Discussions

News: .Net not converting Java developers

  1. .Net not converting Java developers (158 messages)

    An article on SDTimes writes that .NET has not proven itself compelling enough to convert significant numbers of Java or Linux developers. A research report done by Gartner in September among midsized businesses in North America showed that only 34 percent claimed they were looking forward to using .NET. 41 percent said they either don't understand .NET or believe it to be too complex.

    Read Delivering on a promise.

    Threaded Messages (158)

  2. .Net not converting Java developers[ Go to top ]

    Developers these days will convert to anything which may provide an employment, or even some hope of finding one.
  3. .Net not converting Java developers[ Go to top ]

    The IDE is not available free, darn it. Java developers are used to free development tools. Unless Microsoft makes VisualStudio .Net along with runtime libraries free, it will not flourish...
  4. Im not sure I understand this article, is it talking about developers or Companies, title and summary conflict. Anyone know which it is?
      I do find it funny though that the guy they interview says "the learning curve was steep". Imagine that, he goes from Visual Basic(aka High Shcool Algebra) to an Enterprise Platform(aka Plasma Physics) and he finds it hard. It can't help MS that they are trying to sell .NET to a group of "programmers" who have never written anything beyond form.visible=true.
  5. form.visible=true

    You seem to know alot about VB Mr Rasmussen, just what side of the fence are you on? :-)
  6. .Net not converting Java developers[ Go to top ]

    using (form) { visible=true }
  7. .Net not converting Java developers[ Go to top ]

    I'd learn it if a) I could get the environments for free, b) I had the time and c) I thought my company would let me try it. But as far as I can tell, no .NET app will run here any time soon (we are a research group, and few of those are researching with .NET).

    -Jason McKerr
    Northwest Alliance for Computational Science and Engineering
  8. .Net not converting Java developers[ Go to top ]

    Yeah - traitor :-)
  9. .Net not converting Java developers[ Go to top ]

    You can get a free IDE SharpDevelop to go with the free .net sdk from http://www.icsharpcode.net/
  10. .Net not converting Java developers[ Go to top ]

    Lyndon: "You can get a free IDE SharpDevelop to go with the free .net sdk from http://www.icsharpcode.net/"

    That's free as in beer.

    Just to be clear: Neither Java nor .NET are free as in speech.

    However, Sun has clearly worked to make Java ubiquitous on platforms that compete against Sun. Microsoft has clearly threatened to prevent .NET from doing the same (they're holding the "patent card"). Microsoft is an important enough part of the industry that I'd much rather see them work _with_ Sun, IBM, et al ... I think the most disappointing thing is that they went off in their own proprietary direction again when there was really no need to do so.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  11. .Net not converting Java developers[ Go to top ]

    Cameron says "That's free as in beer."

    Well as both java class files are .net exes are really just compressed source files you can generally peruse the source to all the supplied libraries.

    You can also distrute commercially code which uses these libraries.

    What more do you want?
  12. .Net not converting Java developers[ Go to top ]

    I would like to convert to QBasic.

    :-)

    It's not really OOP, but it has a very small amount of commands. easy to remember, you know :-)

    Not like those Java or .Net libraries.

    Maris
  13. .Net not converting Java developers[ Go to top ]

    I am a Java developer and I would like to get a hands-on experience in developing .Net applications..

    Can anyone compile links to:

    A. Good Technical Articles to improve understanding about the .Net architectures
    B. Free downloadable software components of .Net
    C. Tutorials explaining concepts
    D. Sample applications developed in .Net
    E. Step by step how to develop a small application like Web Services
    F. FREE Development tool
    G. What all the latest standards/ versions in this area.
    H. How a Java person should approach .Net

    Thanks,
    Parag
  14. .Net not converting Java developers[ Go to top ]

    A, C, D, E, G - Visit msdn.microsoft.com/default.asp and http://www.gotdotnet.com/ and www.devx.com

    B - Good luck. There are a few at Sourceforge. Be prepared for MS to make it illegal.

    F - See the above replies.
    H - With caution, you back against the wall, eyes wide open, use C# and realize any useful code you write with MS.Net will probably not work without major mods on any platform but Windows.
  15. .Net not converting Java developers[ Go to top ]

    A few pointers to start with:

    A. Good Technical Articles to improve understanding about the .Net architectures
        http://msdn.microsoft.com/net/guide/
        http://www.microsoft.com/seminar/shared/asp/view.asp?url=/seminar/en/20010710devt1-21/manifest.xml

    B. Free downloadable software components of .Net
        http://www.icsharpcode.net/OpenSource/SharpCvsLib/default.asp
        http://www.icsharpcode.net/OpenSource/SharpZipLib/default.asp
        http://nunit.org/
        http://nant.sourceforge.net/
        http://www.gotdotnet.com/userarea/default.aspx

    C. Tutorials explaining concepts
        http://samples.gotdotnet.com/quickstart/

    D. Sample applications developed in .Net
        http://www.ibuyspy.com

    E. Step by step how to develop a small application like Web Services
        http://msdn.microsoft.com/webservices/understanding/readme/default.aspx?pull=/library/en-us/dv_vstechart/html/vbtchGettingStartedWithXMLWebServicesInVisualStudioNET.asp

    F. FREE Development tool
        http://www.asp.net/webmatrix/
        http://www.icsharpcode.net/OpenSource/SD/default.asp

    G. What all the latest standards/ versions in this area.
        http://msdn.microsoft.com/net/ecma/
        http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dndotnet/html/deicazainterview.asp

    H. How a Java person should approach .Net
    Most of what you know will work just right: OO concepts, low level patterns, even the syntax is pretty much the same. The class libraries cover pretty much the same as the Java API, but the namespace distribution is different. If you take the whole thing in a possitive light you will be running in no time.
  16. .Net not converting Java developers[ Go to top ]

    I think a recent Microsoft advertisement said it best when they describe putting your services on one big fat unclustered Windows box (as opposed to many little clustered machines) as "scaling up."

    It just made me laugh. As long as Microsoft tries to lie, no one will believe them, even if they get around to making a decent product.
  17. .Net not converting Java developers[ Go to top ]

    I feel that only java developers can ever become serious .net developers if they want to. Reasons: Microsoft screwed up their developer community by mixing all bad to worst practices, their tools encouraged adhoc programming. Now when MS comes up with a cleaned up version, only the guys who worked with clean environments can appreciate and use such tools. An average MS veteran, can at the max do a few "this.that = bug" kind of apps and give head ache to the users. And offcourse, you have those service packs and wizards to improve your bugs! All the best Java to .Net jumpers!
  18. .Net not converting Java developers[ Go to top ]

    And again...Loads of posts saying that vb coders are not as clever or enterprise as java coder. HHHHHahhahhahhaaaaaa

    Still, the weekend now. Hurrah!
  19. .Net not converting Java developers[ Go to top ]

    java developers do not have to bother - http://www.forbes.com/2002/12/05/cz_qh_1205hp_print.html

    The New HP Way: World's Cheapest Consultants
    Quentin Hardy, 12.05.02, 11:26 AM ET

    Tech giant Hewlett-Packard has seen the future of technology consulting. It's on the other side of the globe and it's really, really cheap.

    "We're trying to move everything we can offshore," HP Services chief Ann Livermore told Wall Street analysts at a meeting Wednesday. "We're aggressively realigning our resources." Short term, that means adding to the software and services personnel HP (nyse: HPQ - news - people ) already has in India. Further out, HP expects China to also turn into a major consulting center.

    The plan addresses a trend toward lower-priced consulting that's been hurting HP. The company's $3.1 billion fourth-quarter services revenue (total fourth-quarter revenue was just over $18 billion) was off 3% from a year before. Consulting and integration revenue was the weakest part of services, down 17% on the year. Livermore, with the blessing of Chief Executive Carly Fiorina, is betting that HP can both lower its costs and damage industry leader IBM (nyse: IBM - news - people ) by slashing services prices with cheaper bodies.

