .NET! Yes, Mr. Tollerud, eBay and Microsoft had actually signed to have eBay use .NET...

    microsoft and eBay made a strategic alliance to have .NET as eBay's backend architecture, but eBay switched to J2EE instead.

    sample URL from Microsoft itself:

    "eBay and Microsoft Announce Strategic Alliance to Expand Global Online Presence"

    This was trumpeted all over by Microsoft, and heralded as one of the premier showcases for .NET (with ballmer probably doing his monkey dance).

    sample URLs:

    "eBay to adopt Microsoft .Net"
    http://www.itworld.com/AppDev/1513/ITW_3-13-01_msebay/

    "More light shed on Microsoft, eBay deal"
    http://www.zdnetindia.com/news/features/stories/16913.html

    "Microsoft, eBay Seal E-Commerce Alliance"
    http://www.ecommercetimes.com/perl/story/8093.html

    "eBay Links Up With Microsoft"
    http://www.cbsnews.com/stories/2001/03/12/archive/technology/main2783...

    "eBay, Microsoft cozy up for new alliance"
    http://maccentral.macworld.com/news/2001/03/13/ebay/

    Unfortunately for microsoft, eBay found out J2EE provided better solutions, and they did a really EMBARRASING (to microsoft) about turn.

    here's the quote on WHY from eBay:

    -----------------------------------------

    According to Geiger, the application platform decision was based on the needs to "increase productivity by creating a component-based architecture, which would do away with the monolithic code base." eBay's evaluation centered on the platform's ability to meet the high quality-of-service requirements (flexibility, adaptability, manageability, scalability, and security) and the need for the architecture to be comprised of frameworks focused on reuse. The two candidates were Microsoft's .NET initiative and industry-standard J2EE technology, with the latter initially considered as a long shot. Geiger recalls, "This was not in the initial vision of V3: If you were to ask me back in December [2000] if we would go to Java [technology], I would have said 'No.'"

    eBay's evaluation of Microsoft's .NET initiative concluded that, at the time, it was not ready to meet its needs. As Geiger puts it, "We were not going to adopt a brand new technology." This was somewhat surprising since eBay's existing "architectural foundation was firmly rooted in C++ [technology]. But, as we dug into the different issues, we reached the realization that we needed a more open architecture accompanied with overarching process changes. This paradigm was integral to our decision to go with Java technology."

    ------------------------------------------
  56. "Bea & IBM are so successful"
     
    Then why are the so difficult to find real life projects then? Why do not Floyd Marinescu pull forward the "great stories that we're trying to choose from" that he supposedly is sitting on? Why do you have to drag up a three year old case from the time where people actually were deceived and thought that Big EJB Java Servers was a viable technology? Do you know that eBay had to ditch the EJBs to make it work?

    As Geiger puts it, "We were not going to adopt a brand new technology". That was his great mistake which he has paid in full. The J2EE technology is not only a mere mistake, it is the greatest catastrophe ever to happen in the IT industry responsible for the loss of billions and billions - probably the main cause to the recession 2000-2003. That there are people that still believe in it is incredible. But what the heck. You can add and eBay to your meager list.

    If BEA are so successful how comes they have financial problems and people are leaving in droves? BTW, have the management managed to stay out of jail yet? On the other hand if there are somebody that really should be behind lock and bars it is the Websphere people compared to the BEA executive are scouts IMO. They make highway and train robbery look like respectable professions.

    Regards
    Rolf Tollerud
  57. J2EE created DotCom Bust[ Go to top ]

    Thanks, Rolf, I know I can always count on you for a little morning humor!
  58. Rolf,

    I think most of us can agree that EJB Entity Beans was a bad idea and it has been heavily misused in the past, BUT, to say that J2EE is the greatest catastrophe ever to happen in the IT industry is only ridiculous.

    You know very well that J2EE is not equal to EJB and by using Spring together with Hibernate, WebWork2 and Velocity you get the most flexible, powerful and testable system that I can think of.

    And yes, I have used .NET on a number of projects as well. It has worked well but I still use Java when I have the choice to do so. .NET and J2EE will of course coexist in many years from now on and
    both will have their success stories. What you can do in .NET you can do in Java/J2EE and vice versa (almost :-)).

    Customers will always "demand" some "XXX-technology" no matter if it is .NET, Java/J2EE or PHP...

    /Tommy
    VisionProject
    Bugtracking made simple
  59. "Bea & IBM are so successful" Then why are the so difficult to find real life projects then?

    I don't know about you, but I've spent the last year or so as a consultant, and I've travelled around various parts of the world for client engagements, mainly in the financial sector. I could reel off a very long list of large-scale J2EE implementations using BEA, IBM, or <insert J2EE server provider here>. I don't think that finding case studies is a real problem. I can also see some .NET penetration, but there is still a huge imbalance, and I can't see a definitive industry-wide push towards .NET that will rectify this any time soon. I mainly see it from long-term, committed Microsoft customers and partners. Having said that though, .NET uptake is certainly on the increase.
  60. corporate culture[ Go to top ]

    From my first hand experience, large corporations are very secretive. Even when there's no reason for it. Other sectors are beyond my experience so this may not necessarily apply. Very few large corporations are willing to go on record about what they use. Even when they do, it's filled with mis-direction, or are so vague that it's worthless. The best way to see, is to become a consultant and get in there.
  61. Then why are the so difficult to find real life projects then?

    We've posted a few of our own J2EE success stories in the past year.
    I've seen loads of stories on IBM's and BEA's sites too. They're not at all difficult to find .. our own biggest problem is finding the time to write them up. My guess is that's Floyd's biggest challenge, considering all the other stuff he's involved with.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  62. etc[ Go to top ]

    Hey Rolf:

    You basically ignored all the links I posted above and went on in another tangent...GREAT way to argue....

    With regards to eBay...did anyone ever tell you J2EE != EJB?
    And that they chose J2EE over .NEt also because they wanted a more open architecture, as that quote so plainly says...

    And of course, that does not explain all the more recent moves to J2EE from Microsoft, including the move by one of the largest online gambling casinos to J2EE because of security problems.

    With regards to EJBs, I've used them before and have no beef with them, but you know what? that's the great thing about Java, there are so many choices for each functionality for customers...can you name other ways to get the job done without EJBs? I doubt it, since you seem to be a pure troll with no technical knowledge of J2EE whatsoever, not to mention the other facets of Java.

    I noticed all your "arguments' basically devolve into calling people names and accusing companies of vague perversions against humanity. Perhaps you should turn your gaze instead to the monstrosity that is windows, and the illegal monopoly maintained by your idol Microsoft?
  63. "you seem to be a pure troll with no technical knowledge of J2EE whatsoever, not to mention the other facets of Java"

    Thank you!

    "the monstrosity that is windows, and the illegal monopoly maintained by your idol Microsoft?"

    To give you an idea of the relationship that Microsoft has with its developers read this post.

    An Open Letter To Microsoft

    There is no way back
    Rolf Tollerud
  64. it is IBM that is the evil monstrosity[ Go to top ]

    trolfy rolfy my deary deary friend... i waiting for the wonderful messages you creating .. we loving you very bery much
  65. "you seem to be a pure troll with no technical knowledge of J2EE whatsoever, not to mention the other facets of Java"Thank you!"the monstrosity that is windows, and the illegal monopoly maintained by your idol Microsoft?"To give you an idea of the relationship that Microsoft has with its developers read this post.An Open Letter To Microsoft There is no way backRolf Tollerud

    I don't see what pathetic tirades from MVP bottom-kissers have to do with what we are talking about?
    Stalin is dead. So is Mao. Russia is capitalistic again. ...

    Apart from believing that Java lobbyists hate you, that OpenSource commies are plotting conspiracy against you and so on, I wonder what other symptoms you may have? Do you have a feeling that someone is trying to poison you? To you have sensation that you are constantly followed? Do you hear voices?
  66. Actually, monopolies like Microsoft on the desktop are closer to the commies than any of the open source proponents ever could be. Microsoft as the politbyro of the IT land :) *shivers*
    Stalin is dead. So is Mao. Russia is capitalistic again. ...Apart from believing that Java lobbyists hate you, that OpenSource commies are plotting conspiracy against you and so on, I wonder what other symptoms you may have? Do you have a feeling that someone is trying to poison you? To you have sensation that you are constantly followed? Do you hear voices?
  67. Rolf! Do you work for M$?
  68. The J2EE technology is not only a mere mistake, it is the greatest catastrophe ever to happen in the IT industry responsible for the loss of billions and billions - probably the main cause to the recession 2000-2003.

    Buh...buh...BuWAAAAAHHaaaaaHAaaaaaaaaaa!!!

    Folks. Enough. It's quite obvious that the gentleman is being serious; he's just out to provide us with a much needed laugh. More power to him I say.
  69. Whoops![ Go to top ]

    I mean't he isn't being serious.

    ... or at least, I hope he's not ...
  70. The J2EE technology is not only a mere mistake, it is the greatest catastrophe ever to happen in the IT industry responsible for the loss of billions and billions - probably the main cause to the recession 2000-2003.Buh...buh...BuWAAAAAHHaaaaaHAaaaaaaaaaa!!!Folks. Enough. It's quite obvious that the gentleman is being serious; he's just out to provide us with a much needed laugh. More power to him I say.

    But he's being anti-American. Europe did not have much of a recession in that period, and the UK in particular did well. But guess what? They used J2EE too! So - he's saying US developers can't keep up with their European counterparts!
  71. Joking?[ Go to top ]

    I am always dead serious.

    Try to estimate the cost of a project like this, taking into account not only the direct costs but also the cost of the lost business:

    "In the last project (a major customer-care callcenter app), they had used Entity Beans, and Websphere, and there were 500 EJBs, and 4700 distinct application classes. It took 2 days of continuous processing just to 'deploy' the beans,"http://www.mail-archive.com/general at jakarta dot apache dot org/msg03376.html

    Imagine thousands upon thousands of similar projects..
    All in all I think my estimate was rather conservative.

    Regards
    Rolf Tollerud
  72. Bad developers[ Go to top ]

    I am always dead serious.Try to estimate the cost of a project like this, taking into account not only the direct costs but also the cost of the lost business:"In the last project (a major customer-care callcenter app), they had used Entity Beans, and Websphere, and there were 500 EJBs, and 4700 distinct application classes. It took 2 days of continuous processing just to 'deploy' the beans,"http://www.mail-archive.com/general at jakarta dot apache dot org/msg03376.htmlImagine thousands upon thousands of similar projects..All in all I think my estimate was rather conservative.RegardsRolf Tollerud

    All I have to say is those cases do exist and they occur more often than most people like to admit. But it is not the result of the specification. It's the developer! I've heard some horror stories from seasoned BEA consultants. You can do similarly horrible things with COM+ too, but that doesn't mean COM+ is evil.

    Easy to abuse is not the same thing as flawed design. Though I try to avoid using EJB and COM+ when I can and only use it when I absolutely need it. You'd be shocked at how many times I've heard developers say "we should use COM+ for all the data access". In this case, all includes select, update, and inserts. My problem is that I haven't found a way to convince managers "finding bodies" is not the same thing as hiring the "right guy" to compliment the team. That's the real key to building a great product. It's like football, basketball, hockey, etc. There are specific rolls and each member of the development team should cover a certain area.

    peter
  73. Bad developers[ Go to top ]

    All I have to say is those cases do exist and they occur more often than most people like to admit. But it is not the result of the specification. It's the developer! I've heard some horror stories from seasoned BEA consultants.

    I've seen some horror stories myself .. quite a few. We used to do a lot of J2EE "performance tuning" (i.e. saving apps that should have never gone to production).

    On the other hand, I've witnessed some very well run J2EE projects that suffer no down-time, including one processing over US$10MMMM (yes, that's trillion) per year. In real money transactions. Unfortunately, I can't say who it is, but you can do the math ;-)

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  74. Fairly easy to guess[ Go to top ]

    If it's NY or Boston area, there's only a few companies generating that kind of volume :) It would probably only take 3 guesses, or I could just ask my friends. I think the top 3 public fund companies are based in either NY or Boston, so it's likely one of them.

    peter
  75. Using the right language[ Go to top ]

    "On the other hand, I've witnessed some very well run J2EE projects that suffer no down-time, including one processing over US$10MMMM (yes, that's trillion) per year. In real money transactions. Unfortunately, I can't say who it is, but you can do the math ;-)"

    Something in wall street or one of those organizations handling mutual funds i'd guess.

    I just had a talk with a rather nice guy who's heading a global research site using perl. no relational database on the backend, large amounts of data (inc images). He's pretty proud how blindingly fast the system is, and sorta made a swipe about the alleged "slowness" of Java in comparison.

    I asked him who coded the thing and knew how to code it, and it turns out he's the only one (i think). makes you kinda wonder what would happen to the whole project if something happens to him...especially since i know some perl and know how much spaghetti code it is...

    In very large financial institutions as above, I would hope they are more worried about stability, security, and maintainability of such large systems over the long haul, rather than just "blinding speed".
  76. Using the right language[ Go to top ]

    Yes. I tend to think of Perl as a write only language!
  77. Bad developers[ Go to top ]

    All I have to say is those cases I've witnessed some very well run J2EE projects that suffer no down-time, including one processing over US$10MMMM (yes, that's trillion) per year.

    So, this system is PURE J2EE? Or is there a mainframe involved. If there's a mainframe involved, then congrats to the mainframe guys...
  78. I'm guessing here, but AFAIK the top 3 public fund companies all have mainframes and all of the trading systems have to integrate with it. I don't know if the rankings are still the same, but it used to be JP Morgan and Wellington were the top 2. I think the biggest private fund company is Fidelity. I think it would be hard to find a "pure J2EE" environment. If you look at the big financial firms in Boston, the positions are normally J2EE.
  79. including one processing over US$10MMMM (yes, that's trillion) per year.

    I just think that stating the above is FUD if there's a mainframe. It would be like me saying we use .Net for 90% of the transactions that go into our multi billion dollar billing system. So, we've deployed a successful multi billion dollar .Net application...

    In all reality, the mainframe is doing all the real work, and the J2EE system is just extra business logic/entry point into the mainframe.
  80. not necessarily[ Go to top ]

    Take fidelity for example, their corporate facts says this http://content.members.fidelity.com/Inside_Fidelity/fullStory/0,,3038,00.html.

    I know some people that work at fidelity and they have a ton of Unix systems and mainframes. What fidelity uses is very similar to most of the other fund companies. Without getting into details, most of them use a combination of JMS and tuxedo to handle XA, which includes saving data to mainframes. So in most cases, a significant percentage of the transactions include multiple databases and multiple insert/updates. Some of the transactions save to the mainframe, while others hit Oracle. I forget where I read it, but fidelity generates something like 5% of the trading volume on NYSE. There are some old articles from late 90's that provides a glimpse into how big fidelity is. While smaller firms may have 50-200K accounts, fidelity has some crazy number like over 20 million. I apologize for the vague statistics, but I don't have time to look up the articles right now. Google is your friend.

    peter
  81. FUD?[ Go to top ]

    "I just think that stating the above is FUD if there's a mainframe."

    1. How exactly would it be FUD since we're all supposed to be Java developers here (exceptin' Rolf the troll of course)

    2. Who knows? If there is a mainframe (and that's only an if since purdy is probably gone), then mainframe may be running JAVA apps. Plus, even if there was a mainframe, so long as most of the load is being handled by java then that's ok.

    3. eBay handles probably hundreds of millions of dollar transactions per day, and its backend is Java
  82. please wake up[ Go to top ]

    I hate to disturb the atmosphere but I have to remain you gentlemen that there is not a single piece of evidence of that any high-performance EJB transaction system exist anywhere on the planet earth.