    As if the high-priced, oversupplied consulting industry didn't have enough to worry about. The field has suffered during the past two years' downturn in technology spending. How bad has it gotten? PricewaterhouseCoopers, which almost sold its PwC consulting business to HP two years ago for $18 billion, managed to finally unload its concern to IBM this year for just $3.5 billion. IBM added the 30,000 consultants to its Global Services Business in an effort to smother consultancies like EDS (nyse: EDS - news - people ) and Accenture (nyse: ACU - news - people ). Competition remains fierce, too: As Livermore pointed out, a laid-off consultant isn't like a factory that gets mothballed during a recession--he's still out there looking for business.

    "The oversupply doesn't go away," she said, "consulting and integration is going through a tremendous transition, with constant price pressure."

    According to Jurgen Rottler, vice president of marketing, strategy and alliances at HP Services, HP will grow in India, building on the "several thousand" services people the company already employs there. "In an ideal world," he said, "you'd migrate as much as you possibly could to India."

    Many of HP's Web applications for Microsoft's (nasdaq: MSFT - news - people ) .Net initiative will be written in India, Rottler predicted. HP and Microsoft recently announced HP would be a prime global integrator for .Net, a "Web services" move to put more interactive software on the Internet.

    HP figures a good high-end programmer in India costs about $20,000 a year, about a quarter the U.S. cost. And things could get even cheaper. "We see China gaining on India about three or four years from now," said Rottler. HP is also developing staff there.

    HP stands to have plenty of company developing the Indian services industry, however. Already, local firms such as the Tata Group (which is traded on the Bombay Stock Exchange), Infosys (nasdaq: INFY - news - people ) and Satyam (nyse: SAY - news - people ) have boosted their software and consulting arms. Microsoft's recent $400 million investment in the country, building on a strong programming presence already there, is designed to boost education, business partnerships and software development.
  20. .Net not converting Java developers[ Go to top ]

    I'm not sure what you mean. I always thought "scaling up",
    met uses bigger boxes, while "scaling out" met using
    clustering. How is Microsoft's wording wrong?

    Also, MSN is the second largest website out there, next
    to AOL, and it runs on Microsoft platforms, so obviously
    it can scale; both "up", and "out".
  21. scalability[ Go to top ]

    Terry: "Also, MSN is the second largest website out there"

    By what measurement? HTTP requests? Every time I mis-type a URL in IE, it goes to MSN three times (judging by the number of "clicks" I hear and the URLs in the "Address" bar)! Every time I install or upgrade something from Microsoft, MSN automatically becomes my "new" home page, thank you very much. If MSN is still only #2 then it must be really bad.

    Anyway, that's all off topic. Yes, you can scale out easily with Windows web servers. It's more expensive (mainly OS and operating costs), but there is very little technically difficult about it. MSN doesn't have the OS costs, and it would look bad to run Linux there, and it's a great test case for Microsoft, so of course it runs Windows/IIS ;-)

    Scalability doesn't usually get difficult until you get into the data processing tiers, such as transactional servers and database servers. That's where MSN had all of its early problems. (I never did get billed if I remember correctly.)

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  22. Your punch line is: If you take the whole thing in a possitive light you will be running in no time.

    How many j2ee developers take .net "managed" C++ and/or ATL server based "unmanaged" C++ development seriously? How many j2ee developers know even that ATL dev is not easy and one needs to know C++/ATL seriously enough? Do they even know what ATL is, for that matter?

    After facing enormous uncertainties revolving around the JAVA performance related to extracting enterprise level ERP data from SAP r/3 HOST both in terms of sheer volume and parallel scheduling, we have decided to switch to SAP .net and/or C++/ATL based data extractors based on SAP's RFC SDK.

    This is because JAVA 2 platform does not let us run concurrent (asynchronous as well as synchronous) RFC calls from SAP host to the JVM. There has been a serious performance issue that made the deployment costly to maintain and support actually. We are looking into SAP .net connector as prototypical alternative to C++/ATL/RFCSDK since time to develop is at stake.

    So dear java folks, please shut up when it comes to hugh buffered I/o both on sockets and on the server disks!
    C++ still prevails there.
  23. This is because JAVA 2 platform does not let us run concurrent (asynchronous as well as synchronous) RFC calls from SAP host to the JVM.


    I seriously doubt this. The JVM gives you all the basic tools for
    heavy data processing. The rest is architectural patterns.

    You're falling into the same trap as the Java advocates you describe. Misinformation and preconceptions seems to be the primary source of discussion on both sides.

    I've always wondered how people can both accuse MS of stealing and copying concepts of J2EE and at the same time deny that this provides for a scalable platform. .NET is/will be a serious competitor to J2EE - the latter has an advantage in maturity, and openness, the former was able to learn a few lessons that went wrong in the J2EE world.
  24. This is because JAVA 2 platform does not let us run concurrent (asynchronous as well as synchronous) RFC calls from SAP host to the JVM.


    >I seriously doubt this. The JVM gives you all the basic tools for heavy data processing. The rest is architectural patterns.

    We have used all the JVMs avaiable in JDK131 (client, server and classic), together with jdb debugger with every time error showing up as a "memory leak" between RFC calls, as none of the JVMs debug-tested could garbage collect the JNI objects used by the RFC library.
    We found out that the "induced" garbage collection works with all the 3 JVMs shipped with BEA WLS 6.1 SP3. Here, induction meant manually doing a "null" on the object that represents the RFC call. We still have not been able to solve the parallel RFC call problem.

    > You're falling into the same trap ...

    While not inviting any furthur rebuttal on this, I suggest you to understand the RFC/JCO basics before you "ignore" lessens learnt by an experienced J2EE/C++/ATL developer such as myself.

    We need to be positive to know where JAVA works (portability across platforms) and at what expense (performance, when compared to native platform specific binaries (such as the VC++/ATL executable with static linkage to RFC library of SAP R/3).

    No more rebuttal please on this subject!
  25. .Net not converting Java developers[ Go to top ]

    Don't I just love these discussions..

    The "war" between MS and Java is a truly productive one. Here's how the evolution has gone so far (more or less):

    round 1:
      MS comes up with com+ and mts (a great platform)
      Sun responds with the J2EE spec (a great spec)

    round 2:
      MS responds with the .NET platform and CLR
      Sun counterattacks with J2EE 1.4 (web services)

    round 3:
      What next?

    It'll be interesting to see.. sharp..

    (Personally I prefer j2ee..)

    //ras
  26. Parag,
    There is a Patterns and Practices group at Microsoft that only develops documents and applications for customer use. The electronic copies are free of charge. I know there are in the process of building another .Net suite of apps. Look in http://msdn.microsoft.com/practices/subject/default.asp and let me know if it helps
  27. .Net not converting Java developers[ Go to top ]

    "Developers these days will convert to anything which may provide an employment, or even some hope of finding one." - you


    Exactly my sentiment, there has been hordes of Microsoft developers attempting to flood J2EE career opportunities including Microsoft gold consulting firms. They
    xonsistently get their ass handed to them on phone tech interviews.

    The question should be how many Microsoft developers are struggling to enter J2EE. Have a look at the the job sites and how many are on TheServerSide.com trolling.


    as Stan Lee would say: 'Nuff Said.

    Excelsior
  28. Reasonable[ Go to top ]

    Each has its pros and cons.
  29. .NET for Java developers[ Go to top ]

    There is a light-weight version of VisualStudio.NET that is free called the WebMatrix project.

    You can also build .NET apps from a command-line as well like Java. There's even a port of Apache's Ant to NAnt, an open-source Ant-like tool for .NET. There's also .NET version of JUnit, NUnit.