    Sorry to destroy the good mood
  83. please wake up[ Go to top ]

    I hate to disturb the atmosphere but I have to remain you gentlemen that there is not a single piece of evidence of that any high-performance EJB transaction system exist anywhere on the planet earth. Sorry to destroy the good mood

    J2EE != EJB how hard is that to handle Rolf
  84. Comprehension[ Go to top ]

    "J2EE != EJB how hard is that to handle Rolf"

    He hasn't understood it yet, or he's willfully ignoring it. I can't tell.
  85. please wake up[ Go to top ]

    I hate to disturb the atmosphere but I have to remain you gentlemen that there is not a single piece of evidence of that any high-performance EJB transaction system exist anywhere on the planet earth. Sorry to destroy the good mood

    The thing is, I would love to quote some stats, but I don't want to get in trouble. When I say large financial firms are secretive, it's not a lie. From what I know first hand talking to people who actually build and maintain these systems, EJB's are doing just fine thanks. And no, they don't blindly use EJB for everything. They use JMS or tuxedo when needed. No one responsible for these systems are stupid and jump into it blindly. There's this thing called due diligence, not sure if you've heard of it or not :)

    It usually means coming up with a few designs, testing the top candidates and doing smoke tests. the approaches that look promising are explored further. Once it goes through several phases of due diligence, a full development cycle starts. When a system has to handle x million in transactiosn every single day 24/7, the room for error is very thin. Developers who are diligent and thorough can deliver rock solid applications regardless of the platform. but of course you don't hear that.

    peter
  86. please wake up[ Go to top ]

    I know for sure that 90% of software in UBS Bank (here in Zurich) is powered by WebSphere and 80% of SW in Credit Suisse is powered by BEA WebLogic. Those are second and third biggest banks in the world.
    Hope big enough for Rolf. And yes Rolf, we know abour Einstain relitivity theory...
    One thing is for sure: you won't find many Windows boxes in Zurich Bahnhofstrasse! ;-)
  87. something never changes[ Go to top ]

    "x million in transactions every single day 24/7, the room for error is very thin"

    "you won't find many Windows boxes in Zurich Bahnhofstrasse!"

    You forget something. This yada, yada have I heard before - when we routed the mainframes back in the nineties. The truth is that this "Big Iron" applications are simple and trivial compared to an ordinary desktop application as Eclipse. The arrogance and incompetence (500 EJBs, and 4700 distinct application classes..) of "mainframe type of staff" is the same everywhere in the world.

    It seems that when you have too much money to spend, your brain goes to sleep. On the other hand necessity breeds innovation. (=good old american ingenuity).

    Must be a natural law or something! :)

    Regards
    Rolf Tollerud
  88. something never changes[ Go to top ]

    "x million in transactions every single day 24/7, the room for error is very thin""you won't find many Windows boxes in Zurich Bahnhofstrasse!"You forget something. This yada, yada have I heard before - when we routed the mainframes back in the nineties. The truth is that this "Big Iron" applications are simple and trivial compared to an ordinary desktop application as Eclipse. The arrogance and incompetence (500 EJBs, and 4700 distinct application classes..) of "mainframe type of staff" is the same everywhere in the world. It seems that when you have too much money to spend, your brain goes to sleep. On the other hand necessity breeds innovation. (=good old american ingenuity).Must be a natural law or something! :)RegardsRolf Tollerud

    Please explain how to build a .NET application that can handle 2K concurrent transactions :) I don't mean 2K per second or 2K users. 2K concurrent as in 2K transactions either starting, processing or finishing an insert or update. Queued transactions are not concurrent, since they are processed asynchronously.

    From what I know, mainframe apps are far from simple, but you go ahead and believe what you want :) Obviously no one builds apps as complicated as what you work on. God knows all this regulatory compliance stuff like SOX is trivial and childs play. All those fines the financial firms paid last year was just a figment of my imagination. I'll make sure to tell Fidelity, Wellington, Citigroup, JPMorgan and Charles Schwab they've been doing it all wrong. NOT!!!

    man you make me laugh.
  89. When you are working at a bank in Zurich, or Fidelity, it is difficult to compare what you are doing with others than a select few. It can be compared to participating in a Polo championship. When you make an application like Eclipse it can be compared to participating in the Olympics, running 100, 200 or 400 m.
  90. by the way[ Go to top ]

    you know that microsoft uses fidelity to manage their 401K right :) Of course you didn't bother reading the link I provided, which says Fidelity has over 1 Trillion under management. but I'm sure you consider that small cookie to your big .NET cake.
  91. Queen Elisabeth’s summer-cottage[ Go to top ]

    Fidelity uses mainframes and this is at should be, what has it to do with Java and .NET? Nothing. NET can handle any project that J2EE can + then some, since COM+ performs far better than EJB. (all 10 places in the top 10 Top Ten TPC-C benchmark by Price/ Performance use COM+ and EJB is totally absent).

    I wish you Java followers were able to see yourself from the outside. In the debate with .NET you take refuge in bigger and bigger projects, until you end up with the world’s largest project, which have absolutely nothing to do with Java,

    That reminds me of the story with the two Irishmen. One Irishmen has bought a summer-cottage and his friend is envious and tired of listening to all talk about it. At last he says; "But Queen Elisabeth, she have a nice summer-cottage!"

    Regards
    Rolf Tollerud
  92. Fidelity uses mainframes and this is at should be, what has it to do with Java and .NET? Nothing. NET can handle any project that J2EE can + then some, since COM+ performs far better than EJB. (all 10 places in the top 10 Top Ten TPC-C benchmark by Price/ Performance use COM+ and EJB is totally absent).
    So why in the world aren't these banks using .Net or COM+ instead of Java, when .Net/COM+ are such a marvellous piece of technology? Bugs me.
  93. "So why in the world aren't these banks using .Net or COM+ instead of Java, when .Net/COM+ are such a marvellous piece of technology? Bugs me."

    Because those companies always stay on the trailing side of adoption curve due to their conservative technology policies. What Java replaces is OS/2, mainframe and proprietary client/server solutions.

    They are behind in technology as they are behind in everything else in our modern world. Why is it not possible to trade stocks over internet for tiny fraction of the ordinary courtage for example? Why does an simple bank transaction take 2 days?

    Regards
    Rolf Tollerud
  94. "So why in the world aren't these banks using .Net or COM+ instead of Java, when .Net/COM+ are such a marvellous piece of technology? Bugs me."Because those companies always stay on the trailing side of adoption curve due to their conservative technology policies. What Java replaces is OS/2, mainframe and proprietary client/server solutions.They are behind in technology as they are behind in everything else in our modern world.RegardsRolf Tollerud
    Sorry Rolf, MTS+COM/COM+ is older than J2EE, so by your logic Banks should have adopted that before adopting J2EE. Your argument is false.
  95. I am not working for a bank, but for major European power and automation tech manufacturer. We had .NET frenzy, too. But what we were doing last year mostly was rewriting mid-size .NET stuff with J2EE stuff, in many cases using JBoss. No one had nerves to stress himself with maintaining that crap anymore!
    And here one more example: we have a range of high-end industrial controllers that we produce, probably the best in the world at the moment by tech used. But, as those were designed on the of 90s, MS OPC standard (OLE for Process Control) was used. So we ended up having trouble to sell them just because customers didn't won't to hear about MS stuff on their shopfloors! So, now we are preparing new ones with embedded Java. There are even some JINI initiatives! :-)
  96. Sorry Henrique, you are wrong. Even if COM+ builds on COM, all bits-an-pieces was not in place before februari 2000, with the release of Windows 2000. At that time EJB was in zenith. And besides, .NET was needed too.

    "When Windows 2000 finally ships, though, we'll see the next big set of enhancements to COM, all packaged together under the name COM+".
    http://www.findarticles.com/p/articles/mi_m0FOX/is_18_4/ai_60471628

    Regards
    Rolf Tollerud
  97. Sorry Henrique, you are wrong. Even if COM+ builds on COM, all bits-an-pieces was not in place before februari 2000, with the release of Windows 2000. At that time EJB was in zenith. And besides, .NET was needed too. "When Windows 2000 finally ships, though, we'll see the next big set of enhancements to COM, all packaged together under the name COM+".http://www.findarticles.com/p/articles/mi_m0FOX/is_18_4/ai_60471628RegardsRolf Tollerud
    Windows 2000 was released in February 2000, final J2EE platform specification was shipped in December 1999, the first product fully supporting J2EE was released only on July 2000. Remember: J2EE != EJB.

    COM+ needs .Net? Isn't it the other way around?
  98. You sure on those facts?[ Go to top ]

    http://www.informit.com/articles/article.asp?p=25753

    Before COM+ came around, there was MTS. Since COM+ is a refinement on COM and MTS, historically speaking, it had more time to develop and mature. Therefore, from a time perspective, COM+ should scale better than either J2EE or EJB. EJB being one component of J2EE. There are many opinions about COM+, but here are some thoughts from anders http://www.artima.com/intv/abstract.html.

    If I remember correctly, COM and CORBA appeared on the IT scene about the sametime, but Microsoft got COM out to the market first. OMG has this page http://www.omg.org/gettingstarted/history_of_corba.htm on the history of CORBA. I can't find the history of COM, but maybe you know a definitive source for it.
  99. Rolf,
    Why does an simple bank transaction take 2 days?

    Your arrogance in spite of your ignorance is astounding.

    Bank transactions have to pend for two days without an objection being raised before they are considered processed. It's a legal requirement, not a technology problem. From what I have heard, the rules date back literally to the "paper era".

    Having once worked (very briefly!) on a banking clearinghouse system (I am not allowed to say whose, but you can assume that each country has one) that handles all those "2 day transactions", I can attest to the hoops that the technology has to go through to "emulate" the behavior of the old paper system.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  100. even worse[ Go to top ]

    For certain securities, there is a 3 day rule. This is also a requirement of government regulations. It is to avoid scams and to allow the systems to check an account has sufficient funds. There are even more regulations these systems have to comply with, but I won't bother getting into them. I don't fully understand most of them and they are quite confusing. Just go look up 1940Act and see how many pages that law is and track how many revisions there are on the subsections or sub-subsections :) I had to interpret 1940Act and subsection 2A7 the last two years. Boy it's like understanding a riddle.

    peter
  101. didn't we go over this already ?[ Go to top ]

    Fidelity uses mainframes and this is at should be, what has it to do with Java and .NET? Nothing. NET can handle any project that J2EE can + then some, since COM+ performs far better than EJB. (all 10 places in the top 10 Top Ten TPC-C benchmark by Price/ Performance use COM+ and EJB is totally absent). I wish you Java followers were able to see yourself from the outside. In the debate with .NET you take refuge in bigger and bigger projects, until you end up with the world’s largest project, which have absolutely nothing to do with Java,That reminds me of the story with the two Irishmen. One Irishmen has bought a summer-cottage and his friend is envious and tired of listening to all talk about it. At last he says; "But Queen Elisabeth, she have a nice summer-cottage!"RegardsRolf Tollerud

    COM+ is closer to a transaction monitor than a stateful application container. By the way, how would you manage a transaction that spans 5 countries and more than a dozen systems? Would you insert the original transaction into Sql Server and than trigger a separate process? If you did that, it would mean your database is the sole place for state management. Now here's one catch. How many insert/update/delete would result from doing it that way 5x, 10x, 15x or more than using a stateful container? Let's say you have to integrate with 3 systems which use stored procs and triggers heavily to code the business logic. An insert into table X starts a specific process, which may take 1 second or longer.

    Now, tell me again how you would do this in a stateless design and support a constant load of 1K transactions per second for 15 million customers? Even on a mainframe, doing things this way will bog down the database in no time. Keep in mind some of these system do things like fraud detection and reconciliation, along with the normal auditing required by the federal government. But I'm sure you could code all of this in .NET in no time and have it handle the loads Citigroup handles :)

    Oh wait, I forgot you have zero experience in this domain. I find it funny you keep saying "that is child stuff." I would think a seasoned developer knows complexity grows rapidly regardless of the domain.
  102. no problem[ Go to top ]

    Do you need help Peter? Just create a temporary email address and we take this offline. When we are finished you are free to bash my solution in TSS. You understand of course that I need a little more info as for example what are the requirements for respons time.

    "Be seduced by the Canal du Midi, built from 1666 by up to 12,000 men, crossing rivers, tunnelling through hills, irrigated from the mysterious Montagne Noir, with grace and beauty that lives on to this day. It runs 235 kilometers, requires over 100 locks. It is a love affair of man, nature, and engineering. The canal was created by Riquet, the tax farmer for Languedoc who wanted to join the economies of the Mediterranean to the Atlantic and stands until this day."

    Regards
    Rolf Tollerud
  103. obvious[ Go to top ]

    Needless to say, Riquet had not a "mainframe type personality".
  104. just respond here[ Go to top ]

    why are you afraid to respond here? it should be simple right :) Having worked with guys that build this stuff for 20 years, this is my understanding. Having the database maintain all state reaches a bottleneck very quickly because the relationship between tables, temp tables, and triggers/stored procs gets extremely complex. What is worse is inserting several rows in this complex process, only to delete the row later. To get a rough idea of what this is like, create a datbase model with 100 tables, then create about 50 stored procs and triggers that do aggregations, auditing, fraud detection and reconciliation. Now, run some simple benchmarks and tell me how long the whole process takes.

    even before you do it, I can tell you right now it will take several seconds. This means at best on a 4 CPU server, Sql Server will only be able to handle 1-10 transactions per second involving this complex process. That doesn't include a simple insert that doesn't trigger the process. for entertainment, say you have to handle a constant stream of 2K transactions per second with 5% involving the complex process. One of the aggregations should perform a sum, average or median for an entire table and the table should have atleast a half million rows. You can use an index view to accelerate the aggregates if you like.

    Or better yet, just go up to TPC and look at the decision support benchmark, which is measure by query per hour :)
  105. just respond here[ Go to top ]

    Or better yet, just go up to TPC and look at the decision support benchmark, which is measure by query per hour :)

    Rolf will only look at statistics that make him feel better about himself for having chosen an obsolete and proprietary technology platform .. and for being one of the seven or eight people left in the world to not have already moved on to something more mainstream.

    We used to deal with this same Wounded Pride Syndrome (WPS) from the mainframe programmers back when it was obvious that the mainframes had become legacy systems. You may be tempted to pity him, but unfortunately there's really nothing you can do to help him if he's not willing to help himself.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  106. yeah, it's fun to watch him rant[ Go to top ]

    I may be demented, but I find it hilarious to see him spout off about things he has no clue on. I'm sure the good .NET developers I know would cringe at the things Rolf says. Most of the people I know in the finance world would be ROTFLTAO at Rolf's responses.

    Rolf would probably cringe at STP (straight through processing) protocol the securities firms use for electronic processing of trades.

    peter
  107. don't take this personal[ Go to top ]

    Do you need help Peter? Just create a temporary email address and we take this offline. When we are finished you are free to bash my solution in TSS. You understand of course that I need a little more info as for example what are the requirements for respons time."Be seduced by the Canal du Midi, built from 1666 by up to 12,000 men, crossing rivers, tunnelling through hills, irrigated from the mysterious Montagne Noir, with grace and beauty that lives on to this day. It runs 235 kilometers, requires over 100 locks. It is a love affair of man, nature, and engineering. The canal was created by Riquet, the tax farmer for Languedoc who wanted to join the economies of the Mediterranean to the Atlantic and stands until this day."RegardsRolf Tollerud

    anything you have to say regarding this domain would be rather worthless to me. I know guys who have 15-25 years of experience building these systems and I've had detailed discussion with them. Though honestly most of it was way over my head. The level of complexity is far beyond my current knowledge. for example, I know a gentleman who has built compliance systems for fidelity, wellington, state street, and a few others I don't remember. It would literally take me 20 years to reach his level of understanding and depth. This is why IBM and Microsoft pay fidelity to manage their 401K.
  108. don't take this personal[ Go to top ]

    "I know a gentleman who has built compliance systems for fidelity, wellington, state street, and a few others I don't remember. It would literally take me 20 years to reach his level of understanding and depth"

    Sounds rather inbreed to me. I advice you to get some fresh blood. I you really want to learn, (instead of trying to impress people) go to Shadowfax

    "The Enterprise Development Reference Architecture (EDRA) – previously code named "Shadowfax" – provides architectural guidance that an organization can use to standardize and improve the development of distributed systems. The EDRA includes an extensible application framework, four QuickStarts, an application template, and supporting documentation. The Global Bank Reference Implementation (GBRI) is also available. The GBRI is a sample application that uses patterns & practices guidance (including the EDRA) in an online banking scenario. Sign into the workspace and download EDRA, GBRI, and the documentation from the "Releases" section. Also the Wiki knowledge base for EDRA can be found at" http://channel9.msdn.com/wiki/default.aspx/Channel9.EDRAWiki and the GBRI Wiki at http://channel9.msdn.com/wiki/default.aspx/Channel9.GlobalBankWiki.

    with greatest respect
    Rolf Tollerud
  109. that's old stuff[ Go to top ]

    http://www.law.uc.edu/CCL/InvCoRls/rule2a-7.html rather than try to summarize 2A7 regs, there is a link to it. Read that about 20 times and tell me how you would implement it. By the way, I've already implemented those rules for one system.

    If you end up being totally confused, it' ok. No one can understand those laws in a few hours or even weeks. Go ask microsoft if they think regulations are easy to support. Oh wait, Microsoft gave a talk already and they said it was very hard. So either MS is wrong or you're living in an idealized world :)

    have fun

    peter
  110. With all due respect Peter I think you take on problems in a wrong way. Whenever you get a contract for a new task you need to find the super user, the intelligent computer savvy specialist in the particular business domain you are going to implement.

    Such was the Canal du Midi made, back in 1666, one of the greatest and most spectacular architectural feats of all time. All the established experts said it was impossible.

    Regards
    Rolf Tollerud
  111. ROTFLMAO[ Go to top ]

    I don't know how you work, but usually the way it works in the industy is this. Usually there's several people that have to work together to interpret a specific section of the regs. It might include but limited to:

    1. a lawyer who specializes in interpreting regs
    2. a business analyst who understands the securities domain like the relationship between legal entities, securities, backers, under-writers, etc
    3. a compliance expert who interprets 1 and 2 to define a clear algorithm
    4. a rule expert who translates the mathematical algorithm into executable algorithms
    5. a developer who implements the algorithm
    6. a functional team to create test database with sample data
    7. a qa team to run the sample rules against "realistic" data
    8. a lawyer who looks at the input and outputs of the test to verify it meets the legal requirements
    9. a report to the SEC verifying the company supports the required regulations and has passed rigorous certification

    So the answer id "DUH! it's not possible to know everything yourself." Not supporting regulations is not an option, unless you like living in a 8-12 cell with a boyfriend. The upside is you get to workout on a regular schedule, meals are provided for and every day is interesting. This computer savvy super user "does not exist". Oh wait, I forgot, you qualify as "the computer savvy super user." </joke>

    peter
  112. ROTFLMAO[ Go to top ]

    The development process proposed by Rolf has been scientifically proven inefficient and ineffective:
    http://news.bbc.co.uk/1/3013959.stm
  113. the Waterfall-method? ;)[ Go to top ]

    There should be two persons responsible, no more no less. This is a system that has showed it value again and again through history, in all areas of human activity. Other people can be hired in temporarily.

    On a side issue, in my opinion, the Java world’s almost superstitious belief in experts and authorities is the main thing that keeps us most apart.

    Why do you not read the history of the canal and learn how they were in an almost continuous state of improvising?

    Regards
    Rolf Tollerud
  114. I'm not worthy[ Go to top ]

    There should be two persons responsible, no more no less. This is a system that has showed it value again and again through history, in all areas of human activity. Other people can be hired in temporarily.On a side issue, in my opinion, the Java world’s almost superstitious belief in experts and authorities is the main thing that keeps us most apart.Why do you not read the history of the canal and learn how they were in an almost continuous state of improvising?RegardsRolf Tollerud

    Since you are "the computer savvy super user" who is able to understand everything. I look forward to your explanation of 2A7 regulation as it pertains interpretation, algorithms and implementation. The rest of the world is still trying to evolve from normal humans to the "super human" level you've achieved. Thank god we have a semi-diety to worship. All those lesser dieties like Jesus, Rama, Zeus, and Adonia should all bow down to Rolf "soothe sayer of the .NET masses."

    You should write a book called "I am Rolf, hear me bark: truth for the masses of unwashed developers." This way everyone can read it and know how not to develop software.

    </sarcasm>
  115. personality change?[ Go to top ]

    It is not I who is the computer savvy super-user I am the professional, remember?, like yourself.
    I do not understand, deliberately misunderstanding is not your normal behavior that is the mark of Cameron. But as things are, in the words from "Raiders from the lost ark": "Now you're getting nasty", better end the discussion.
  116. it's a joke[ Go to top ]

    it was meant as a joke. not successful, and a lame attempt. what can I say, I'm not a comedian.
  117. personality change?[ Go to top ]

    It is not I who is the computer savvy super-user I am the professional, remember?, like yourself. I do not understand, deliberately misunderstanding is not your normal behavior that is the mark of Cameron. But as things are, in the words from "Raiders from the lost ark": "Now you're getting nasty", better end the discussion.

    Rolf: Before you accuse others of getting nasty:
    Just remember to go to your old shool and get the money back.
    You can see from Steve’s answer that he do not understand enough statistics...
    I am tired of discussing with the immature, emotional, ignorant and uncultivated.

    and, finally:
    deliberately misunderstanding ... that is the mark of Cameron

    I can't see you are in a position to complain.
  118. You underestimate him[ Go to top ]

    Since you are "the computer savvy super user" who is able to understand everything. I look forward to your explanation of 2A7 regulation as it pertains interpretation, algorithms and implementation.

    You have to understand that the 2A7 regulation is peanuts for geniouses like Rolf. He could probably grasp the 2A7 reading it with one eye while memorizing every line from all episodes of M.A.S.H (fast forwarded) with the other, and in the end generate the script for every episode in hex form (all this in only 10 minutes, remember he's a master coder)

    Writing simple code for implementing the 2A7 is nothing for .NET-gods like Rolf, just look here:

    http://www.dotnet247.com/247reference/msgs/10/53783.aspx

    /Mikael
  119. Thanks for that laugh[ Go to top ]

    Obviously that question Rolf asked is much harder than calculating the weight of a mutual fund for an entire firm or fund. And it definitely harder than calculating the weight of any single issuer within a group of accounts, taking into account the money market, hedge and mutual funds making sure the absolute exposure by issuer grouped by security type is accurate. In fact, I sure he can do it in his sleep with 1 pinky. Silly rabbit, compliance is for kids!!!
  120. last nail in Linux coffin[ Go to top ]

    Today was Chrismas day. Avalon Preview Released for XP!
    Download here:
    Public November 2004 "Avalon" Community Technology Preview Including WinFX™ SDK

    And before you make any stupid jokes..
    Netcraft Confirms--Linux Is Dying:

    http://it.slashdot.org/comments.pl?sid=136123&threshold=0&commentsort=0&tid=109&tid=156&tid=218&mode=thread&cid=11368912

    I told you so! (my favorite sentence!)
    Rolf Tollerud
    (hi hi, it is the last laugh that count)
  121. Last laugh[ Go to top ]

    Have you solved the 2A7 regs offline yet, as you proposed to do earlier? Or are you still busy with that "Event Bubbling Control Sample"? Looking forward to your future genious "Canal Du Midi" (comparable) inventions.

    /Your big admirer,
    Mikael

    (I admire you, you are so 1337!!!! Your last post in the following thread is so - what should I call it - "manly"? http://weblogs.asp.net/sjsmith/archive/2004/01/21/61468.aspx)
  122. The Voice of Common Sense...[ Go to top ]

    Plz, don't fight with trolls. Don't feed them instead. They will die slowly in pain.
  123. last nail in Linux coffin[ Go to top ]

    And before you make any stupid jokes..Netcraft Confirms--Linux Is Dying:http://it.slashdot.org/comments.pl?sid=136123&threshold=0&commentsort=0&tid=109&tid=156&tid=218&mode=thread&cid=11368912I told you so! (my favorite sentence!)Rolf Tollerud(hi hi, it is the last laugh that count)

    You are using a Slashdot post marked as 'flamebait' as evidence?
  124. last nail in Linux coffin[ Go to top ]

    You are using a Slashdot post marked as 'flamebait' as evidence?

    Yeah, so out of character for Rolf ;-)
  125. to read or not to read slashdot[ Go to top ]

    Now and that you can find bits of wisdom even in a pigsty.

    What do you say to this little pearl for example?

    "Instead of discussing the technology (which is actually pretty cool...they do have smart engineers at Microsoft), I have a feeling this will be a bunch of Microsoft-bash posts.."
    One third referencing some obscure GUI from the past where something almost like this has been done already, another third referencing some future project not released yet doing the same, and the rest a bunch rehashing old Microsoft jokes from the last eight years.

    With best respect
    Rolf Tollerud
  126. to read or not to read slashdot[ Go to top ]

    Now and that you can find bits of wisdom even in a pigsty. What do you say to this little pearl for example?"Instead of discussing the technology (which is actually pretty cool...they do have smart engineers at Microsoft), I have a feeling this will be a bunch of Microsoft-bash posts.."
    One third referencing some obscure GUI from the past where something almost like this has been done already, another third referencing some future project not released yet doing the same, and the rest a bunch rehashing old Microsoft jokes from the last eight years.
    With best respectRolf Tollerud

    Well that is Slashdot. You won't find much support for .Net there (or Java). It's a hard-core open source site.

    I have to say that I have nothing against the new Microsoft APIs and tools - they look pretty good. But they are irrelevant to me and most other UI designers. I have to code for ALL customers: to put up a site that was Microsoft-only would be commercial suicide. I am not going to include specific interface features in my website that deliberately alienate 10% of my users (well, actually 100% if I used Avalon, as Longhorn is not out yet). Not to mention that as I am setting up sites that may be seen in certain European countries, supporting non-MS browsers is part of their requirement.

    Look how long it took for WinXP to get a significant fraction of the PC market - there are still huge numbers of Win98 users out there. Even if I were coding purely for the very small client-side app market, I would still have to wait at least a decade before there was sufficient Longhorn penetration of the market to make using these new features worthwhile.
  127. an unattainable standard[ Go to top ]

    Steve,

    "I would still have to wait at least a decade before there was sufficient Longhorn penetration of the market to make using these new features worthwhile."

    It is not Avalon that are going to create the great paradigm shift but all the competing technologies that will spring up as a result.

    How often does it happen that we are allowed a peek 6-7 years into the future?
    How often does it happen that we see fantastic new technology created from bottom up without any regard for "backward compatibility?

    Avalon sets a standard, the equivalent of a role model if you want. Soon the market will be flooded with similar solutions with more or less comprehensive subsets. You will see it in Flash, for Mono, for Java, for Compact frameworks, everywhere.

    All will look to the Avalon for inspiration and guidance.
    That is why it is so important

    Regards
    Rolf Tollerud
  128. scary[ Go to top ]

    Wow....you're not really loony in the "hilarious way", you're actually a religious loon in a most scary way.
  129. perhaps Rolf really is a shell script[ Go to top ]

    Think about it, it's not all impossible. I mean there's Annanova, so it's possible the posts are the result of a bot. I could probably write a simple program using JESS and a couple hundred rules. Getting the pages from TSS would be easy with HtmlParser library. I could easily build a list of links and saying for the knowledgebase.

    then it's just a matter of simple pattern matching to return semi-topical, but totally fluffy responses.
  130. so classic[ Go to top ]

    Now and that you can find bits of wisdom even in a pigsty. What do you say to this little pearl for example?"Instead of discussing the technology (which is actually pretty cool...they do have smart engineers at Microsoft), I have a feeling this will be a bunch of Microsoft-bash posts.."
    One third referencing some obscure GUI from the past where something almost like this has been done already, another third referencing some future project not released yet doing the same, and the rest a bunch rehashing old Microsoft jokes from the last eight years.
    With best respectRolf Tollerud

    So between now and when the day .NET rules the solar system, you plan on answering technical questions with real technical answers? Or would that be beneath you? I must say reading your posts is kinda like listening to Howard Stern. Though atleast howard gives facts every now and then, whereas I see no concrete technical facts aside from the usual references to bogus statistics or market studies that amount to PR. By the way, Microsoft also never lies or claims their rule engine is RETE.
  131. perhaps Peter is a bureaucrat?[ Go to top ]

    "do you plan on answering technical questions with real technical answers?"

    I will be happy to answer any computer software related question but instead you keep pestering me with this non IT related nonsense. Regulation 2A7, 2A8 2A9. etc These government regulated papers are the same all over the world.

    I feel sorry for you if interpreting these idiotic rules is all what you do all the days.

    Regards
    Rolf Tollerud
    (i didn't see you answer any question on Expert-Exchange..)
  132. of course[ Go to top ]

    "do you plan on answering technical questions with real technical answers?"I will be happy to answer any computer software related question but instead you keep pestering me with this non IT related nonsense. Regulation 2A7, 2A8 2A9. etc These government regulated papers are the same all over the world. I feel sorry for you if interpreting these idiotic rules is all what you do all the days.RegardsRolf Tollerud(i didn't see you answer any question on Expert-Exchange..)

    As usual you ignore my 5 other technical questions about how to calculate aggregations, manage transactions, and scale an application. I find it hilarious, because really you're doing .NET developers a great disservice. One more than one occasion, I've given detailed description of how I've built high availability systems with .NET. I've also given detailed descriptions of how other people can achieved high throughput using .NET based my analysis of existing case studies.

    You're going to have to try harder than that, if you think an intelligent (not that I am) developer be it .NET, C++ or Java is going buy your flimsy excuses. by the way, I fully expect you to only see what you want to see and never address real technological issues with details like "how do i rewrite correlated subqueries to use joins instead to optimize performance," or "when should I use OLAP to aggregate millions of rows of data for semi-realtime applications." that would actually require first hand experience to understand how Sql Server handles/manages execution plans and when it is appropriate to use triggers to update summary tables. None of these techniques are platform specific. They are derived from practical first hand experience.
  133. Is that so.[ Go to top ]

    I think your practical experience is very limited, Peter, and at any case this is not the forum for the "experienced soldiers from the field", that is Experts-Exchange.com.

    Here is the place for "Well-meaning Impractical Theoreticians". Such as yourself. "Stating some problem in vague and uncertain terms and bash people when they asks for clarification" always have been the practice of bureaucrats. Whom do you think you are kidding?

    We can take this to Experts-Exchange to decide this matter but you have to prove that you are worthy. Pick any target-area and show that you can take one of the 10 top places first.

    Not that it settles anything, the core issue is that the Java/J2EE tools is crap. Please enjoy the top segment of the market where you can overprice 10-20 times without the customer complaining as long as you can.

    It is not going to last very much longer.

    Regards
    Rolf Tollerud
  134. You want specifics, that's easy. Using the following data model, perform aggregations calculations using either Sql Server or OLAP.

    Security
    - ticker
    - cusip (non text code equivalent for ticker)
    - GISC sector
    - GISC industry group
    - GISC industry
    - GISC sub industry
    - legal entity id (this is the issuing company)
    - exchange
    - security type (bond, money market, mutual fund, stock)
    - maturity
    - duration
    - asset backed

    Account
    - first name
    - last name
    - userID
    - address

    AccountItem
    - cusip
    - quantity
    - purchase date
    - security type
    - itemType (buy long/short, sell long/short)
    - sell date
    - status (open, close, waiting period)

    legal entity
    - id
    - name
    - address
    - parent legal entity
    - entity type (public, private, federal, state)

    rating
    - rating type (s&p, moodies)
    - rating code
    - rating value (numeric value)
    - rating name
    - rating description
    - legal entity

    transaction
    - cusip
    - account id
    - quantity
    - price
    - price range
    - status (open, close, pending)

    transactionSet
    - transaction (an array)
    - trader
    - isProxy (a grouped trade, to be dispersed across several accounts later)


    this is a simplified version of what many trading firms use. the actual model usually has several tables. Now for the scenario.

    each account should have 20-50 account items, which is also refered to as positions in the industry. the database should have 100-200K distinct securities. create 200K account and vary the number of account items.

    Now, try to calculate/evaluate the compliance of any given account with the following rules.

    1. the weight of sector X should not exceed 10% for the portfolio. Warn if the weight is 5%
    2. the weight of industry group X where country is equal US, FR and GB should not exceed 8%
    3. the absolute weight of all issuers in an account cannot exceed 12%.
    4. the weight of security Type X cannot exceed 10% by country


    Try to see how many transactions per second Sql Server can handle with transactionSets of 20. I've already done this, and it's not applicable to just compliance. Many firms today are using these same techniques to do data mining and business intelligence. Microsoft has been pushing heavily into this area. In fact the new reporting features in Yukon are there to cover these areas better.

    Have fun.
  135. not so difficult[ Go to top ]

    There you see, it was not so difficult- I didn't hurt did it? Now we are almost there, 50% is done, when everything is clear, we will probably see that the problem is very simple.

    Being that I am such a philanthropic and kind person I am always willing to give moral education for my TSS.com fellows.