    If you're a Java developer and need to use .NET, there's a good book called C# for Java Developers that's pretty fair unbiased, considering it's from Microsoft Press. You can even read the positive book review on Slashdot.
  30. .NET for Java developers[ Go to top ]

    For me C# it`s an insult, a dirty copy of Java, why do i have to move to Net if Java does the same and even better these MS guys always are trying to appear as "The wet-water inventors", they copied every concept involt in Java frameWork
  31. .NET for Java developers[ Go to top ]

    ASP came around way before "JSP"

    ODBC came around way before "JDBC"

    EJB was a knockoff of MTS Services

    There are plenty more.

    Microsoft developers have moved on from ODBC to OleDb...so we dont care if you ride our coattails.

    I still dont get how Java is "about choice"? If you choose the Java platform, you are stuck with the java language. Once you start working with a J2EE container, you are pretty much stuck with it...(because you get bound to it)

    All those stupid containers work differently. There are tons of unemployed java developers out there...because "they dont know weblogic" or "they dont know websphere".

    Great....Vendor skills are a way of life...for java developers...as well as microsoft developers.
  32. .NET for Java developers[ Go to top ]

    If you look at the C# specification, you will see that C# is actually much more like C++ than Java.

    If you look at the spec, you will also learn that C# does a lot of things much better than Java.
  33. .NET for Java developers[ Go to top ]

    Thanks for the "official history from Redmond" bit. Sounds a bit Soviet to me.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  34. .NET for Java developers[ Go to top ]

    its annoying how java developers talk about java like sun "gifted it to the world" to liberate the developer community from microsoft.

    please. the current flavor of java/j2ee steals tons of ideas from microsoft and is owned and controlled by Sun Microsystems, Inc.

    Guess what? If sun had the ability to dominate/influence like Microsoft does, they would do the same things microsoft does.

    you want a truly open environment..nix java and go straight to ANSI C++!
  35. .NET for Java developers[ Go to top ]

    ODBC came around way before "JDBC"


    Yes, ODBC came out of MS, but I wonder if MS would have bothered standardizing database driver plug-ins if they had been able to get traction with their own SQL server.

    > EJB was a knockoff of MTS Services

    I doubt it. App server technologies have been around for a while - back when MS sold just MS-DOS and threw in BASIC. I know that BEA's Tuxedo product (which includes async as well as sync service invokes, messaging, and transactions) dates back twenty years. In fact, I'm fairly sure that MTS conforms to the XA transaction standards, which were developed originally by AT&T (for Tuxedo) and IBM. I'm also fairly sure that IIOP pre-dates MS services as well (could be wrong here).

    I work at BEA, but these opinions are my own.

    Tom, BEA
  36. .NET for Java developers[ Go to top ]

    The books that teaches the so called .NET are expensive. We normaly get Java Book free on PDF. Sorry I will not change to .NET
  37. Just now team leader of our MS developer group called and told they are not moving their application from VB to .NET because .NET is just more talks then serious development enviorement and they are not sure whether .NET will perform or even work on clients machine still using windows 95 and windows98.

    I would like to know how other MS developer think about .NET
  38. One of the big problems with .NET is their data access strategies and how they coupled it with a bunch of the other frameworks. Again, this makes for fast development, but is terrible when it comes to maintain the beast. JDO is a much better solution and allows you grow a natural domain model. Personally, I really Java's approach much better. Maybe I'm just an object biggot, but I won't switch to a language that doesn't advocate a free choice in the use of object-oriented principles. I also don't like how it assumes so much about the patterns and class libraries that are you going to use. It eliminates a lot of the choices that you have as an architect. Again, it makes for fast development, but I can't see very complicated apps being built using .NET like this. At least I'm not confortable with that. I think these reasons are important to a lot of other java programmers as well.
  39. I doubt if you have written one line of code in c#/.net. From your comments, it is quite obvious.
  40. what you are you talking about man. JDO is a clear copy of ADO.net. Have you have used ADO.net. Do some reading on the power of DataSets,DataColumns,DataRows,DataTable.
  41. Mackie: "what you are you talking about man. JDO is a clear copy of ADO.net. Have you have used ADO.net."

    Have you used JDO? ;-)

    JDO precedes ADO.NET and JDO is not comparable to ADO, despite the similarity of acronyms.

    If you said "JDBC is a copy of ODBC", you'd be correct, at least initially (JDBC 1.0). That was purposeful, because Sun wanted to be able to use ODBC (and UDBC) drivers from Java to kick-start the data access capability.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  42. Please Mackie,

    JDO is about O/R mapping and persistence. Have absolutely nothing to do with ADO.NET.

    Regards
    Rolf Tollerud
  43. .Net not converting Java developers[ Go to top ]

    And having used ADO.Net and tools like Hibernate I would rather have the later. I would love it if something like Hibernate would show up for MS.Net (cause I gotta use .Net sometimes and I have love having to avoid writing CRUD SQL).
  44. .Net not converting Java developers[ Go to top ]

    There are some experimental work called "Objectspaces", but it is to be replaced by "Business Component Framework" as part of Visual Studio. As I understand it BCF will be built around an IL code enhancer that adds persistence to regular serialisable objects, an "entity" framework as a rebuttal to J2EE entity beans.

    For what it’s worth.

    Regards
    Rolf Tollerud
  45. P Singh:

    "Just now team leader of our MS developer group called and told they are not moving their application from VB to .NET because .NET is just more talks then serious development enviorement "

    An interesting MS developer leader, .NET *is* a serious development platform, whether it's better than Java is open to debate but you can do a great deal with .NET (at least two banks in my country are migrating their core systems to .NET) and most studies show that in the mid-term the development pie will have very big Java and .NET segments (which one will be bigger is again open to debate).

    P Singh:
    "they are not sure whether .NET will perform or even work on clients machine still using windows 95 and windows98"

    It won't work in Win95. Windows Forms apps will work in Win98. I wonder why your team leaders didn't even bother to try...
  46. .Net not converting Java developers[ Go to top ]

    <Q>
    at least two banks in my country are migrating their core systems to .NET
    </q>

    And at least one isn't - it has moved to Java. (You are in Brazil?)

    I have three clients that are MS development shops - mostly set against using another platform (one uses Oracle for their database). None are interested in moving to .Net no matter how I or others push [suggest]. For a variety of reasons. Since they are set on MS products I would like at them at least to move to .Net because so much can be gained. I did get one other client to move to Java from VB. I would love them all to move to Java but some just won't see the light so I am settling for a least a glimmer.
  47. .Net not converting Java developers[ Go to top ]

    <Q>
    Just now team leader of our MS developer group called and told they are not moving their application from VB to .NET because .NET is just more talks then serious development enviorement and they are not sure whether .NET will perform or even work on clients machine still using windows 95 and windows98.

    I would like to know how other MS developer think about .NET
    </Q>

    I do "Windows" and non-Windows specific programming. I would say .Net is no less a serious programming environment than VB6 and ASP. Having used VB Classic for years I can say that if your group is set on MS products only - then they should move immediately to .Net. Especially for server side programming. VB Classic, and ASP for that matter, is not all it is cracked up to be. I am constantly having problems with server side issues. Server side debugging with logs - what fun. How about a stack trace even? .Net will perform just as well as the old stuff - if not better. If you have clients that are running 95 - you have bigger problems. Funny - clients running Win9x and being concerned with serious development. :)
  48. .Net on Win9x[ Go to top ]

    Just now team leader of our MS developer group called and told they are not moving their application from VB to .NET because .NET is just more talks then serious development enviorement

    > I would like to know how other MS developer think about .NET

    Since you asked ... .NET is a large beast and cannot be adequately described in a single catch phrase. .NET _does_ provide a simpler component-based framework for Windows desktop applications and for traditional client-server apps. ASP.NET is _much_ simpler to develop for than ASP. Finally, .net also encompasses useful technologies not previously addressed by the Windows api, such as regular expressions and xml parsing.