    "When the physiologist professor explains something and his students do not understand anything he is proud. When a philosophy professor says something and his students do not understand anything he is ashamed."

    Which one was insecure with an inferiority complex, do you think?

    Regards
    Rolf Tollerud
  136. not so difficult[ Go to top ]

    There you see, it was not so difficult- I didn't hurt did it? Now we are almost there, 50% is done, when everything is clear, we will probably see that the problem is very simple.Being that I am such a philanthropic and kind person I am always willing to give moral education for my TSS.com fellows."When the physiologist professor explains something and his students do not understand anything he is proud. When a philosophy professor says something and his students do not understand anything he is ashamed."Which one was insecure with an inferiority complex, do you think? RegardsRolf Tollerud

    Yup, not so difficult to handle 500-4K transactions per second on a 4 or 8 CPU server using .NET. You still haven't answered with anything resembling technical answer. I suspect I will never see any kind of technical answer from you. It's also not so difficult to handle 500 concurrent TransactionSets while doing these kinds of operations for ever single transaction in a TransactionSet.

    I look forward to the day you actually provide real concrete answers that contain specific technical facts that aren't blog entries by other people. Before you even try, I can tell you exactly what kind of performance you can expect after 8 months of development and in depth performance tuning. 1-4 transactions per second, if you get into Sql Server excution plan and really squeeze every ounce of performance out.

    But don't take my word for it, do it yourself. It's a great learning experience. what I described in the previous post is approximately 1/20th of the actual process one would have to implement for basic compliance. For full pre-trade compliance, the scenario above covers 1%. I don't have time to spend the next 12 months explaining every minute detail to you. Plus you wouldn't listen anyways :)

    have fun trying to answer those questions.
  137. "I don't have time to spend the next 12 months explaining every minute detail to you"

    To bad, I would have loved to show you how much faster and how much cheaper the MS solution would have.

    The Windows Advanced server 2003 is more reliable, more secure, faster both as Webserver and fileserver than Linux. .NET has outperformed J2EE in all known tests, and SQL Server/COM+ debunks Oracle in the TPC tests. In short, the Java components are all crap.

    Still with all the different part significantly better than the competition you claim that by some magic the whole is more than the parts?

    Cedric Beust:
    Some scalability stories: major insurance company (200 tps, 700,000 tx/day), Merrill Lynch: 21,000 tps (yup, twenty-one thousand), 75M tx/day, CNET 4400 tps, 1.5M tx/day.
    http://www.beust.com/smackdown.html

    To me frank, I never enjoyed myself as much as I've done the last two years when it has become obvious that MS is going to blow everything away. ;)

    Regards
    Rolf Tollerud
    (Now, back to Avalon!)
  138. To [sic] bad, I would have loved to show you how much faster and how much cheaper the MS solution would have [sic].

    It's pretty obvious that it doesn't begin to meet the requirements, so it could be free and that would still cost too much. If you're around, drop by FinExpo in London next week for my talk .. I'm speaking on this very subject.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  139. BTW[ Go to top ]

    "I look forward to the day you actually provide real concrete answers that contain specific technical facts"

    Why don't you pick up my thrown glove regarding Experts-Exchange? I won first place in July, August and September last year for my target area. I look forward to an honest competition.
  140. Respect your domain[ Go to top ]

    I'm not even going to bother compete in webdev domain. I haven't done anything remotely hardcore webdev in 5 years. Even when I did graphic design in the mid 90's and ASP/JSP stuff in late 90's it was still stock web dev. When I was focusing on webdev, I knew plenty of guys who knew javascript better than I did. Since 99 I've focused on mid-tier and backend, so webdev is really my weakest area. I don't plan on focusing on it either, since I prefer to focus on high availability/high performance applications.

    Back to the topic of answering with specifics. Anyone with decent understanding of Sql (the language) can point out the various approaches for calculating aggregates in real-time and see the limitations of each approach. Keep in mind I'm not an expert, since I know sql and oracle admins with 8-12 years of experience.

    For example, one approach to calculating aggregates is to query against the table or select into a temp table. since the scenario would have close to 1 million rows of account items, doing queries against the table or dynamic view might be in-efficient. Selecting the account items into a temp table should be faster, but the catch is you have to delete those rows at the end of the process, so there's a significant trade off. If you the stored procs didn't delete the rows by account in the temp table it would quickly be larger than the main account items table.

    In this case, the rate of transactionsets coming into the system plays a big roll in the implementation. If the rate is 10-20 transactions per second, it's probably going to be eaiser to query against the account items table directly. On the otherhand, if the rate is 500/sec it might be better to select into a temp table and then periodically flush the table of old data on a set interval. both of these approaches have a significant draw back because the calling application won't know when the results are done. Now with Yukon, one could use the new messaging bus to send out a MSMQ message. But that doesn't get around the fact that you might have to lock account items for a given account while the aggregates are calculated to make sure the result is accurate. Obviously optimistic locking is preferable, so one can solve the locking issue by changing the process.

    I see no reason why you're unable or unwilling to provide basic description of different techniques and their pros/cons. this has nothing to do with platform and has everything to do with one's knowledge of sql and how to perform aggregations.

    the description I just provided has plenty of flaws, which I didn't point out. If you know sql well, you should be able to see it and provide alternate approaches.
  141. Well Peter I have insufficient information but I try to answer some of your questions now that it seems to boil down to how to use the SQL Server in the most efficient way.

    First some words of advice. I am a performance freak, no doubt about it. Unlike some that advocate "build now, optimize later" I think that performance is something that you should think about all time when you are developing , morning, mid-day and evening, even in the night, and not to be slapped down later.

    As to sub queries forget them, even normal joins should be avoided as far as possible, Another important measure is to of course to separate the reporting server from the production system (failover Clustering and lots of lots of memory of course) where you do inserts/Update/deletes.

    That is just normal common sense. But in your case I would advocate the more powerful SQL Server Analysis Services with OLAP cubes and dimensions and program against it with multidimensional expression (MDX) statements. Also, with Analysis Services there is some real performance advantages to be gained by using the new 64-bit version of the SQl Server.

    I apologize in beforehand if this is something you know all about already.
    http://www.microsoft.com/technet/prodtechnol/sql/2000/evaluate/ansvcs64.mspx

    For a MDX primer look here,
    http://www.microsoft.com/msj/0899/mdx/mdx.aspx

    SQL Server 2005 (scheduled for release in early 2005) hosts the .Net Common Language Runtime (CLR). More info about SQL Server 2005 is here: http://www.knowledge-management.com/news/v4n8/yukon.html

    Regards
    Rolf Tollerud
  142. I am a performance freak, no doubt about it. Unlike some that advocate "build now, optimize later" I think that performance is something that you should think about all time when you are developing , morning, mid-day and evening, even in the night, and not to be slapped down later.As to sub queries forget them, even normal joins should be avoided as far as possible

    Sorry to pick up on something addresses to someone else, but... Wow! This is scary! It goes against almost every good principle of software design. I would certainly hope that anyone coding any financial or business service I use put correctness and validity as the absolutely key priority for design, not performance. Performance can be achieved after getting correctness right, but its very hard to do things the other way around.
  143. as I said[ Go to top ]

    "On a side issue, in my opinion, the Java world’s almost superstitious belief in experts and authorities is the main thing that keeps us most apart."

    "It goes against almost every good principle of software design"

    Such as in Sun's petstore application?
  144. as I said[ Go to top ]

    "On a side issue, in my opinion, the Java world’s almost superstitious belief in experts and authorities is the main thing that keeps us most
    apart.""

    Could you explain exactly how this relates to the matter of correct code? Or is 'correct' just a matter of opinion, secondary to speed? If you were using a bank website, would you rather it was written by authoritive experts or a hacker more interested in performance?

    By the way, just repeating a quote does not make it valid. I believe your quote comes from Miguel De Icaza, the Mono guy. Well, it's strange isn't it, that you are quoting someone who is so keen on Linux.

    Oh, and also, if Java is so awful, why was Microsoft so keen on it:

    "First, the cadre of highly intelligent and incessantly curious Microsoft engineers discovered how much easier it was to work with Java than with C or C++. Second, Microsoft discovered that Java is an ideal language for implementing the kind of components on which it had based its whole corporate strategy." (www.kamath.com)

    The only reason MS abandonded Java was because they were not legally allowed to modify the core language and libraries to add OS-specific features, and because they refused to implement the complete standard (as they were required to do by the license).
     
    It goes against almost every good principle of software design"Such as in Sun's petstore application?

    Did Sun's pet store application give wrong answers?

    You can't have it both ways - you can't criticise Sun's J2EE application for not following good software design (devised by 'experts') and then say that experts are a bad thing.
  145. again[ Go to top ]

    Steve,

    This thread has gone out of steam. To dig up the old issue again is too boring. IBM has done exatly was MS was forbidden to do, and regarding not implementing the whole specification there was other players (Netscape comes to mind) with a less complete implementation that was not sued. That Sun won this lawsuit is just proof that the American juridical system is broken.
    But that did we know already.

    To think that Sun with their insignifant company could dictate great Microsoft without allowing MS any influence at all was ridiculous. Please note that it happened during the great IT boom where Sun arrogance was at its worst.

    Remember my view of things:

    "Sun was desperately trying to sell Unix workstations, a business idea that had failed, and was saved in last minute by the miracle of the Web. Certainly no by any effort or brilliance from Sun."

    "A company that doesn't make neither its own hardware nor OS and that has not had even one single successful software product ever? That hire friends and relatives without proper screening? That have a pointy haired manager that never been a professional? That had a lame business idea with Unix "workstations"? That only exists because of the Internet boom? That tries to take credit for Java that is mainly developed by the international community? That makes "specifications" instead of real software? And when other companies finally have dragged the "product" to something useable claim ownership and charge 500 000+ for certification? The company that says they are not earning any money from Java when they actually has taken in approximatly 20 times as much as they have invested? That has the most arrogant staff on earth? That spend more time in court and inventing invent outrageous slanders about Microsoft and Bill Gates than develop own products?"

    I hope you get the gist of my opinion about Sun.

    "On a side issue, in my opinion, the Java world’s almost superstitious belief in experts and authorities is the main thing that keeps us most apart."

    Is my quote, not anybody elses. But you may quote me!

    Regards
    Rolf Tollerud
  146. P.S.[ Go to top ]

    Click here to learn what happens when Sun try to do something themselves, for once.
    http://www11.brinkster.com/monoasp/sun-internal-memo.htm
  147. again[ Go to top ]

    IBM has done exatly was MS was forbidden to do, and regarding not implementing the whole specification there was other players (Netscape comes to mind) with a less complete implementation that was not sued.

    IBM did not add specific features that tied Java to a specific operating system. Other companies had incomplete versions of Java, but did not state that it was their intention to keep the implementation incomplete. Of course, if these companies did wrong things (in your opinion), this does not make Microsoft right.
    "A company that doesn't make neither its own hardware nor OS and that has not had even one single successful software product ever? That hire friends and relatives without proper screening? That have a pointy haired manager that never been a professional? That had a lame business idea with Unix "workstations"? That only exists because of the Internet boom? That tries to take credit for Java that is mainly developed by the international community? That makes "specifications" instead of real software? And when other companies finally have dragged the "product" to something useable claim ownership and charge 500 000+ for certification? The company that says they are not earning any money from Java when they actually has taken in approximatly 20 times as much as they have invested? That has the most arrogant staff on earth? That spend more time in court and inventing invent outrageous slanders about Microsoft and Bill Gates than develop own products?" I hope you get the gist of my opinion about Sun. "On a side issue, in my opinion, the Java world’s almost superstitious belief in experts and authorities is the main thing that keeps us most apart." Is my quote, not anybody elses. But you may quote me!RegardsRolf Tollerud

    I confess that I actually did raise this issue to reveal your attitude. You have an issue with Sun. If Microsoft had come up with Java, you would have had no problem with it. You simply have an irrational faith in Microsoft, and will use any argument to back this up.

    You simply refuse to accept facts. Sun has never had a successful product ever? Makes specifications but not real software? These statements are random nonsense gibberish.

    As for Sun only existing because of the internet boom - they have been around for over 20 years. I guess at least one acknowledgement that you got this fact wrong might be progress. Go on - admit you got this fact wrong. You can do it! Then we can work on the other 'facts' you state.
  148. Sun has existed for 20 years[ Go to top ]

    I never said it was facts, I just presented my opinion of Sun.

    Yes that is true, but they would have gone out of business but for the IT Boom. IMO.
  149. Sun has existed for 20 years[ Go to top ]

    I never said it was facts, I just presented my opinion of Sun.Yes that is true, but they would have gone out of business but for the IT Boom. IMO.

    I think the problem is that you state opinions as if they were facts. Putting 'IMO' in front of more things might lead to a more rational debate in future!

    But anyway, have fun with Avalon....
  150. Partial BS[ Go to top ]

    Well Peter I have insufficient information but I try to answer some of your questions now that it seems to boil down to how to use the SQL Server in the most efficient way.First some words of advice. I am a performance freak, no doubt about it. Unlike some that advocate "build now, optimize later" I think that performance is something that you should think about all time when you are developing , morning, mid-day and evening, even in the night, and not to be slapped down later.As to sub queries forget them, even normal joins should be avoided as far as possible, Another important measure is to of course to separate the reporting server from the production system (failover Clustering and lots of lots of memory of course) where you do inserts/Update/deletes.That is just normal common sense. But in your case I would advocate the more powerful SQL Server Analysis Services with OLAP cubes and dimensions and program against it with multidimensional expression (MDX) statements. Also, with Analysis Services there is some real performance advantages to be gained by using the new 64-bit version of the SQl Server.I apologize in beforehand if this is something you know all about already.http://www.microsoft.com/technet/prodtechnol/sql/2000/evaluate/ansvcs64.mspxFor a MDX primer look here,http://www.microsoft.com/msj/0899/mdx/mdx.aspxSQL Server 2005 (scheduled for release in early 2005) hosts the .Net Common Language Runtime (CLR). More info about SQL Server 2005 is here: http://www.knowledge-management.com/news/v4n8/yukon.htmlRegardsRolf Tollerud

    You're too funny. I've already mentioned in past threads that in some cases OLAP is not faster than SQL. I also gave a detailed explanation of the London Stock Exchange, based on my own research. Yet, you're still unable to provide concrete details. You can provide links until hell freezes over, but it doesn't demonstrate you know the subject. It onlly shows you can google. I find it hilarious that in all these posts, you haven't been able to give one concrete example. Someone will real experience could rattle off a hand full of examples. I'm sure Cameron could rattle off some stuff without any problems.

    The superior tool argument is bit lame and is no substitute for learning how Sql Server or Analysis service really works. You go ahead and keep using that excuse. I hope the next time you go for an interview, the prospective employer does a search and sees all your fluffy posts.

    I definitely wouldn't hire anyone that claims to be an expert developer and can't give write a query off the top of their head. If you want non-fluffy information on OLAP so that you can actually provide technical answers, I would recommend reading the olap report http://www.olapreport.com/. The author of the site has over a decade of in depth experience. Analysis Service is a good product by the way, but it's not a magic bullet. Proding you is getting much too boring now, so I'll let someone else point out the flaws in your logic for a while.
  151. an old rerun[ Go to top ]

    "What surprised me most was how the topic of Java, even among and between perfect strangers, always returned not to the technological flaws of our platform, but to the "jerk" attitudes of our developers".
    http://weblogs.java.net/pub/wlg/1343

    I hope you don't think you impress anybody Peter, your attitude is exactly the same as when we swept the floor with the mainframe programmers in the nineties. That was highly satisfying.

    I look forward to doing it again.
  152. Always keen to provide moral support to the good members of the society today we will examine two "sayings". The first is "He drank himself from his farm and business". The second is "Proudness goes before fall".

    When I was young and innocent I took these words literally. He lost his farm and business because he drank. Now I am older and wiser. When a person sees that his business soon will go into bankruptcy, he often takes to drinking because "it's more cool to fail because of drinking than because of incompetence".

    Let us also examine "Proudness goes before fall". Likewise, when I was young I took these words also at their face worth, "He become proud when he was successful and become careless and therefore failed."

    Now when I am a cynical and corrupt old fox (what happened to the young Rolf?) I know the real reason. The proudness comes after you know you are going to fail and is a way to postpone the catastrophe.

    Now you know.. If you are going to fall, you have basically two ways open to you.

    (Unless you are British of course, they never use squirming excuses.)

    Best regards
    Rolf Tollerud
  153. I wonder if some people are seeing Rolf as a representation of MS' community way of dealing with competition.
  154. nah, most people are reasonable[ Go to top ]

    I know a few .NET developers and Rolf doesn't represent the .NET community. Most of the developers I've met be it Java or .NET are reasonable and can see pros/cons. Those who can do. Those who can't start flame wars :)
  155. nah, most people are reasonable[ Go to top ]

    Well, the way I see it, that guy alone is doing a greater damage to MS community name than all Java zealots together could in a decade.
  156. nah, most people are reasonable[ Go to top ]

    I know a few .NET developers and Rolf doesn't represent the .NET community. Most of the developers I've met be it Java or .NET are reasonable and can see pros/cons. Those who can do. Those who can't start flame wars :)

    Absolutely. The big thing these days seems to be .Net/J2EE integration, with .Net being particularly successful on the client side. Both Sun and MS seem to have realised this: Rolf is one of the few left behind.
  157. :/[ Go to top ]

    Once again - why do u all post replies to his worthless FUD? Discussion with stupid, annoying and blind person like Rolf is pointless, ignoring ppl like him is the only valid way. Lets keep this site clear...
  158. :/[ Go to top ]

    Once again - why do u all post replies to his worthless FUD? Discussion with stupid, annoying and blind person like Rolf is pointless, ignoring ppl like him is the only valid way. Lets keep this site clear...

    Well, this is a well-known public forum where FUD can have an effect. Rolf has a technique of publishing what seem to be reasonable links. It would take a casual reader some research to discover that a link might be to a meaningless survey from years ago, or a few unrepresentative selected words from a blog entry. I have encountered such biased people at times in my career - they can have influence in the right places!
  159. :/[ Go to top ]

    You got point! :>

    On the other hand every reply to Rolf's humbugs brings another portion of FUD...
  160. :/[ Go to top ]

    Right Steve. Those who seem close to the truth are more dangerous than those who are obviously way out in left or right field.

    I have the same problem with research groups and mainstream content providers (Computerworld, SD Times, etc.)

    Find the FUD in the following article - http://www.computerworld.com/softwaretopics/software/story/0,10801,98910p2,00.html


    It kills me that Rolf keeps saying J2EE is so expensive when a full, quality Java/J2EE stack can aquired for $0. But he isn't alone in saying it.
  161. Google's search system FUD?[ Go to top ]

    "It kills me that Rolf keeps saying J2EE is so expensive when a full, quality Java/J2EE stack can aquired for $0. But he isn't alone in saying it."

    The Open Source alternatives are only expensive if you don't value your time, Mark. If we take Lucene for instance, used by this site, read carefully.

    <blockqote> On the surface, Lucene is "free". But how does it compare to commercial offerings?

    In terms of "quality", we've been very impressed with what we've seen. What's in Lucene seems of high quality, and is well documented. Since it is intended to be extensible, great effort has been taken to document it.

    But since this is not a turnkey product, so by default you will not have things like:

    Not an out-of-the-box turnkey solution. There is no "installer" or "setup wizard"

    No out of the box administration or command line tools. The demo code does offer some command line demos which could be leveraged.

    No "spider" to index your web site, though you may be able to find some code to do it

    No built in support for HTML format files, though one of the demos shows how to add it

    No built-in support for office documents such as MS Word, though you could add it

    No support for advanced XML queries, though some articles have been written about this topic.

    No support for Adobe's PDF format, but again, this could be implemented.

    No database gateway, though you could build one.

    No built-in Web Interface for searching, though they do have some sample JSP code

    No "help desk" or Tech Support phone number to call. </blockqote>

    http://www.ideaeng.com/pub/entsrch/issue03/article03.html

    Regards
    Rolf Tollerud
  162. News at 10[ Go to top ]

    Scientists have discovered that learning a new tool requires work. This previously unknown fact has astounding implications. With the discovery, scientist hope to get rid of all the diet fads and show the best way to loose weight is to do some exercize. Like wise, scientists believe this discovery will fix numerous problems in the IT world.

    Doctor Iam Knowitall states, "I don't know why my colleagues didn't discover this sooner. I believe it will cure many ills in our society. The days of over budget projects will come to an end. Things like the FBI fiasco reported this week by the news wouldn't happen."

    Will this new discovery prove to be the magic bullet? Only time will tell. We now return you to your scheduled programming (pun intended).
  163. Google's search system FUD?[ Go to top ]

    "It kills me that Rolf keeps saying J2EE is so expensive when a full, quality Java/J2EE stack can aquired for $0. But he isn't alone in saying it."The Open Source alternatives are only expensive if you don't value your time, Mark. If we take Lucene for instance,

    Correct me if I am wrong, but I thought Lucene was a search system, not a J2EE stack?

    By the same logic, permit me to criticise Microsoft database technology, using my dislike of Windows wireless networking as justification....
  164. Google's search system FUD?[ Go to top ]

    Yes, Lucene is an API.
    You can buy products that do all you've said. And I am sure a few that use Lucene.

    The question is, do you want flexiblity or the "straight and narrow" (don't get off the path)? I've found that the MS "straight-an-narrow" approach usually ends up costing more weh developing applications.
  165. Google's search system FUD?[ Go to top ]

    Yes, Lucene is an API. You can buy products that do all you've said. And I am sure a few that use Lucene.The question is, do you want flexiblity or the "straight and narrow" (don't get off the path)? I've found that the MS "straight-an-narrow" approach usually ends up costing more weh developing applications.
    Question is, is the cost of adapting such a flexible open source product higher than the cost of purchasing, consulting, support and maintenance of a "ready-to-go" closed source one? I don't think so, in most cases.

    Regards,
    Henrique Steckelberg
  166. an old rerun[ Go to top ]

    "What surprised me most was how the topic of Java, even among and between perfect strangers, always returned not to the technological flaws of our platform, but to the "jerk" attitudes of our developers".

    You quote this blog entry a lot. Let's see what else it says:

    "I've come to the conclusion that there is still a great deal of bitterness toward Java developers among the .NET rank and file."

    "As I attended various sessions on .NET technologies, especially one about developing web applications with ASP.NET, it became fairly clear to me that we [J2EE developers] still have a tremendous and important role to play in the future of enterprise computing, of persistent transactional object frameworks, of distributed component frameworks, of ultra robust application servers."

    "Microsoft does not seem to have an answer to EJB, or at least the developers I've spoken with don't spend much time considering distributed component architectures. They're still coaching developers to hand-write their SQL code—the very thing that drove me from ASP technologies back in 2000."

    You realise you are taking just one sentence from a blog entry of someone who suports the use of EJB and J2EE?

    Can we therefore assume that everytime you use this quote you are backing ALL his views?

    Let's look at more of the views of this person who you think is so wise:

    "Open standards allow multiple vendors to offer their own implementation of valuable technologies. The Sun and JCP open standards are particularly good at helping small vendors participate in the technology market. Competition makes software cheaper, forcing the vendors to improve the quality of their products and services. Open standards also lower the risk for the end users by at least providing for some common ground between technological platforms."
  167. The world is round[ Go to top ]

    "I know for sure that 90% of software in UBS Bank (here in Zurich) is powered by WebSphere and 80% of SW in Credit Suisse is powered by BEA WebLogic. Those are second and third biggest banks in the world. Hope big enough for Rolf. And yes Rolf, we know abour Einstain relitivity theory... "

    He's a troll..you could tell him the world is round and he would start talking about how the rain in spain stays mainly in the plains.

    I'm just glad he's being roundly spanked for being such an a**.
  168. J2EE[ Go to top ]

    What the heck you talking about? J2EE is a spec. it runs on mainframes also.
  169. ok[ Go to top ]

    oh, ok, i got it...you're talking about legacy mainframe apps.
  170. Bad developers[ Go to top ]

    You can do similarly horrible things with COM+ too, but that doesn't mean COM+ is evil

    Yes and to make thing more worst in MS world you even don't know it is a bug, an implementaion detail or both:-) I am amazed how MS guys keep pushing Java bees on the back foot by always saying J2EE is complex. I challenge any EJB developer to come to MS world and try to master COM+ and COM/DCOM and then go back to EJB. After that EJB would look choclate cake in ICE coating.... of course by then the big ego of being Java/J2EE architect and "Holier than thou" attitude would be vanished in space:-) My 2 cents...
  171. "Thank you, Alexander Tomorrow is the 10 of January. Exactly two years since the famous "TheServerSide Calls for Real World J2EE Project Stories"

    see the link

    http://www.sdn.sap.com/sdn/index.sdn

    The whole SAP NETWEAVER server (As you know SAP applications are the mother of all business applications) is based on Java and J2EE. If this is not the enterprise level application/Success story then I don't care how many years of experince you have, you simply hain't happen. And if SAP choose Java to do so, you better start believing in Java/J2EE too.

    And yes please don't just wait for TSS to give you the right story.... Java world is beyond these forums and BB.
  172. Cameron, your my hero.
  173. You forget that the so called monopoly was won in fair fight and competition with OS2, Apple, Unix etc.. At the time nobody had expected or predicted the Web including Tim Berners-Lee et all. Sun was desperately trying to sell Unix workstations, a business idea that had failed, and was saved in last minute by the miracle of the Web. Certainly no by any effort or brilliance from Sun.

    Of all the stupid things you have posted this has to be the worst. Either that or you were not coding in the early 90’s.
  174. the end of the browser[ Go to top ]

    Hello Rolf,

    as a professional computer scientist I did many projects with many different technologies. And history repeats itself quite frequently in our business. That means most times there isn't a radically new programming language or framework or technology. The education and experience I have, the technologies I learnt in the past, help me to learn new technologies very quickly when I need to use them. Also: The technologies from Microsoft have always been very mass compatible, that means quite easy to learn for every it professional. I doubt that Microsoft can come up with a technology I cannot learn in two weeks. And if so, there isn't much advantage learning something that still has the probability to fail. I'll learn it when I need it...

    Cheers
    Michael
  175. basically you are right. But,[ Go to top ]

    Hi Michael,

    "I doubt that Microsoft can come up with a technology I cannot learn in two weeks. And if so, there isn't much advantage learning something that still has the probability to fail. I'll learn it when I need it..."

    For the first time I have a set of my own developed widgets - tab control, tree view, lists and buttons etc looking completely professional in my own personal style, and it was not even hard! In spite of that the last thing I would call myself is a designer.

    And to astonish your colleagues with a interface that compound 10 normal web pages into one!

    I can not resist, it is too fun.

    Regards
    Rolf Tollerud
    (to think that Gerald Bauer was right all along..hmm)
  176. Learning a new framework in 2 weeks?[ Go to top ]

    I doubt that Microsoft can come up with a technology I cannot learn in two weeks.

    To say that a new technology can be learned in two weeks is a bit far fetched, to say the least.
    Aer you talking of actually learning remoting, web services (using .Net APIs of course), code access security etc. all in two weeks??
  177. the end of the browser[ Go to top ]

    First you have to have the language in place. To release .NET was a major undertaking, more expensive than sending man to moon according to Bill Gates.http://theserverside.com/news/thread.tss?thread_id=30730#150569RegardsRolf Tollerud
    What percentage of that was paid for abusing testers?
  178. What I can see ......[ Go to top ]

    As a firm believer in market reality, I am learning .NET not to stay with outdated skills ....

    If you are a good Java developer, shouldn't take very long. Mostly you'll have to figure out how to do things with out all that is available in Java.
  179. Meanwhile back on the thread[ Go to top ]

    I don't know how much of a trend this really is but because I develop with both sets of technologies I get a fair number of enquires. I’ve seen two projects in the last year that were .NET based (both proof of concept) that were moved to J2EE for full development, and another couple of ASP/VB production projects that were the same. Both the .NET projects hit problems around exception handling (not implementing checked exceptions – big mistake), and performance (a lot of the ASP components seem to run very badly when put under strain and the framework as a whole seems to struggle under heavy load). A lot of time was wasted working round bugs in the framework and so on – all the signs of something that was rushed out. All in all .NET is very good for instant gratification, but maybe not yet ready for prime time.

    I've also seen a number of vendors with VB/ASP based solutions taking ht plunge to J2EE. A major reason I find for this from vendors is a sense that since Microsoft has made our existing ASP/VB/COM+ code base obsolete at a stroke and moving to .NET offers us zero return on investment, we might as well take the plunge to J2EE and open ourselves up to non-windows companies (ex-MS employee and former Excel VBA team leader Joel makes this point rather well in
    <ahref="http://www.joelonsoftware.com/articles/APIWar.html">
    this article</a> though he stops short of saying "I'll move to Java then."

    I can't say I've ever seen a J2EE project migrated to ASP or .NET.

    On the whole I tend to think that .NET is pretty good news for J2EE. Whilst Java has a pretty competitive internal market new competition is generally a good thing and there are some quite good ideas emerging in .NET that J2EE can take inspiration from. I do think that .NET was a huge mistake for Microsoft as a whole though. As well as annoying a lot of their develops there is the pretty obvious point that post the dot com bubble bursting there was a brief opportunity for a new, low cost vendor to enter the data centre’s of large enterprises (where Sun, IBM et al had been so successful). Most analyst reports I read around 1999 were convinced that Microsoft would be able to make serious inroads here. This relied though on Microsoft spending a lot of time and money making a version of Windows that was suitable for such a deployment – one where, for example, the GDI didn’t run in Kernel space and where an application couldn’t just arbitrarily post anything to another applications event queue (the later being the most obvious example of why Windows cannot ever be secure), and the OS understood memory management (lessons the rest of the industry worked out in the 70's) and had a properly scalable threading model and…I digress. Had Microsoft spent its money on getting a version of Windows that was suitable for such a market and embraced J2EE as a development platform it could have hit that opportunity hard around 2001 and probably made some serious inroads. As it was, of course, since Bill Gates suffers from "Not invented here syndrome" (or penis envy) it poured money into building a J2EE clone so he could say "look how innovative we are – we've built this revolutionary new way of writing enterprise apps that is nothing like Java, honest", took money away from the core Windows development team, and missed the opportunity (the hole now being plugged by Linux). The net (pun intended) result of all of this is that Garnter's prediction that Microsoft would rule the data centre by 2005 starts to look rather foolish and .NET (which was far too late into the space anyway) starts to look rather lame.
  180. The power of foolishness[ Go to top ]

    Don't underestimate the power of marketing and human foolishness.

    There are a lot of small and mid-sized businesses that trust Microsoft (mainly due to the fact they are familiar with it) and have been infected with the wrong impression that J2EE is overly complex for their in-house people to learn. I blame this on the propensity of some vendors to try to sell their very expensive app servers and use stuff like EJBs for what should be lightweight apps.

    Thus, the encroachment of inexpensive, open source Java tools, servers, etc is a godsend that hopefully will change this attitude over the long run.
  181. This is a valid trend[ Go to top ]

    It proves nothing but I am unfortunately seeing the opposite trend (increasing interest in / use of .NET) in many of my clients (mostly large financial organisations).

    I've yet to see a single back-end system in a financial services company being built in .NET. Forget "production use" or whatever, I've yet to see one even planned. Not to say that there aren't any .. just I haven't seen any.

    I _have_ seen Windows UI projects being done in .NET for financial services. Most are fronting J2EE back ends .. pretty funny.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  182. This is a valid trend[ Go to top ]

    I _have_ seen Windows UI projects being done in .NET for financial services. Most are fronting J2EE back ends .. pretty funny.Peace,Cameron Purdy

    Absolutely. This is a huge market for .Net - simply as an upgrade for Visual C++/VB6 for Windows UI development. Not competing with Java at all.
  183. This is a valid trend[ Go to top ]

    The day you see financial services use .net, or Windows for that matter, will be the day you see Pres. Bush saying Iraq was a mistake.
  184. last in last out[ Go to top ]

    "Insurance and banking business will always stay on the trailing side of adoption curve due to their conservative technology policies. What Java replaces there is OS/2, mainframe and proprietary client/server solutions".
  185. last in last out[ Go to top ]

    "Insurance and banking business will always stay on the trailing side of adoption curve due to their conservative technology policies. What Java replaces there is OS/2, mainframe and proprietary client/server solutions".

    Woah! Wait a minute. You can't have it both ways. You can't put forward a quote saying that financial businesses are implementing Java (obviously J2EE) solutions, after having said that it's so difficult to find real life projects.