    How well .NET scales to enterprise apps is a topic with which I have no experience. And, of course, these are really not adequate reasons for a java shop to switch.

    > they are not sure whether .NET will perform or even work on clients machine still using windows 95 and windows98

    .NET can be installed on Win98 and NT 4 but not on Win95.

    It seems to me that your team leader needs to learn how to do his own research. I have no idea how he concluded that ".NET is more talk than serious development"; imho it's a far more serious development environment than (say) any version of VB prior to VB6.

    Just my 2 cents. I'm a longtime Delphi developer who has been using c# and .net for six months.

    Art
  49. I am now using Java and J2EE but for about a year (Oct 2001 to Oct 2002) I was using C# and .NET. .Net is much more "serious" then VB. It is a great development environment. There are many things I like about C# over Java. It does not work on 95 but will work on 98. In my opinion any one that still uses 95 in 2003 has no real business expressing an opinion on any technical issue because there is no reason to use 95.
  50. You mean people still write fat client applications? We have begun the conversion process to .Net and haven't really encountered the "steep learning curve" that so many talk about. I have done VB, Java, VB.Net, and C# and I didn't think it was that difficult. If you are writing enterprise applications, Java is way too expensive. I love that everyone talks about all the free app servers available, but when you get right down to it, spending the dollars to aquire Weblogic or Websphere to run your corporate applications is retarded. What is it, 15-20k per cpu? The other thing I like is if I have a problem, I call Microsoft because it is their products end to end. I don't have all the "best of breed" vendors pointing their finger at each other while nothing gets resolved like all the java projects I worked on.
  51. .Net not converting Java developers[ Go to top ]

    The body of the article is rather detached from the headline in a misleading sort of way... as though they figured they'd better put in the word "Java" for theserverside.com audience ;)

    It's really about the fact that MS shops will put up with the pain of moving to .NET cuz what else are they going to do? Would have been nice if they'd found a J2EE shop that said, "here's why we're going to move to .NET."

    But the bit that rang true and interesting for me is the familiar theme of folks implementing with a sub-set of what they are handed. J2EE (buying BEA as a glorified servlet runner) and now .NET suffer from the same symptom... but that's okay. Everybody needs a learning curve... especially IT shops in big companies that have to put up with politics and shifting winds...

    David
  52. .Net not converting Java developers[ Go to top ]

    Yeah I agree, The headline on this serverside posting is totaly inconsistent with the name of the article. The article name is delivering on a promise (A .NET Promise) but says very little about what the serverside headline is. I have to question the motive behind this for the serverside. They should take care to match posting to actual article. A windows Java developer can go into this article and start to sway to .NET. Not to say it is wrong or right, but the original reason for reading the article. (To see why Java developers aren't being swayed) is not what I get out of it. It is actually trying to convinve me to sway if I am deploying to Windows.
  53. .Net not converting Java developers[ Go to top ]

    I have not used .Net, but from what I understand, if a person wants to develop a Windows application in .Net the runtime file the programs needs to use to run on a system is unbelievable huge. Is this a problem?

    MS markets .Net to groups that are developing straight forward applications. Java and J2EE is made for large scale complex applications. The group that I work develops n-tier apps. We don't have any compelling reason to switch.
  54. .Net not converting Java developers[ Go to top ]

    <Q>
    </Q>
    Not any less than any other Windows program. If a Windows PC is not up to date, then there usually is alot to install anyway. If it is up to date then it will have the .Net framework
  55. .Net not converting Java developers[ Go to top ]

    Max,

    <Q>Java and J2EE is made for large scale complex applications</Q>

    At least you talk much about large scale complex applications.

    Regards
    Rolf Tollerud
  56. .Net not converting Java developers[ Go to top ]

    Need a compelling reason to switch to .net?:

    ASP.NET = Purely compiled machine code
    JSP = Interpretted code

    Build a small webapp in Java/JSP
    Build the same thing in ASP.NET


    Compare the response time. We have several java/jsp apps in our firm...so people were used to the doggy java.

    I recently built our firm's first ASP.NET application. People were blown away by how the app's pages just "pop". Blindingly fast.

    People who argue that Java is "almost as fast as C" are dreaming. Compiled Java is about as fast as interpretted ASP. Its a joke.

    By the way...try out VS.NET...and you will realize why you pay for it. "You get what you pay for." VS.NET is the best IDE on the market (java or not) The only Java IDE that performs is Eclipse (non-swing). Good luck trying to set up the debugger!....oh wait...Java developer dont need debuggers because they like to do everything the hard way.
  57. .Net not converting Java developers[ Go to top ]

    Tom Cole: "ASP.NET = Purely compiled machine code, JSP = Interpretted code"

    Regarding "interpretted code", feel free to come back and have a nice discussion when you at least get your technical ducks in a row.

    Regarding performance, JVMs are generally speaking about 10-15% faster than the Microsoft CLR, and in many cases are significantly better than that.

    In addition to the performance tests (posted here by a .NET would-be evangelist) that I easily showed Java to be 30% faster at, take a look at:

    http://javalobby.org/members/jpr/methodopt.jsp

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  58. Mr Cameron,
    Please refer to my feedback on this subject, put under the message #68050.

    C++/ATL still prevails when it comes to low-level stuff, such as making "parallel" asynchronous RFC call to a SAP R/3 host for huge enterprise data, that you wan't to do want all these JVMs cannot!

    No wonder enterprise software maker SAP recommends their newly released .NET connector over its own JAVA connector for anything of large scale!

    Thanks,
    /Ravi
  59. C++/ATL/COM+/OLE-DB combo really is blazingly fast. The trading platforms that we have at our day trading firm prove this point. From the server side, the market data server and execution server consolidate all the activity from all ECNs, market makers, Nasdaq, NYSE, AMEX etc. From the clients side the trading platforms run many open windows with a lot of algorithm & graphics intensive work in real-time. The amount of activity that happens on client machines is really amazing (I'm guessing that it is more than what happens on the servers). I have never seen the computer utilization going beyond 30% on any of the high volume days.

    I would suggest that if speed is your concern then stick with C++/ATL.
  60. Srikanth: "C++/ATL/COM+/OLE-DB combo really is blazingly fast. The trading platforms that we have at our day trading firm prove this point."

    It certainly can be fast ... there's nothing in particular about any of the above that force an application so constructed to be slow. I used most of the above extensively for years ... it's a PITA to develop with and a PITA to debug and a PITA to maintain, but it's workable as long as you're stuck on Windows.

    Srikanth: "From the server side, the market data server and execution server consolidate all the activity from all ECNs, market makers, Nasdaq, NYSE, AMEX etc. From the clients side the trading platforms run many open windows with a lot of algorithm & graphics intensive work in real-time. The amount of activity that happens on client machines is really amazing (I'm guessing that it is more than what happens on the servers)."

    We see those same apps done in Java, and it's plenty fast to accomplish the same. We work with a handful of trading companies (not "day traders", but professional traders) and their infrastructures and presentation is all done with Java now. Similarly banks.

    Of course, that's not to somehow degrade the "C++/ATL/COM+/OLE-DB combo" ... it's workable technology if you have to get it to work, I just don't happen to think it's the future. Neither do most CIOs/CTOs.