    So - either Java IS replacing older solutions (in which case you were wrong about the number of projects), or it isn't (in which case the above quote is wrong). Which is it?
  186. last in last out[ Go to top ]

    Rolf: What Java replaces there is OS/2, mainframe and proprietary client/server solutions.

    Rolf, does "proprietary" mean "not Windows" or "not .NET"? I'm confused.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  187. hopeless case[ Go to top ]

    Cameron you are always confused. I can't help you.
  188. Dinosaurs[ Go to top ]

    The day you see financial services use .net, or Windows for that matter, will be the day you see Pres. Bush saying Iraq was a mistake.

    Exactly. Financial services won't adopt something until it's PROVEN itself stable and secure. Which is why they're going to J2EE, mostly, not .NET, for their upgrades.

    Call them technical dinosaurs if you want, but they have to operate this way.
  189. what? no one wants to loose millions[ Go to top ]

    why should banks and financial firms worry about stability :) Not like the system going down is going cost x million per hour it is down. Just do what Quicken Turbo tax used to do, have sevaral hundred NT boxes and have a team of MCSE run around rebooting the servers when it crashes.

    </sarcasm>
  190. everything is relative[ Go to top ]

    Thats bad, but at least it is not as bad as the J2EE servers, which is down nearly seven hours of time per week, or worse according to Wily:
    http://www.eweek.com/article2/0,1759,1388603,00.asp

    But of course the best reason to run Windows is security:

    Linux and open source software accounted for 16 out of 29 advisories. Traditional UNIX software - 16. Windows - 7.
    http://www.unixsucks.com/unixsucks.aspx

    Regards
    Rolf Tollerud
  191. Perspective[ Go to top ]

    My first hand experience is Unix + Java is 2-4x more reliable. Even though I was joking around, Turbo Tax did have a lot of reliability and performance issues in the past. I chalk that up to lack of experience and under estimating public demand. Many of these differences are the result of hiring good developers, so I don't attribute it to a specific platform.

    I think in large part Unix has a higher learning curve than Windows, which ends up like a filter. I have no concrete proof of this. Being easy enough that anyone can learn has its negatives. It means you have to go through a larger number of applicants to find one that is really top notch. What makes it worse is when certifications are so easy that it ends up being rote memorization.
  192. the truth about Unix[ Go to top ]

    Peter, please read the whole http://www.unixsucks.com/unixsucks.aspx
    carefully.

    Not only are Windows more secure and more stable but,

    "Windows (I assume Server 2003) has shown on average 2x performance in file server tests and almost 3 times performance in web server tests".

    Regards
    Rolf Tollerud
  193. the truth about Unix[ Go to top ]

    Rolf,

    You're using a site called UNIXSUCKS.com as backup? I mean come on, learn how to at least have a modest appearance of independent thought.

    Really though, that whole site quotes a bunch of studies out of context or ones that are almost embarrassing in their total lack of independence or poor methodology (MindCraft, Mi2G, Aberdeen).

    I'm almost embarrassed by your little diatribes. In reality, you probably do more harm to to the Windows world than J2EE or Linux.

    Jason McKerr
  194. you're too funny[ Go to top ]

    that page presents a pretty weak argument. plenty of people have written much better critique of Unix than that blog. Let's keep in mind that NT and Unix both were influenced by VMS and "time-sharing" systems. both were also influenced by microkernel, which was created by academic researchers. You know, the people you keep saying are ivory towers :)

    i won't bother listing all the architectural differences between NT and Unix, since you're probably already aware of them. Each system has strengths and weaknesses. If someone were to tell me "windows can replace Unix systems for backend Telco" I'd laugh pretty hard. Same is true if someone tells me "solaris is the best choice for a local file server."

    Every case should be decided based on the needs and the staff. If a company is made up of Unix guys, why would one want to use windows? Likewise, using windows for a local fileserver is much nicer and easier for business users. Also remember that users change and their needs change. No one solution is always right or right for forever. At best, one can hope it meets the needs long enough to make the investment worth while.
  195. the truth about Unix[ Go to top ]

    As somebody said, tour problem (apart from the personal one) is that your statements are not based on real life experience but on all of the MSFT marketing, and all we know what type of marketing MSFT has.

    I used to work with BEA on project all over Latin America (and of course not on little stores projects) and i can promise you those systems are not down a day per week, if you say so, you need to probe it in a better way that just pointing us to that king of reports.

    Just clarify (i am sure nothing is gonna change your point of view as i have seen on all your posts) but the report of your reference an infamous one. In fact that report lead by Middleware but FINANCED BY MICROSOFT was a real joke. Middleware itself recognize that

    http://www.middlewareresearch.com/endeavors/021028J2EEDOTNET/artifacts/Rematch%20Message.pdf

    Microsoft indeed has to recognize it to, a shame i don't found the link on the Microsoft site, oh well, may be was erased after couple hours on their site.

    I invite you to expose commentaries based just on your experience and not on this MSFT biased studies.

    Of course, just a commentary from one of the IMMATURE,EMOTIONAL, IGNORANT and UNCULTIVATED around.

    NOT peace (sorry Cameron not me)
  196. Rolf[ Go to top ]

    Rolf is a troll. Why so many bother to respond to his posts is beyond my ability to comprehend (and most of you are far more intelligent than I am). His trolling isn't even unique or special in some way. He just sits on these boards and makes snide comments about Java/J2EE subjects in an attempt to ruin the mood of the people who enjoy discussing these topics. He even does this on multiple websites. His sole goal in life is to dog Java/J2EE technology wherever and whenever he can. This is classic trolling folks. How many specious arguments does he have to make before it is clear?

    Ignore him and he will eventually get bored and go away.

    And Rolf, respond to this all you want. I don't give a damn what you have to say about anything. *hugs*
  197. Marc[ Go to top ]

    I don't consider myself a troll, period.

    But if I were, I would be the most successful troll ever!
  198. everything is relative[ Go to top ]

    Thats bad, but at least it is not as bad as the J2EE servers, which is down nearly seven hours of time per week, or worse according to Wily:http://www.eweek.com/article2/0,1759,1388603,00.asp

    Invalid. That conclusion was from a small self-selected sample. Being trained in statistics, I can tell you it's meaningless. Also, the survey was from a company that sells J2EE monitoring services. Are they likely to publish a survey that mentions J2EE installations that have no problems with downtime?
     
    But of course the best reason to run Windows is security:Linux and open source software accounted for 16 out of 29 advisories. Traditional UNIX software - 16. Windows - 7. http://www.unixsucks.com/unixsucks.aspxRegardsRolf Tollerud

    Nice try. That report is more than 2 years old. Again, statistically, if you pick the right time period for a sparse data set, you can always 'prove' one system more secure than another. It also compares all 'open source and Linux' with just 'Windows and Microsoft products' (not all products on Windows), so they are not sampling equal sets ('Linux' tends to include the entire distribution of kernel, tools, X11, and bundled apps). Another factor is the open source vunerabilities are far less serious, as apps almost never run with administrator permissions.
  199. everything is relative[ Go to top ]

    But of course the best reason to run Windows is security:Linux and open source software accounted for 16 out of 29 advisories. Traditional UNIX software - 16. Windows - 7. http://www.unixsucks.com/unixsucks.aspxRegardsRolf Tollerud

    Have you ever thought about taking your act on the road?
  200. but how would Rolf post?[ Go to top ]

    If Rolf went on the road, how would he post to TSS? It would be a bit challenging to type out long responses on a WAP phone. On the upside, it would be a great workout for the thumb. It's funny, the last 2 weeks there have been 2 exploits for windows. Today cnet is running an article about hackers using windows media player's DRM to install spyware.
  201. the folly of self-suggestion[ Go to top ]

    Peter,

    Windows Advanced Server 2003 (a far more secure OS than any Unix/Linux system) doesn't come with a Media Player.But I forgive you. After all it is quite common to confuse client and server.

    Henrique,

    I suggest you read the book again..

    Fred,

    "What surprised me most was how the topic of Java, even among and between perfect strangers, always returned not to the technological flaws of our platform, but to the "jerk" attitudes of our developers".
    http://weblogs.java.net/pub/wlg/1343

    "The best argument against Java is a five-minute conversation with the average Java user." - Rolf Tollerud

    You may quote me.
  202. I've used 2K3 server[ Go to top ]

    Having used it last year, you're absolutely right that by default it does not install windows media player. An admin that installs media player on advanced server needs to slapped down real hard.

    Then again, I've heard horror stories in the past at user groups for MCSE. The funniest one is when newbies installed a ton of stuff and completely opened up the server. Needless to say the server gets hacked.

    The default settings for 2K3 server and advanced edition is very restrictive and the shutdown opens are much better than professional. Back in 2003 when virii were going around like crazy and causing professional edition to reboot non-stop, several people where I worked had their machines booted from the network by the admins. A dozen people didn't get their machines working for 2 weeks, so they had to borrow other people's systems. I'm a bit paranoid myself, so I update and patch regularly and put 2 firewalls between the net and my systems.
  203. oops typo[ Go to top ]

    that should "shutdown options". win2K3 server has additional options and is 3-4x more restrictive than 2K server. atleast that's my bias opinion.
  204. Plea[ Go to top ]

    "Windows Advanced Server 2003 (a far more secure OS than any Unix/Linux system)"

    The desperate pleas of an entity whose pie is slowly being eaten away by Linux.
  205. Plea[ Go to top ]

    "Windows Advanced Server 2003 (a far more secure OS than any Unix/Linux system)"The desperate pleas of an entity whose pie is slowly being eaten away by Linux.
    Some people are complaining they are unable to play Warcraft 3 on Windows Advanced Server 2003... Bad Microsoft, bad!!! :)
  206. not only that[ Go to top ]

    other RPC services don't work too. On one of the server we couldn't get COM+ authentication to work at all. I spent like a month digging around to find the cause and fix it, but we never could figure out why a client on a fully patched win2K advanced server going out to another server using COM+ authentication wouldn't work. We finally gave up and just used OleDB and bypassed COM+ altogether :)
  207. When you say that both EJB and COM+ should be avoided as much as possible I agree with you. But sometimes you have to.

    And in that case it seems that COM+ has done something right, seeing that all 10 places in the top 10 Top Ten TPC-C benchmark by Price/Performance use COM+ as TP Monitor, even place no 4 which is IBM DB2 running on Linux and place no 7 which is Oracle 10g running on Linux!

    On the other hand I can not see any EJB action at all.

    http://www.tpc.org/tpcc/results/tpcc_price_perf_results.asp?resulttype=noncluster

    Regards
    Rolf Tollerud
  208. I would say both COM+ has come a long way since MTS and has benefited from the lessons learned buidling Tux/Tuxedo. Let's remember that COM+ owes a lot to the pioneering work by ATT researchers. Where would COM+ be without Tux and Tuxedo?
  209. Rolf- you're a loon! LOL[ Go to top ]

    Rolf:

    You ARE a loon...as I've said many times to many people I've debated with since 1997 (and i've known many who are now gone from usenet), I'll still be here laughing at you 10 years from now, and have no doubts, Java will STILL be here - whereas .NET, well, we ALL know how microsoft likes pulling the rug from under their developers when it suits them, eh?

    And btw, Java is

    1. The #1 (or #2, depending on the source) most-wanted IT language JOB skill today...

    2. Is second only to C in that tiobe index (which btw, measures only popularity in terms of hype on the web, so guess where c-sharp is, eh?)

    3. Supports a much larger and faster growing book publishing ecosystem than other languages, including PHP, Perl, C++, C#, and VB. The survey looked at the number of books published in amazon.com for each year since 1995.

    http://www.blueboard.com/phone/publishing_capacity.gif
    http://jroller.com/page/kalimantan/20040510#java_accelerating_above_perl_php

    The fact that you spend so much of your time (and perhaps someone's money eh?) jumping at shadows over here indicates how strong the platform is!
  210. It is easy to have a fabulous growth when you have such a tiny market share. Even with a predicted growth of 20-30% per year until 2008, market share figures will be: Linux, 13%, Windows, 39.1%, and Unix, 38.1%.

    But all the growth is taken from Unix which still is the dominant server OS. Windows has a steady growth at 8%, the inflection point for Windows to take over Unix as the dominant market share server is 2007.

    When Linux has taken all the Unix share it can take (I guess something will be left), - the story ends.

    http://asia.cnet.com/enterprise/specialreports/0,39035905,39202771-4,00.htm

    Regards
    Rolf Tollerud
  211. It is easy to have a fabulous growth when you have such a tiny market share. Even with a predicted growth of 20-30% per year until 2008, market share figures will be: Linux, 13%, Windows, 39.1%, and Unix, 38.1%. But all the growth is taken from Unix which still is the dominant server OS. Windows has a steady growth at 8%, the inflection point for Windows to take over Unix as the dominant market share server is 2007. When Linux has taken all the Unix share it can take (I guess something will be left), - the story ends.http://asia.cnet.com/enterprise/specialreports/0,39035905,39202771-4,00.htmRegardsRolf Tollerud

    That is only for Pacific Asia. IDC forecasts that Linux server shipments will have 25.7% of world-wide market share in 2008 - more than double your figure.

    But, of course, this is all statistical nonsense. Anyone who tries to predict figures to this accuracy three years ahead is wasting their time.

    By the way, that link you gave has bad news for Windows: "Unix will continue to be the platform of choice for transaction and computer intensive applications.".

    As for 'tiny' market share: Linux has 50% of the server blade market, 20% of rack-mounts, and 11% of stand-alone machines.
  212. IDC forecasts that Linux server shipments will have 25.7% of world-wide market share in 2008 - more than double your figure.
    Maybe till that time someone will pull couple of ideas from Windows 3.0 API and will provide Linux (oh, I am sorry, X Window... or Gnome or KDE) with decent clipboard support.
  213. But I though everybody predicted that just the Asia market would embrace Linux?

    "But, of course, this is all statistical nonsense. Anyone who tries to predict figures to this accuracy three years ahead is wasting their time."

    Agree. And just for the record what is important is only the enterprise market, real applications hosted on servers set up and owned by the company. The low-level market with simple html pages hosted on shared servers is uninteresting.

    That the low-level market is included in all surveys and statistics skew the results heavenly in Linux favor.

    Regards
    Rolf Tollerud
  214. F*rting 'gaints the wind[ Go to top ]

    Btw, Rolf, you DO know you're just f*rting in the wind here, right?

    Everyone actually knows how lousy a debator you are, and how loosely you play with facts, and how easily you cringe away from real arguments.

    Hint: You're actually hurting your cause (Microsoft's salvation) by being here, not helping it - LOL.
  215. do try to contain your excitement[ Go to top ]

    "Everyone actually knows how lousy a debator you are"

    But I don't have to be good debater San Juan, because the whole Sun/J2EE community is like a house of card. Java survives at all only because of the effort from guys like Rod Johnson, Juergen Hoeller, Howard Lewis, Richard Öberg, Gavin King, etc.

    But it is not enough. Java have now dropped down to 17% in the TIOBE Programming Community Index, from 25% last year.

    Why don't you jump the ship? Read Cedric Beust summing up the year 2004:

    "Never mind the fact that the past eight years have all been predicted as the "year of Linux", there are quite a few signs that make me think that if anything, this year will be the year of Windows"
    http://www.beust.com/weblog/archives/000070.html

    Regards
    Rolf Tollerud
  216. Donald Duck[ Go to top ]

    btw, are you actually gonna look at the valid points or just duck and cringe into new and trivial territory when your "arguments" are punctured?

    I believe I'll call you Donald Duck from now on...you duck and duck, and quack, quack, quack... ;-)
  217. do try to contain your excitement[ Go to top ]

    Java have now dropped down to 17% in the TIOBE Programming Community Index, from 25% last year.Why don't you jump the ship?

    You need better links to prove your case. This same index shows Visual Basic down to 8%, and C# at an negligible 2.2%!

    If you are using this link to show Java is decreasing, you also have to accept that .Net is dead or hardly used at all.
  218. Steve,

    "By the way - there is something else that skews the 'market'. Linux can be free. There is a huge volume of Linux installations that don't could as 'sales of Linux'"

    That is true but that is more than offset by all pirate installations of Windows. In many places in the world it is normal to sell PCs with Linux to avoid the MS-license but the first thing they do is to install Windows.

    "the enterprise market is currently (and likely to remain) dominated by neither Linux, or Windows but Unix"

    Yes, but Windows is close behind. I think it is safe to say that all big organizations has a lot of both Unix and Windows.

    "It may be uninteresting, but it's vital for businesses"

    Naturally, only that which is vital for enterprise businesses, is interesting.

    There are high level phones and low-level phones. I, I am interested in phones that can interact with the companies business systems. For example with the CRM and Sales Automation systems. To play melodies and simple games I gladly gives away to Sun and Java (I heard that they talked a lot about that at JavaOne) or to anyone that is interested.

    The same goes for hosting companies that host hundreds or thousands of users on cheap Linix boxes. That makes sense, Window Server would be overkill, but again - not my business.

    As long as the Sun/Open Surce/Java world are so desperate to appear successful that count 1 Server with 1000 IP-addresses as 1000 Apache-users we will never any correct statistics over the important question: "What technology is used in today’s enterprise business".

    "C# at an negligible 2.2%!"
    The only thing we can count upon with TIOBE Index is that they accurate measures trends, the absolute numbers remain uncertain. C# remains constant, Java has gone from 25% to 17%, that is a (25-17/25)*100 = 32% decrease in one year.

    Regards
    Rolf Tollerud
  219. In many places in the world it is normal to sell PCs with Linux to avoid the MS-license but the first thing they do is to install Windows.

    That is a serious allegation. Do you have evidence?
    "the enterprise market is currently (and likely to remain) dominated by neither Linux, or Windows but Unix" Yes, but Windows is close behind.

    No, not according to your link. 'Dominated' means there is no 'close behind'. See your dictionary.
    I think it is safe to say that all big organizations has a lot of both Unix and Windows.

    So now you agree there is a lot of Unix.
    There are high level phones and low-level phones. I, I am interested in phones that can interact with the companies business systems. For example with the CRM and Sales Automation systems.

    Like the most successful current phone/pda - the Java-enabled Blackberry.
    The same goes for hosting companies that host hundreds or thousands of users on cheap Linix boxes. That makes sense, Window Server would be overkill

    This is because you CAN host hundreds of users on Linux boxes.
    As long as the Sun/Open Surce/Java world are so desperate to appear successful that count 1 Server with 1000 IP-addresses as 1000 Apache-users we will never any correct statistics over the important question: "What technology is used in today’s enterprise business".

    So, Windows is better because you need more installations (and need to pay a lot more) to support the same number of users?
    "C# at an negligible 2.2%!"The only thing we can count upon with TIOBE Index is that they accurate measures trends, the absolute numbers remain uncertain. C# remains constant, Java has gone from 25% to 17%, that is a (25-17/25)*100 = 32% decrease in one year.RegardsRolf Tollerud

    So by the same argument, you would recommend that developers leave the non-existent .Net and switch to PHP or the rapidly growing Delphi?

    If you want to use search engine results (as against something more sensible - the job market) to judge the popularity of a language (a highly suspect thing to do), that is what you have to conclude....

    I think these figures are nonsense - even I don't believe that .Net is that rarely used!
  220. Blackberry[ Go to top ]

    "Like the most successful current phone/pda - the Java-enabled Blackberry. "

    I believe, although i haven't checked recently, the higher end blackberry's are not only "Java-enabled", they are actually "java-powered", as in Java is the OS or a prime component of it, or all apps are java apps.

    http://mindprod.com/jgloss/blackberry.html

    The interesting thing about this pda/phone is that more and more customers seem to want to standardize on it in the pda realm rather than palm or windows. A potential client mentioned this to me just a couple weeks ago, and to be honest about it, pleased the heck out of me.

    I use palm, but the darn company has given short shrift to Java for the most part, although its recent steps into the enterprise market with IBM and MIDP are heartening.
  221. same old story from palm[ Go to top ]

    "Like the most successful current phone/pda - the Java-enabled Blackberry. "I believe, although i haven't checked recently, the higher end blackberry's are not only "Java-enabled", they are actually "java-powered", as in Java is the OS or a prime component of it, or all apps are java apps.http://mindprod.com/jgloss/blackberry.htmlThe interesting thing about this pda/phone is that more and more customers seem to want to standardize on it in the pda realm rather than palm or windows. A potential client mentioned this to me just a couple weeks ago, and to be honest about it, pleased the heck out of me.I use palm, but the darn company has given short shrift to Java for the most part, although its recent steps into the enterprise market with IBM and MIDP are heartening.

    back in 2000, I worked on a wireless portal platform. we tried to work with Palm, but they were hesitant. If I remember correctly, many people including Sun tried to get Palm to go for J2ME, but they did for one reason or another.

    It's nice to see wireless applications start to take off, but back in 99/2K, we build a wireless portal capable of tracking where users. We could get a user's location based on the network or GPS. Other people could then track the user on a website and see where the person was. The accuracy varied depending the accuracy of the triangulation, but in some cases it was 100 feet. We even had prototypes with a custom client, which could download b/w maps from our server. We also did things like scrape Caltrans data to show traffic conditions and try to send the user a WAP alert. We never got as far as re-routing the user, due to a shift in priorities. Even today, the single quality of TDMA/GSM/CDMA in the US is so bad that it's really hard to make wireless applications reliable.

    The company I worked for had a research division, which designed and built CDMA scanners, so the phone engineers could do some pretty cool stuff. We also wanted to but a Java RTOS on the phone, but the cost was too high back then and the memory requirements were much higher than comparable client written in assembly.
  222. When someone mentions "enterprise" I think of a company with like 10-100K employees. When someone says "enterprise software" I read it to mean nothing, since it could be an office suit, or the company accounting system. Not all "enterprises" are created equal :)
  223. it is hopeless, all is in vain[ Go to top ]

    "When someone mentions "enterprise" I think of a company with like 10-100K employees"

    Yes, but companies with only 100-500 employees is still interesting, but shared hosting is not.

    Anyway I get a lot of flack because I post here. You could perhaps guess that people would think it interesting to get an second opinion. But no.

    Reason, logic and common sense. How rare it is! You can see from Steve’s answer that he do not understand enough statistics to keep separate absolute numbers and significant trends.

    BTW, another thing you never can get the TSS members to understand even in a 100 years is the effect of the combination of the copyright-law and the GPL-license.

    So why try to educate the masses? It is hopeless anyway. For 20 (or was it 30?) years ago they introduced a new kind of doors in the public busses in my home city, that opened automatically when you stepped down towards the exit. But still, after all this years, most people stand there and shout for the driver to open the doors.

    Regards
    Rolf Tollerud
  224. it is hopeless, all is in vain[ Go to top ]

    Anyway I get a lot of flack because I post here.

    and you deserve it well! :-)
    You could perhaps guess that people would think it interesting to get an second opinion. But no.

    what you offer is not a 2nd opinion, it's poor FUD and provocation.
    So why try to educate the masses? It is hopeless anyway. For 20 (or was it 30?) years ago they introduced a new kind of doors in the public busses in my home city, that opened automatically when you stepped down towards the exit. But still, after all this years, most people stand there and shout for the driver to open the doors.

    ok, understood: people are idiots, except you of course.

    Rolf, there was a time when you at least *tried* to successfully make your case. It really seems that you have given up completely and just chose to take the easier path (which is "trolling without a cause")

    best,
    Joe
  225. it is hopeless, all is in vain[ Go to top ]

    Reason, logic and common sense. How rare it is! You can see from Steve’s answer that he do not understand enough statistics to keep separate absolute numbers and significant trends.

    Excuse me! You were the one who (falsely) said that linux growth was irrelevant because it was from a small base, then later when I pointed out the small base of C#, you said that all that mattered was trends!

    I think you need to study statistics more. Perhaps I should point you to some of the resources I used to use in the statistics courses I taught.
  226. it is hopeless, all is in vain[ Go to top ]

    Steve,

    What you do is called "guilt by association".

    a) The TIOBE index says nothing about the real percentage of various programming languages. It is most probably wrong.

    But from that a) is wrong does not follow that b) is wrong. You make fun of a) and everybody laughs and say, those morons can not have right in anything. That is your mistake.

    On the contrary, b) is statistically significant (I am sure you know what that professional meaning of that) and is independent from a).

    Regards
    Rolf Tollerud
  227. I'm sure it wouldn't take more than a few minuts for you to write up a formal mathematic proof :)

    I'm just joking.
  228. it is hopeless, all is in vain[ Go to top ]

    Steve,What you do is called "guilt by association".a) The TIOBE index says nothing about the real percentage of various programming languages. It is most probably wrong.But from that a) is wrong does not follow that b) is wrong. You make fun of a) and everybody laughs and say, those morons can not have right in anything. That is your mistake.On the contrary, b) is statistically significant (I am sure you know what that professional meaning of that) and is independent from a).RegardsRolf Tollerud

    Look. If the TIOBE says nothing about real percentages, then, statistically, it can say nothing about real changes in percentages. If you have a nonsense measurement, you can't make sense of changes in it. I could link Java popularity to the number of geese flying South over my house each year. It's nonsense - there is (I hope!) no connection. Suppose fewer geese fly South over my house this year - do I assume that this is a statistical measure of decreased Java popularity?

    The TIOBE index is based on analysis of search engines. Search engines don't index more than a fraction of the web. You can interpret the figures any way you like. If you want some reasonable, independent, and statistically meaningful measure of popularity, look at the job market, and sales and downloads of developer tools and books. But you won't, as this clearly shows you are wrong in almost every respect.

    I have no more time to waste on this. I have been posting because I worry that some developers will see your posts on this site and assume that they have some validity, and I also get annoyed about bad use of statistics. I shall leave it to more patient people to counteract your FUD.
  229. it dawns upon me[ Go to top ]

    " have no more time to waste on this"

    OK.

    Just remember to go to your old shool and get the money back.

    Now I know the reason I spent so much time in TSS.COM. It is the only place I can feel like an intellectual giant! :)

    Regards
    Rolf Tollerud
  230. delusions of granduer?[ Go to top ]

    " have no more time to waste on this"OK. Just remember to go to your old shool and get the money back.Now I know the reason I spent so much time in TSS.COM. It is the only place I can feel like an intellectual giant! :)RegardsRolf Tollerud

    that is a funny joke. I suspect without any proof, you like to goad people on. on a totally unrelated note, I hope the latest miniMac becomes a huge hit and people abandon windows in droves. that is one sweet little box.
  231. it dawns upon me[ Go to top ]

    " have no more time to waste on this"OK. Just remember to go to your old shool and get the money back.Now I know the reason I spent so much time in TSS.COM. It is the only place I can feel like an intellectual giant! :)RegardsRolf Tollerud
    So you really think that you are getting smarter everytime someone prooves you're wrong... hum... You must be really smart by now then, Einstein!!!! :)
  232. it dawns upon me[ Go to top ]

    Now I know the reason I spent so much time in TSS.COM. It is the only place I can feel like an intellectual giant! :)RegardsRolf Tollerud

    Well, Rolf, why didn't you admit that on the beginning? Now, when we know the root of your problem, we can focus on how to help you free yourself from whatever complex you have and make you feel confident in the real world, too!

    Lets not waste more time!
    So, how old were you when those terrible frustrations started appearing? Did other kids teased you for some reason back then?...

    And remember: thou you sometime annoy us, we still love you!!! :-)
  233. Yes, it is hopeless... call a doctor[ Go to top ]

    Hmm, Rolf...
     
    Funny how your name/surname is similiar to the word 'troll' (yeah, cheap shot, but still funny). I don't believe that this is coincidence :>

    Anyway, this is my first and last reply to your FUD. I really don't know why ppl keep trying to write sensible replies to your stupid, tiring posts - it is pointless.
  234. cheap form of entertainment[ Go to top ]

    the way i look at it, it's a cheap form of entertainment. Kinda like reality TV, but without the TV or reality.
  235. by the way win2K3 rocks[ Go to top ]

    http://news.netcraft.com/archives/2005/01/12/apple_store_macworld_expo_sites_slowed_by_heavy_traffic.html

    Looks like win2K server has improved tremendously, but more work is needed. I wonder why win2K server crashed and took hours to recover at macworld.

    care to guess :)
  236. for those humor impaired[ Go to top ]

    In case someone didn't realize it was a joke, Apple uses Akamai to distribute the load, so chances of their site going down is rather unlikely. I'm guessing Macworld doesn't use Akamai, since it's expensive.
  237. Longest Uptimes for servers[ Go to top ]

    Another interesting stat I came upon:
    http://uptime.netcraft.com/up/today/top.avg.html

    Linux is not usually detected by the survey which explains its absense. The list is dominated by UNIX-type servers. Only 1 windows.
  238. As long as the Sun/Open Surce/Java world are so desperate to appear successful that count 1 Server with 1000 IP-addresses as 1000 Apache-users ..

    Yeah, those grapes were sour anyways ..

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  239. VB Hamsters tired of MSFT Treadmill[ Go to top ]

    "You need better links to prove your case. This same index shows Visual Basic down to 8%, and C# at an negligible 2.2%!
    If you are using this link to show Java is decreasing, you also have to accept that .Net is dead or hardly used at all."

    Yep, C-sharp has hardly moved from that spot since a couple years back, which REALLY bodes well for .NET if one goes by that single index - hehe...and note that since September, Java (and C, the leader) has shown a very large spike upwards (look at the bottom graph) after both falling continuously in 2004.

    VB though has to be going down, since all other stats show it is going down - one survey done by gartner i believe showed that up to 31-32% of VB shops were moving to Java - i guess being on the microsoft treadmill like good little hamsters just wasn't good enough for some people ;-)
  240. LMAO[ Go to top ]

    But I don't have to be good debater San Juan, because the whole Sun/J2EE community is like a house of card. Java survives at all only because of the effort from guys like Rod Johnson, Juergen Hoeller, Howard Lewis, Richard 0berg, Gavin King, etc.

    Man you can't be serious. That like saying, "I use .NET and C# because I'm in love with Bill Gates." I definitely don't use Java because of any of the people you mentioned. I would none of my friends use java because of those guys. If that were true, one could just as well say "Rolf loves .NET because he wants to be Bill Gates." Which I'm sure is not true. I like to think you use and believe in C# and .NET because you find it fits your needs.
    Why don't you jump the ship? Read Cedric Beust summing up the year 2004:"Never mind the fact that the past eight years have all been predicted as the "year of Linux", there are quite a few signs that make me think that if anything, this year will be the year of Windows"http://www.beust.com/weblog/archives/000070.htmlRegardsRolf Tollerud

    Why must one jump ship? That sounds a bit mental and crazy to just "jump ship." That would be like saying, "I bought this great new shirt, so I should throw out all my old shirts." If you were the CEO of a large multi-national corporation, and the business has 500 billion dollars invested in technology X. Along comes a new CTO who says to you, "We need to throw out all of our existing software and hardware for this great new thing. It's like totally cool!"

    Would you A) fire the guy, B) wait until the guy calms down and comes to his sense, or C) throw away a half trillion dollars of investment and spend another half trillion to replace it? Personally, I'd choose A if the guy was serious.
  241. And just for the record what is important is only the enterprise market, real applications hosted on servers set up and owned by the company. The low-level market with simple html pages hosted on shared servers is uninteresting.

    It may be uninteresting, but it's vital for businesses, with matters of reliability and stability. That is why Linux+Apache is so popular here.
    That the low-level market is included in all surveys and statistics skew the results heavenly in Linux favor. RegardsRolf Tollerud

    And also in favour of Windows! As your reference said, the enterprise market is currently (and likely to remain) dominated by neither Linux, or Windows but Unix.

    By the way - there is something else that skews the 'market'. Linux can be free. There is a huge volume of Linux installations that don't could as 'sales of Linux', and a significant number of those are installed in parallel with or over Windows. Just looking at the 'market' for Linux is inadequate.
  242. Server shares[ Go to top ]

    Hey, Rolf, you DO know of course that since Linux is open source that most installations are probably not actual sales but downloads right? Therefore, all these data actually significantly underrepresent actual linux usage (which represents sales lost to windows - unix too, but i consider Linux to be a unix variant)

    Plus, you're looking at one market (east asia) and forget to mention that the rest of the world counts too (see quote below) ;-)

    Face it, Microsoft is running scared of Linux (as is Sun to be honest about it, but that's another story), and all the boasting and puffing and huffing you do ain't gonna change that.

    There IS a reason why microsoft's stock is stuck you know...

    Here are SALES figures worldwide for Linux, but note that these represent only sales, not actual usage.