    For Windows-only front-ends, you'll see most new dev being done in .NET. For all other front-ends, it will be done in Java, HTML, etc. For back-ends, most of it is already Java.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  61. Cameron :
     " It certainly can be fast ... there's nothing in particular about any of the above that force an application so constructed to be slow. "

     An application can only be as fast as it's underlying infrastructure allows it to be. These are the resources occupied by a single threaded java "Do nothing" application ( I just had Thread.sleep(15000) in the main function so that java.exe hangs on long enough for me get this info ) :

      Peak Mem usage : 4MB
      Number of theads : 8

                  Priority Context Switches
                  -------- ----------------
      Thread 1 : Normal 203
      Thread 2 : Normal 9
      Thread 3 : Normal 5
      Thread 4 : High 5
      Thread 5 : Normal 511
      Thread 6 : Normal 3
      Thread 7 : Normal 2
      Thread 8 : Normal 14

      What kind of application performance can you expect when the underlying architechture behaves like this ?

     I/O performance, memory consumption and threads are the areas in which the underlying architechture can effect the performance of applications. I/O performance because most applications today are distributed, memory consumption because a large memory usage results in more page swapping and finally threads because thread count/context switches tax the OS.

     A similar ATL/C++ application costed me 1 thread, 500KB of memory.

     " I used most of the above extensively for years ... it's a PITA to develop with and a PITA to debug and a PITA to maintain, but it's workable as long as you're stuck on Windows. "

      We started our development in Jan ( from scratch ... not a single line of code ) and had our first release in July. 7 month development time doesn't seem to be like "slow development" to me. Sure I had couple of 14 hour work days but isn't that a normal thing in our community ?

     Visual Studio 6.0 for us has been a pleasant experience. The VS integrated debug is such a pleasure... stop on any statement, edit & continue, debug out dump, call stack, any thing you want and more. The PITA with COM/COM+ is handling HRESULTS. I mean it really is PITA. Another area i can think of is synchronization related code in some cases ( think COM apartments ). Regarding maintainability, COM programming model has never been a problem for me. The model allows a beautiful way for extending existing objects.

     " We see those same apps done in Java, and it's plenty fast to accomplish the same. We work with a handful of trading companies (not "day traders", but professional traders) and their infrastructures and presentation is all done with Java now. Similarly banks. "

      No way can a java app speed be compared to C++/COM app (please take a look my "Big decimal" post in the Petshop thread. It's 10 fold difference ) . The speed difference becomes more with the complexity of the app. It's only natural because of the middle JVM ( memory pages being swapped, page faults, thread count, context switches etc... ).

    Oh by the way, we deal with professional traders.. sorry for the wrong usage in my previous post. If you look at the industry standard trading platforms like Redi+, lava, Realtick, etc., ( users in the order of thousands ) they are all C++ and there is reason behind that.

     " Of course, that's not to somehow degrade the "C++/ATL/COM+/OLE-DB combo" ... it's workable technology if you have to get it to work, I just don't happen to think it's the future. Neither do most CIOs/CTOs. "

        You are sounding as if all the people who "made it to work" ( see above names ... ) have barely "made it to work" and it is just going to take one bird to P to make them crash. Get a grip and stop being such a meglomaniac. Before .NET there are literally thousands of IIS/ASP web sites.

      " For Windows-only front-ends, you'll see most new dev being done in .NET. For all other front-ends, it will be done in Java, HTML, etc. For back-ends, most of it is already Java. "

        Nope... it's going to take a lot of convincing to do by MSFT to make us move to .NET and so is the case for a lot of my friends I know. Nobody in their right mind would develop a GUI based front-end these days unless it is an absolute necessary. And I am sorry to break this news to you but java does not own HTML, DHTML, etc. And regarding servers, i would like to remind you about thousands of ASP installations though JSP definitely holds an edge here.
  62. Shrikanth: "No way can a java app speed be compared to C++/COM app (please take a look my "Big decimal" post in the Petshop thread. It's 10 fold difference ) . The speed difference becomes more with the complexity of the app. It's only natural because of the middle JVM ( memory pages being swapped, page faults, thread count, context switches etc... )."

    I found some more interesting "stuff" in addition to these run-time "machine" related issues. I just finished reading the last chapter "CLR EXternals" from COM/CLR guru DON BOX's book on CLR, with ATL guru Chris Sells as a co-author. They distinguish between CLR and "other" runtime systems such as JNI. DON BOX and (his) developmentor has generally been a pro-Microsoft shop, although with equal stress on Java, in recent years. I recommend you to read this chapter and understand what penulties most runtime systems have when integrating with (native) C runtime libraries thru' the so called "thunking layer". Unlike the JNI, CLR does not require a "mode switch" to transition from "runtime mode" to "C mode". This is done by allowing the use of classic C-style memory manipulation when absolutely essential, be it "managed" pointers that go thru' the code verifier or the C-style pointers that support pointer arith additionally.
    Good luck to you all in case you are like me (-a physicist's attitude to be an honest observer of both the camps) and keep in touch with both worlds that are not going to eliminate each other for long time to come.
    ...
    /Ravi
    P.S.:
    Interested folks can ask folks like me at tadwalkar at hotmail dot com, for details (-if you are impatient about reading Don Box's serious readings that require understanding of COM/DCOM/C++/ATL and now FCL/CLR).
  63. Ravi: "C++/ATL still prevails when it comes to low-level stuff, such as making "parallel" asynchronous RFC call to a SAP R/3 host for huge enterprise data, that you wan't to do want all these JVMs cannot!"

    It may be a legitimate gripe that SAP's Java adapter is not multi-thread safe or capable, but that has nothing to do with the JVM specification.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  64. Cameron: "It may be a legitimate gripe that SAP's Java adapter is not multi-thread safe or capable, but that has nothing to do with the JVM specification."

    Cameron, the underlying RFC library (LIBRFC.DLL version 6.2) used by all of the SAP connectors is multithreaded.

    Certain features in SAP connector technology exist today completely due to CLR's features, and that's where there other connectors fall behind, e.g. parallel asynchronous communication with host system.

    As for your comment on JVM specs, refer to what I had to say about JVMs from my earlier message #68050 post. JVMs today do NOT support parallel asynch calls with host, in my experience. Shrikanth Talla's message #68635 has a direct comparison in another context.

    Many of the new breed Java developers sound more like a C++ developer in the early 90s when the latter stated that all COBOL is dead and C++ is the only thing that will servive. Well, every programming language has a place to stay, be it COBOL or C++, jsp or asp/x, and java or C# for that matter.

    /Ravi
  65. Srikanth says "...JVMs today do NOT support parallel asynch calls with host, in my experience..."

    I fail to see how this is something to do with the JVM. Such calls would be JNI calls to a supplied native library.

    If the <i>actual</i> implementation does not allow parallel calls than that is an implementation issue, not a fundamental flaw in the JVM. I would presume that the native code implementor is SAP.

    "...please shut up when it comes to hugh buffered I/o both on sockets and on the server disks..."

    java now has a nio package which addresses alot of the io concerns people had with java.

    Regardless of whether java is lacking any functionality at the moment, the point is, if it's considered usefull someone will enventually add it. And you will still have the advantages of java being 'easy' compared to MFC/ATL/COM+/Win32API etc.

    BTW, Why are you moving such a volume of enterprise data through the SAP apis?, wouldn't you be better off exporting to a RDBMS or using some middleware like tibco RV?
  66. .Net not converting Java developers[ Go to top ]

    quoting error!

    s/Srinkanth/ravi/g

    sorry
  67. Your reply suggests that you have no idea what SAP RFC means, let alone what JVM debugger (jdb) does.

    TMC states it very emphatically- your reply suggests that you belong to the crop of 80% J2EE developers who think java will solve any and every problem on earth with basic understanding that we C++ programmers have gained on large projects, thru' "hardship". You belong to the folks who think jdk 1.4's java.nio will solve the problem, without any idea about what and how real world consulting works at large enterprises.
    I suggest that you need to have something concrete in terms of real world experience, to reason why we experienced JVM crashes if the JNI had memory leaks? We SAP/ABAP/J2EE/C++ developers consulted with RFC/C++/JNI experts before coming to conclusions of switching from JAVA to .NET. I recommend you to read up -once again- my earlier messages on this mini-thread, and come up with more constructive criticism, if you can! Otherwise I do not see any reason to read replies that show uplike misguided missiles!
  68. Ravi, your reply suggests you really are in the Rolf class of unpersonable people aren't you? It appears you are a die-hard C++ programmer with a chip on his shoulder about the popularity of java.

    Anyway, regarding your less than helpful reply.

    I have a basic understanding of SAP RFC, my only SAP experience is with the Tibco SAP R3 Adapter. I know enough about programming theory in principal to know that it is nothing special. It's not like SAP invented RPC or anything, it's all just sockets after all. Please explain to me how something that listens on a socket can't behave in an asyncronous manner, like for example, a web server.

    The problem you are having may be either a problem with the SAP JNI library, or the JVM JNI interface, but I'm sure you have investigated that possibility. Thats nothing to do with the JVM's ability to handle concurrent RFC calls from SAP->JVM. I also find it strange that when you forced GC by nulling a reference to an RFC object that you expected GC to occur before that stage, it is sensible that that object not be GC'd whilst that reference existed!
    Also, the GC can't free objects unless any references ( global and local ) to it are freed, have you confirmed that this is the case with the problem JNI code?

    Regarding the nio package, it supplies the equivilent of Win2K's memory mapped files and aync I/O, which are used to improve application performance.

    I also have enough real world experience to know that the only way you could switch from java to .NET is if you are running Windows2000 in a production enviroment, unless you are running non enterprise critical applications that's an unprofessional decision in my opinion. I've worked with NT4.0/Win2K for a long time, when things are running smoothly they are fine environments, however their black box, dumbed down nature makes them a nightmare when things go wrong. You know under Windows 2000 its possible for a run away process to exist that the Administrator can not kill, that can't happen under Unix where root is God.

    Just as an aside, java is available in a real-time implementation, which sort of indicates the uses the language/JVM can be put to.

    I'm still waiting for you to answer my previous queries, I have no problem with you pointing out if I'm wrong about any of this stuff.

    Please excuse me if you once again consider this as 'show uplike misguided missiles'.

    BTW, jdb is a java source code debugger, it's not for debugging the JVM, the JVM is written in C++...
  69. .Net not converting Java developers[ Go to top ]

    Cameron:
    The only JVM that was ever worth a damn was the Microsoft JVM. Most Java evangelist accept the fact that the MS JVM was superior to equivalent Sun JVM of the time.

    Cameron, you are dreaming about Java performance being so great. Look at the website you sent me: "Javalobby". Come on! I am not even talking about "benchmarking". I challenge you to compare the performance of a simple JSP and ASP.NET application. The result is clearly visible to the naked eye.

    My point is that Java is so slow, that its not even about "benchmarking". Java is so slow, its about a horrible user experience.

    By the way, my facts are correct. Java IS interpretted...per the spec. The CLR converts MIDL (MS Intermediate Language) to pure machine code, upon first execution.

    Java will be gone in two years.
  70. .Net not converting Java developers[ Go to top ]

    Tom: "The only JVM that was ever worth a damn was the Microsoft JVM. Most Java evangelist accept the fact that the MS JVM was superior to equivalent Sun JVM of the time."

    There are a couple of interesting points here:

    1. You like the Microsoft JVM. So did I, five years ago or so. (It was the fastest at the time of its release, and worked relatively well for most Java stuff.)

    2. You say it was "the only JVM ever worth a damn", yet "the MS JVM was superior to equivalent Sun JVM of the time." Since (from my testing) the Microsoft JVM has improved quite a bit (as the "CLR" now) and the IBM and Sun JVMs are both faster than the CLR by 10% or so, it would seem to me that your statements do not bear up under any form of objective scrutiny.

    3. "Most Java evangelist...." This comment added nothing to what you wrote. Your comment would have been quite acceptable as your own opinion, for example. Saying "most" is a simple debating/marketing trick. Most doctors prefer ... most auto mechanics suggest ... most phychiatrists believe ... etc. This is not CNN, and I hope you don't mind if I hold you to a higher level of discourse.

    Tom: "Cameron, you are dreaming about Java performance being so great. ... My point is that Java is so slow, that its not even about "benchmarking". Java is so slow, its about a horrible user experience."

    "Most" modern JVMs on Windows are still faster than the CLR from .NET on Windows. ;-)

    Frankly, your saying something does not make it fact. For anyone to claim that "Java is so slow" (or to claim the same for .NET for that matter) is ridiculous. Here's my comments on the subject from a while back:

    "After Edgar posted some benchmarks to the thread on TheServerSide, I decided to take the bait and see how Java was performing. I'd already run many micro-benchmarks between Java and .NET, and frankly I was surprised how fast .NET was for a 1.0 product. It was almost as fast as Java for most operations. On the other hand, it's not too surprising, since they had the JVM+JIT working back in '97 and five years to "project cool" it. So anyway, I built the .cs and .java files, and ran them with all the defaults, and Java was a bit slower ... probably 15% slower. I thought little of it: I've managed to find benchmarks in which .NET beat Java by up to 20% or so. In fact, if you really do "micro" benchmarks, you can find on some low-level operations a difference of over 100% either way, but those benchmarks change results from release to release (.NET 1.0 to .NET 1.0.2, and basically with every JVM on the Java side.)

    "Back to the tests, though. So I put the test in a loop (to let HotSpot do its thing) and set the -Xms and -Xmx settings to the same (since this test was basically just testing the garbage collector -- no reason to constantly recycle back to the OS what we'll need again in a microsecond) and guess what happened? Java whalloped .NET by over 30%. So after I finished patting myself on the back for all the hard work that Sun did on HotSpot, I temporarily returned to my sanity to conclude thus: It really doesn't matter that Java is faster than .NET. In this test, it was 30% faster. In most tests it is 10-15% faster. I say "so what?" for two reasons:

    "1. Performance is important, but it is not the reason to choose Java over .NET or vice versa. I've said it before and I'll say it again: There are still places where .NET is going to be a better choice than Java, it's just that those places are relatively few and confined to the market of supporting and upgrading Windows applications.

    "2. No matter how fast Java is (or .NET is), we know that tool developers, framework developers, and (of course) application developers will always be able to find a way to make applications crawl. You can't blame the language, and generally speaking you can't blame the platform."

    In other words, both are more than sufficiently performant for most of the tasks that they are being used. If you are choosing based on performance, then you're already lost.

    Tom: "By the way, my facts are correct. Java IS interpretted...per the spec."

    It's going to be hard to maintain an intelligent discussion with you if you cannot dwell in the realm of the factual. Having read the JVM spec, I will point out that the spec does not say that "Java IS interpretted."

    With Java, you do have the choice of many JVMs, some of which support interpretive modes.

    Tom: "The CLR converts MIDL (MS Intermediate Language) to pure machine code, upon first execution."

    Yes, that is called a JIT. Just like Microsoft's JVM. So is Java compiled to pure machine code, or is Microsoft a liar?

    Tom: "Java will be gone in two years."

    A technology such as Java exists and thrives because it solves real-world problems. The set of problems that Java is being applied to is growing. Furthermore, the set of problems that Java is being applied to are long-term solutions that will be used for the next 30-50 years or more. As a result, I find it highly unlikely that Java will be gone in two years.

    I personally find it much more likely that .NET (like COM and DCOM before it) will become yet another deprecated and discarded technology from Microsoft within the next five years or so. After all, it is being built and supported by a total of one vendor, and when it no longer suits that vendor, it will be relegated to the same pile that COM/DCOM finds itself in now.