    --------------------

    End-user adoption of Linux servers for enterprise workloads, hosting ISV applications and databases, is on the rise, fueling a transition in the marketplace. The annual growth rate of Linux server unit shipments has been increasing over the past three years from 15% in 2Q01 to 40% in Q2 2004. Given this rapid adoption, the mix of form-factors for these Linux servers has been changing over time, with the result that dual-processor systems are the predominant form-factor, followed by uniprocessors and four-processor systems. A new IDC Insight study provides data about this trend, and shows the market dynamics, by form-factor segment, within the Linux server market space.

    Linux represents about half of the worldwide server blade market, in terms of unit shipments, compared to 20% of all rack-optimized servers and compared to 11% of all pedestal (standalone) servers. Between 1998 and 2003, the 1-way (uniprocessor) segment of the Linux server market grew at a healthy pace, with a 37.5% compound annual growth rate (CAGR). Growing at a CAGR of 89.8% from 1998 to 2003, dual-processor systems account for fully 74% of total Linux server shipment volume in the market today. Worldwide Linux server customer revenue is expected to reach $9.1 bln in 2008, driven by a CAGR of 22.8%, compared to the broad market CAGR of 3.8% for the worldwide server market. IDC forecasts that Linux server shipments will be 25.7% of worldwide server shipments in 2008, up from 15.6% of worldwide server shipments in 2003.
  243. everything is relative[ Go to top ]

    But of course the best reason to run Windows is security:Linux and open source software accounted for 16 out of 29 advisories. Traditional UNIX software - 16. Windows - 7. http://www.unixsucks.com/unixsucks.aspxRegardsRolf Tollerud
    Have you ever thought about taking your act on the road?
    People have already written a book about this: it's called Don Quixote de la Mancha ;)
  244. everything is relative[ Go to top ]

    Rolf,
    But of course the best reason to run Windows is security:

    You meant this link, right?

    http://www.theinquirer.net/?article=20817

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  245. need to hire PR[ Go to top ]

    Cameron,

    I really think that the "Elder Statesman Role" is more you.

    Best respect
    Rolf Tollerud
  246. This is a valid trend[ Go to top ]

    .They quote reasons including better productivity / development cost and simpler integration using one (MS) architecture. It’s depressing.

    Only because they are clueless about development of software.

    With the right selection of Java tools, Java can be as productive (what they define as productive).

    Over the life of the application, the MS solution will cost more unless it is very simple and the requirements never change and if the only Java products used are big $$$ ones. I am doing both .Net and Java. And every day I think or say to someone I am working with "If I was using Eclipse/Java, I could done it faster/better". I've been using MS products much longer than Java so I am familiar with them.


    MS has lots of work to do getting VS.Net to place where OO development is easy. It also has lots to do get get all of its products on .Net.
  247. .net depressing?[ Go to top ]

    Why do say depressing? Surely our job as people that deliver applications is to deliver the best application.

    It is a fact that Java on the desktop means java on a windows desktop in the financial world. It is a fact that java doesn't work as well as .net on that desktop.

    While i have a natural preference for non-MS product, being historically unix orientated, its blindingly obvious that to deliver the best apps on the desktop, .net is the 1st choice.
    Of course, there are other factors (do you have any .net talent in your organisation?)

    Java server side (unix), .net client side (windows). Use & embrace the strengths of both.
  248. .net depressing?[ Go to top ]

    its blindingly obvious that to deliver the best apps on the desktop, .net is the 1st choice.

    I disagree. With a GUI toolkit like SWT, there is very little that can't be coded in a GUI on Windows in Java that can be done in .Net. Sure, it's likely to be faster to code the app using .Net, and some aspects of integration with the OS may be easier with .Net (although the new Java Native Desktop interface tools will help with this considerably), and the SWT app may not look quite as pretty! However, surely it makes more sense to have a desktop application that does not limit your choice (the consequence of using most Microsoft technologies). Using .Net may look good at first, but then someone turns up with a MacOS/X notebook and needs to use your app...
  249. .net depressing?[ Go to top ]

    With a GUI toolkit like SWT, there is very little that can't be coded in a GUI on Windows in Java that can be done in .Net.
    Until Avalon/XAML is released.
    Using .Net may look good at first, but then someone turns up with a MacOS/X notebook and needs to use your app...
    I am pretty sure that MS will have their .NET engine ready at the time when Avalon/XAML will be publicly available. There is a reason that MSIE is not updated for so long, they surely up to something. They will do what Java could not do: they will grab the desktops. Linux fans always say that Linux is the core, not the Gnome/KDE/whatever. So it will be: .NET apps will work on top of Linux core. It will be a blitzkrieg, and a successful one. I totally back up Rolf here.
  250. XAML for Linux/Mono/C#[ Go to top ]

    Xamlon (http://www.xamlon.com/) is planning to make a version for Mono/C#.

    And more, XAML to SWF converter

    Xamlon has also announced that they have preview versions of XAML for Flash and CF. Xamlon staffer Robin Debreuil has the news. Robin has stated that the product will allow for Flash applications to be coded in XAML and any language you choose as long as it compiles to the CLR (a la C#, VB.NET etc etc).

    It will be interesting to see what this means for Flex and Laszlo.

    Regards
    Rolf Tollerud
  251. .net depressing?[ Go to top ]

    Why do say depressing? Surely our job as people that deliver applications is to deliver the best application.It is a fact that Java on the desktop means java on a windows desktop in the financial world.
    Thank goodness. Then at least some of the desktop apps will be secure.
     It is a fact that java doesn't work as well as .net on that desktop.
    Is that a fact? I don't know of anyone doing Winforms(I'm sure some is doing it). It is all web stuff. I work with some ".Net is the cat's meow" people. And guess what they use to work with databases - SQuirreL. :)

    Eclipse RCP is doing great. Java is just as good on the desktop as .Net. And with Java, you don't alienate alternative or up-and-coming desktop OS's.

    Now if you mean Java doesn't work as well with MS Office as .Net, then I can see that.
    While i have a natural preference for non-MS product, being historically unix orientated, its blindingly obvious that to deliver the best apps on the desktop, .net is the 1st choice.
    Can't see why it is obvious. Having done enough of both MS desktop apps and Java desktop apps - I would choose Java 90 some percent of the time. The reason why is not obvious to those who don't know about the riches of Java.
    Of course, there are other factors (do you have any .net talent in your organisation?)
    If you have Java talent, you have .Net talent.
    Java server side (unix), .net client side (windows). Use &amp; embrace the strengths of both.

    Other than stating it is as truth, you've not shown proof of .Net being better on the desktop. Why use to different platforms when you can use one and not be duplicating code. I've developed apps that, depending on the user, have the some code running on the client and/or server.
  252. And the big hack known commonly as Windows is losing to Linux.
  253. I suppose that from technical point of view J2EE and .NET are very similar.
    So business decisions are not probably based on this aspect.


    But:

    - J2EE is not OS specific

    - J2EE is much more opened than .NET

    - J2EE is supported by SUN, IBM, Oracle, BEA, other 1000 of vendors and huge open source community

    - 2 biggest relational DB vendors (Oracle and IBM) are also on J2EE side


    IMO these are reasons, why J2EE deployment will be always 2-6 times bigger than .NET

    DF
  254. the only thing this tells me[ Go to top ]

    Is that programmers should divorce themselvs from a particular platform and language. Everything changes and nothing is forever.
  255. Programmers forever[ Go to top ]

    Yes, but some things stay longer relative to others...since you're not immortal, i can assume it won't be possible for you to learn everything new under the sun, so having a platform that has a core sameness over a long period of time would be helpful...

    This tells me is that programmers shouldn't stay programmers-only forever...what, you wanna be coding like the young 'uns when you're 40 or 50?
  256. Programmers forever[ Go to top ]

    .This tells me is that programmers shouldn't stay programmers-only forever...what, you wanna be coding like the young 'uns when you're 40 or 50?

    No, much better that the young 'uns! What is wrong with being 40 (or 50) and coding? I'm 40. Doing it much better than most young and old. :)
    so having a platform that has a core sameness over a long period of time would be helpful
    From what I have seen - it makes one stagnant.
  257. Programmers forever[ Go to top ]

    What is wrong with being 40 (or 50) and coding? I'm 40. Doing it much better than most young and old. :)

    Heh. Me (44) too. I believe the decades of experience are invaluable, and I am usually prove to be far better at design and debugging than younger colleagues.
  258. Programmers forever[ Go to top ]

    I want to code til I die. I found that being in management was VERY bad for the blood pressure.
  259. I'll second that[ Go to top ]

    I want to code til I die. I found that being in management was VERY bad for the blood pressure.

    it's funny when I hear people say, "do you want to code the rest of your life?" Happily, my answer is yes, though not necessarily at the same job or as a fulltime occupation.
  260. Stagnant[ Go to top ]

    Noticed that I said, the "CORE" remains the same...If you look at Java now, there is no way it could be called even close to stagnant...not when it is expanding and reaching into so many niches.

    With regards to the age thing, heheh...sorry!
  261. .Not a total failure[ Go to top ]

    The hype of .Not is catching up people are getting fed up of it and ditch as fast as they can. Most of VB/.Not developers are now switching to long trusted Java.
  262. Is "Rolf" a very small shell script?[ Go to top ]

    I was thinking it would be quite amusing to write a little program that was a Rolf emulator. Based on Weizenbaum's Eliza program all this program would have to do is to "scan" posts looking for certain key words (Sun, Java, J2EE, Solaris, Linux etc) and constructing largely random sentences ("But what you need to understand is that (Java keyword) is crap and that (MS technology) will squash it like a bug. Therefore you should rush off an learn (insert MS technology) now before it too late..") Add an occasional psedou intellectual quote from a Greek or Roman Philosopher ("After all, as Socrates has said 'The only true wisdom is in knowing you know nothing.'). And – in all of about 150 lines of sheel script, Java or another programming language of your choice – you could have something that, whilst it understood nothing of what it read and responded to, would be quite a convincing impression of Rolf.

    And then I realised that someone must already have done this. Rolf is, in fact, a very small shell script.

    A quick scan on Google confirmed it. Surely no-one would generate that amount of BS under their own name? I mean, if you worked for a living, or went to school or whatever, you wouldn’t have time to write that amount of stuff.
  263. Rolf is, in fact, a very small shell script.

    LMAO ! Fucking brilliant ! Best joke of the year ! Thank you, thank you, it lit up my morning !
  264. last year i completed a project for a consultancy using j2ee, they blew approx £1 million on a .net for a ERP/CRM type solution previously that almost brought the company down - during the .net web services hype etc.

    Ok that is my 2 pence now, i going to better website for more informations.

    thanking the you
  265. http://www.cooltechzone.com/index.php?option=content&task=view&id=887&Itemid=0

    "eBay announced its decision to abandon the use of Microsoft Passport and .NET services later this month to allow hassle-free and other variants of secure login authentication systems for its users. In October of 2004, world’s top most Monster.com also discontinued the use of Microsoft’s aforementioned secure transaction software to return to other alternatives. Since the launch, Microsoft has been involved in controversial debates over the web-applications’ security. In 2002, it was forced to settle charges of privacy and security of its users by the Federal Trade Commission (United States). Later in 2003, security experts found a flaw in the service, which could’ve compromised highly sensitive user information, such as credit card numbers and addresses, if tampered with by an intruder."
  266. Folks, folks...[ Go to top ]

    ...don't you realize that Rolf's *job* probably is to try to steer java developers away from java to .net? I mean - if you were Ballmer, had N billions at hand, and the goal to draw as much attention as possible away from the main competing platform - wouldn't you set up a staff of evangelists with the single purpose of infiltrating the "enemy" lines in public forums and such? And ask yourselves - which site is "the" J2EE site..? Sinking the morale in this site would probably be like, well, sinking the Bismarck during WWII or the Merrimac during the american civil war. Succeed with that, and consider 80% of the job done. And imagine a whole world bying VS - WinServer2003 - SQLServer licences...that's yummy for little Ballmers!

    I have followed Rolf's postings through the last year or so, and let me say, this man has excellent skills in rethorics: he never answers the questions where his proofs are weak (instead he uses the same technique as skilled polititians, namely, he steers the attention away from the question at another subject), etc.

    This man is probably a trained professional working for MS (think of the reduced marketing costs if he's successful!) - either that, or he's just plain nuts. Why would you otherwise preach .net so fanatically?

    So tell us Rolf, what's your employers name? Website/and or contact info, please. I repeat: what's your employers name? And DO answer this very question, and not a question not asked.
  267. Folks, folks...[ Go to top ]

    ...don't you realize that Rolf's *job* probably is to try to steer java developers away from java to .net?

    I totally disagree.

    As far as I can tell, he's just out to destroy the good name of Microsoft by associating himself with it.

    Rolf is probably one of the best things that's happened to J2EE marketing in a long time. Anyone that actually reads his arguments "for" returning to a totalitarian software market will immediately switch to J2EE and never look back.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  268. Folks, folks...[ Go to top ]

    ...don't you realize that Rolf's *job* probably is to try to steer java developers away from java to .net?
    I totally disagree.As far as I can tell, he's just out to destroy the good name of Microsoft by associating himself with it.Rolf is probably one of the best things that's happened to J2EE marketing in a long time. Anyone that actually reads his arguments "for" returning to a totalitarian software market will immediately switch to J2EE and never look back.
    Cameron, does that mean that his postings do not hold water at all? Come on, XFORMS is still in draft, HTTP was not updated for six years. Don't you expect the storm?
  269. Folks, folks...[ Go to top ]

    Cameron, does that mean that his postings do not hold water at all? Come on, XFORMS is still in draft

    nope, they do not hold any water, especially not in the way he delivers them. he is a j2ee/linux doom-sayer since several years, and until know he was wrong everytime
    HTTP was not updated for six years. Don't you expect the storm?

    well what should microsoft do? go back to the early nineties and attempt to push their own proprietary web protocol?? This simply will not work. The added value this might bring (nice development tools, etc.) will in NO WAY account for the risks of such a technology (total vendor lock in), hence such a technology would not have a chance on the market.

    Microsoft knows that, and i think their primary goal and intention is not to take over the (IT-) world (you used terms like "storm", "blitzkrieg"...), but to defend their most important product, which is Windows. They are creative in that, but it's a defensive strategy, .. so, as long as the Open source and J2EE stacks continue to evolve, i don't see a possibility for MS to destroy them.

    therefore, yours and Rolfs postings fit into one category: FUD!
  270. I think cameron was joking[ Go to top ]

    I believe cameron was attempting to be funny, which is generally more successful than my attempts. I don't know if Rolf is trying to be funny, but I read it like it's meant to be a series of punchlines without the joke.

    peter
  271. obviously I am unfairly persecuted[ Go to top ]

    "Rolf is probably one of the best things that's happened to J2EE marketing in a long time."

    You have said that before Cameron, but when do I get it in writing? Wherever I go people accuse me (unfairly!) of destroying contracts for millions and billions. A written affidavit from such a prominent member of the Big Iron Mafia would be extremely useful.

    With respect
    Rolf Tollerud
  272. obviously I am unfairly persecuted[ Go to top ]

    A written affidavit from such a prominent member of the Big Iron Mafia would be extremely useful.

    Big Iron Mafia? You've got to be kidding .. I'm a Microsoft MVP! ;-)

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  273. obviously I am unfairly persecuted[ Go to top ]

    If Rolf is here who is looking after HELL
  274. obviously I am unfairly persecuted[ Go to top ]

    A written affidavit from such a prominent member of the Big Iron Mafia would be extremely useful.
    Big Iron Mafia? You've got to be kidding .. I'm a Microsoft MVP! ;-)Peace,Cameron PurdyTangosol, Inc.Coherence: Shared Memories for J2EE Clusters

    I can't see how the big J2EE dogs can be happy with Cameron. He costs them (takes away) mucho consulting $s.
  275. Told you so...[ Go to top ]

    See? Rolf never answered my question (check replies in the thread).

    Rolf, you are hilarious *rofl* (pun intended)

    I repeat: what is the name of your employer? (name of the company) :)
  276. Rolf, do you have a real job?[ Go to top ]

    Rolf,

    Judging by the number and frequency of your postings, I am starting to believe that you don't have a real job!
    With job comes experience maybe then you can make intelligent arguments!?

    Unless you have done any real programming in .NET and J2EE, which doesn't seem to be the case judging from some of your postings, shut up!?

    If I were psychologist, I would say you are a lonely man with a desperate need for attention?