    However, that said, I still think that there is a class of applications that .NET is an applicable technology to use, such as VB conversions and COM/DCOM/COM+ based appliations that need to be upgraded and made interoperable with modern enterprise infrastructure. .NET is _not_ bad technology, it is just a proprietary product from a single vendor. The world is as likely to switch en masse to PowerBuilder 12.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  71. .Net not converting Java developers[ Go to top ]

    Java is more than a language. It is a platform. I have no problems with the language...but the J2EE world. Just because you benchmark some "loop" and find great performance advantages for Java, doesnt mean that JSP/Swing/J2EE performs well. Benchmark based on that would not result so favorably. Who writes Java apps without using one of the above? All I am saying is that if you lined up a set of 100 applications (windowing/web) and let me try them out, based on usability alone...the average Joe could identify the ones written in Java with high accuracy.

    Just because "javaadvocate.com" says that Java beats the CLR by 10% doesnt make it so. FYI...the CLR 1.1 is coming out soon and is supposed to be driven toward optimization.

    The reason COM/DCOM are now going away is because Microsoft is investing in their platform in a way that Sun cannot do. If Java were to stick around, you would be using some version of J2EE 10 years from now.


    I would venture to say that .NET is not as proprietary as you say. The C# spec was released to the same body as the HTML specification...whereas, Java is held under lock by Sun. Pieces of .NET are available as shared source, and there are big efforts to port the platform to Unix and other platforms.

    You can take jabs at my credibility all you want, but I think your speculation that Java apps will even exist 30-50 years from now...says alot about your credibility.

    Who is going to drive the Java spec when Sun declares bankruptcy? I hear they're not doing so well.

    Are you also going to claim that there is a single Java IDE out there that compares to Visual Studio 7? Do I write my own before I start a project?! Yea..I know what you are going to say...IDE is not a good argument. Given a set of specs, and comparable developers, .NET development is much more cost effective....simply based on the good IDE. Wow, I can actually debug right out of the box!? What a concept.

    .NET 1.0 is about 10-11 months old, and is going head-to-head with Java in the enterprise.


    W
  72. .Net not converting Java developers[ Go to top ]

    Regarding my point about your using J2EE 10 years from now...I dont want to speculate on the life of J2EE, like you were.

    Let me rephrase that as follows:

    Microsoft has the balls to reinvent their platform. Its called "progress". Whereas, Sun doesnt have the money to do anything but "enhance" its platform. It has already made its investment in Java..now its in maintenance mode.
  73. .Net not converting Java developers[ Go to top ]

    Tom,
    You need to understand that the Java language and APIs are not just something invented by Sun and that other vendors develop to, like Microsoft. It's a wide array of vendors and other interests. So it's not just Sun investing in Java the way that Microsoft invests in its bit of software. It's IBM, Oracle, Software AG, Hyperion, Unisys and many others who help shape the APIs and the language itself. They all have an active and vested interest in the success of the language. So to say silly things like Java will be gone in two years is to not have a solid grasp of the types of solutions and platforms that Java and more specifically J2EE encompass and the level of industry support out there. It's very large and growing. To say it is in maintenance mode also is to not know much about what's going on in the language and APIs today. I suggest actually going to the Sun site and reading up on what's going on. It's making quite a lot of progress quite frankly. Obviously MS can make faster "progress", because they don't have to bother with outside interests.

    Cheers
    Ray
  74. .Net not converting Java developers[ Go to top ]

    I'm usally pretty good with perdictions. I'll give Java fives years or less...give or take two before it's every even considered in any enterpise solutions.
  75. Reply[ Go to top ]

    I'm usally pretty good with perdictions. I'll give Java >fives years or less...give or take two before it's every >even considered in any enterpise solutions?


    Man you are way to stupid to be using a computer.

    Do you realise many existing CRM vendors are porting their software to j2ee? Do you you realise java is replacing ABAP in SAP? Do you realise many telco hardware vendors use JCA to interface to their systems?

    Whatever MS can put in .NET, java can copy, and java will allways have the advantage of not being a MS product.

    Nobody wants to go back to the bad old days of being spoon fed whatever MS want to give them. Even if it tastes good for a while, the same thing over and over makes you sick. NOBODY!
  76. not an issue of Java vs.net[ Go to top ]

    The real factors at play are not being discussed here.
    Java is a tool promotoed by Sun and IBM, .Net is a tool being promoted by Microsoft.

    If you look closely at what is happening with the whole MS-vs.Sun thing, more and more I would put my money on IBM coming back into a major player (they always seem to get lumped as 3rd after MS and Sun when talking about development).

    Jump for a second to the real drivers of developement - the server and the those monster chips. McNealy got rid of all his execs to refocus his entire company into a machine that has one single objective - make his servers (64 bit) crush anything out there.

    IBM made the comment, about an alliance with Intel (where Intel takes over IBM's business of producing the chips instead - ya wouldn't that be nice Intel!), where they responded that there would only be 2 compnaies producing chips in 3 years - IBM and Intel.

    Those two comments in context give me some powerful predictions:

    One McNealy KNOWS Intel is a serious threat, thats not the sign of someone sitting on his unshakable fortress laughing at those who seige him - thats the sound of someone who knows that if that massive army gets to his door he's finsished!
    Prediction: Intel is going to be the force to recon with (especially when you consider that SPARC has been 64 bit since 1995 - an 8 year head start!)

    IBM is setting themselves alone in the list (they purposely made no mention of Sun), and IBM's chips power Nintendos, and Sony's, and they have complete control over their entire server configuration (all IBM parts - can't beat that integration) and have a serious performance edge over using proprietary parts.
    Prediction: IBM may be coming back into its glory years again.

    So the question of whether Java will be replaced by .Net - no way! But the question of whether IBM will replace Java by their own proprietary tool (like MS did with .Net)...thats a possibility (or a IBM-stranglehold on Java to conform to its needs - certainly a demise of purely open-source Java - which seems to be what so many people are grasping to hang on to).
  77. not an issue of Java vs.net[ Go to top ]

    In summary, it may not be .net at all that cause Java's demise - but Sun's fall and IBM's gain that could dilute the Java we know today (or whipe it out altogether).

    If Intel succeeds with its 64 bit chips (I think we see that even McNealy is afraid they will...), all the more power will be given to .net and MS solutions in the enterprise.

    So .Net could win 'indirectly' if Sun falls (I think they might), IBM prevails over Java (last man-with-money standing syndrome), and Intel succeeds in taking a good percentage of that enterprise market share (Sun is afraid of it, and IBM already beleives it are willing to share the stage).
  78. MLM redux[ Go to top ]

    You didn't answer my questions, not that I expected you to, so at least you are playing by the script.

    MLM: "A very interesting point about the whole Java movement (which amusingly enough seems more of life choice like permaculture or communisim now a days than a tool) is that they gained success by 'opening up to the world' in order to spread out development costs...not a bad thing."

    That's how most of the industry works, except for (cough cough sputter sputter) ....

    Honestly, look at Groove. Look at Eclipse. Look at Java. Look at what Microsoft is trying to do with .NET. Get people involved. Make them feel welcome. Don't make them sign brutal EULAs (heh!) ... let them actually contribute to the vision.

    Just because Sun was smart enough to do it with Java does NOT imply that they were smart enough to think of it first ;-) ... it's as old as competition in the high tech industry, and it's a sure sign that competition is alive and kicking.

    MLM: "But acceptance in the development community doesn't pay the bills."

    You, like most blinder-hobbled zealots, miss the point. Sun doesn't make money on software, they make it on hardware. Same with IBM (of yesteryear anyway). So to make their hardware viable (since they aren't Intel), it has to have software that runs on it. Java does that. It allows them to focus on two pieces of software (the OS and the JVM) and spend most of their efforts on the hardware (and selling the kit) because that's what they sell. IBM with their mid-range based on the Power4 (for example) couldn't sell into half those accounts if it didn't support middleware software, which is largely built in Java now.

    As far as your Microsoft / IBM / Sun conspiracies go, it's been talked to death. Get over it. They are big boys, and they will play rough, and sometimes they'll get caught cheating, but it will largely come out in the wash.

    MLM: "If Intel succeeds with its 64 bit chips (I think we see that even McNealy is afraid they will...), all the more power will be given to .net and MS solutions in the enterprise."

    Intel Itanium? The difference is that Sun ships more 64-bit chips each month than Intel has probably shipped in total. Give IA64 another 18 months and we'll see. Either IA64 or x86-64 will get a significant lead in the market. The other will probably have to morph or die, since they both think that they will occupy the commodity space in the server market.

    IBM is IBM. They don't have to be fast or inexpensive because they are IBM after all, but they _do_ have pretty fast chips now (the 4+) and the $/unit is "reasonable" for high-end systems.

    Sun is far behind the other major players in performance, and they need to get their ducks in a row in a hurry. Slow, expensive and hot (thermals) is not a good combo. They'd be _much_ more competitive at a lower cost, or with a faster chip ... pick one. Sun's always beaten the odds before, though, and although they are far behind on $/unit, they still are the volume leader in the high end space (e.g. the Unix market). OTOH they can't take it for granted ... let's hope for their sake that they get more agressive.

    I'm not "taking sides". It doesn't matter terribly to me which one wins. As long as they all get faster and cheaper. (A good thing about Java, eh? The customer wins any way it goes.)

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  79. MLM redux[ Go to top ]

    Good reply, quite agreeable!

    CP: let them actually contribute to the vision

    I completely agree with that statement and have never been opposed to that aspect of open-source. It promotes developer bonding, better code (tested by a more diverse set of developers, usually with broader skill sets than in-house chosen staff), and lets the developer feel they have realy contributed to that product.

    CP: You, like most blinder-hobbled zealots, miss the point...
    If by that you mean success-driven individual who loves money then that is me to a tee!

    CP: The difference is that Sun ships more 64-bit chips each month than Intel has probably shipped in total...
    Sure that is the McNealy's mechanism and scheme to counter - drive their hardware like a stampede while they can (and can justify those higher prices now). Just like drugs under patent - milk it dry while you can! Good marketing to me.

    CP: Either IA64 or x86-64 will get a significant lead in the market...
    Thats for sure, and really it will seriously sway all of our course one way or another (but non of us will loose - we'll just latch onto the winner). Ah yet, more success!

    CP: IBM...have pretty fast chips now (the 4+) and the $/unit is "reasonable" for high-end systems...
    Totally! IBM has learned some lessons over the years - they started out keeping everything to themselves and charging astronomical prices - people started to desert. So they sold the IBM PC patent to everyone in order to raise capital (under-estimating its potential to be a serious contender) - losing an even bigger share of the market in the end.
    Now a days IBM is smart. They have hedged their plans to come under the radar for years, creeping more and more into a player (but never acting like a lose cannon like McNealy to draw 'stupid' attention).
    Their new chip is fast, and even using IBM proprietary technology (which is much faster and highly tuned) is actually reasonable now a days.

    CP: Sun is far behind the other major players in performance, and they need to get their ducks in a row in a hurry...
    Yes, and that is why McNealy has taken full control of the helm (dumped his execs), and steered his ship with that single drive. Will it work - I dunno, they have pulled oiff some amazing stuff before and they do have huge head-start on the rest, and they do curently have the largest enterprise market share (there is always something to be said about the reigning champion keeping his title vs. struggling upstart trying to worm in)....somehow I don't getv the feeling that it will work though. Everything I see and hear from all sides, even the defecting execs says they are in trouble (i.e. they may not have what it takes to stay on top).

    CP: So to make their hardware viable (since they aren't Intel), it has to have software that runs on it...
    Well lets look at that as a viable 'stay-afloat-plan' though...Sun currently makes like 65% of thier total income from hardware, and while they have recently tried to dodge concerns about the Itanium 2 threat by saying customers buy Sun solutions cause of SunOne (the software) - lets be honest, are you going to stick yourself with the lowest performing server, and a waning corporation with a jerk at the helm or choose superior performance from another monster corp. like IBM?

    My moneys on IBM. Furthermore, when clients have to choose another vendor at all, because they feel their current vendor (in this case say Sun) doesn't meet their needs any more - suddenly the clients eyes are much more open when choosing the next vendor - and may very well choose Intel instead of IBM (and all the way down the line).

    Thats the plan baby, and while I too will tag onto who ever wins (so that I can win too, cause I like success), there are more 'good' choices on the horizon than ever before (and they are not all Java).
  80. New life for Sun java[ Go to top ]

    I was just thinking about the whole chip thing, and if Sun loses on the hardware battle (which I beleive they will) - they put everything they have left into SunOne and Java in general. Although they will become more depreciated like HP (packagers instead of makers), Java would have a good chance of flourishing (if Sun can stay afloat with just software).

    In fact I predict that in 3 years Intel and IBM will share rule over enterprise servers (in that order), DELL will bury HP and will become the main packagers (i.e. software enablers) for Intel, and Sun will become IBM's software enablers (IBM will sump their own stuff - or merge with Sun, Sun will become like HP is to Intel).

    That can't be bad for Java or .net though.
    Intel/DELL will be the power I formation for MS - .net will flourish big and small.
    IBM/Sun (still buddies) will be the blitz for Java - Java wil continue to be a dominant player in IBM soltuions.

    Look at these simple facts:
    Sun made 7.8 billion for their 64-bit servers (ol'SPARCy -and it IS old now).
    IBM made 5.7 billion from their 64-bit servers (Power4+ and it IS powerful).
    Intel only sold 100 million dollars worth of the Itanium 2 servers last year (many factors there) - which might seem trivial, but thats cause their Celeron, Pentium and Xeon took 16 billion (more than Sun and IBM's 64-bit put together!). So Intel's only competition is Intel...

    Side line:
    Not really germain to the survival of Java, but interesting is where the big boys make their money and how it turned out in the end...

    While software has a margin of like 85% (IBM's Zeitler says it's the best margin in business - second only to drugs!), Sun makes 65% of their income from hardware and they may lose that 65% very soon to Intel and IBM.

    Although no one could have forseen Sun ever losing on that ground, it may not have totally been a surprise since the hardware industry has a much lower margin (cause costs are huge!) and server consumers are much more apt to buy the cheapest and best product - regardless of who makes it.

    Microsoft makes like 60% of their income from their Office software (yes, like 80% margin), which is far from being in any danger. Their economic base is virtually unshakable cause people don't ever want to learn how to use another office tool (we just want to go home earlier, or have more time to golf, or have more time for coffee and chatting). Even minimal improvement to office software will keep us consumers forever just cause of human nature (and our love for comfort). MS can afford to do the kind of ventures and forays like .net and NT and always come out ahead cause their base in safe.

    Furthermore, in a tight market (such as now) lower sales will be seen in the hardware industry (client's 'live' with what they have), but software always does a good deal better (cause they 'need' it to function).

    I don't think anyone can blaim Sun for having a hardware base, or MS for having a software base - its just where they came from (and man, look where they got from there - simply amazing!) but it does makes one think now a days (if you are starting your own corp.) where to put the base.
  81. Microsoft's nightmare[ Go to top ]

    Intel is a big supporter of both Linux and Java. So is IBM. So is HP. Microsoft is not a shoe-in for the 64-bit world ... they are going to have to earn their keep, which is hardly a bad thing.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  82. Still going on![ Go to top ]

    I didn't realize that this discussion was still going on, though it dead long ago.

    Just for the record:

    Cameron, the role of an elderly statesman doesn't suit you very well.. Remember what I said about Diderot and hypocrisy? Why don't you come out as the Java zealot you really are - I have no problem admitting that I am a .NET zealot - why should you?
  83. ad infinitum[ Go to top ]

    Rolf: "Cameron, the role of an elderly statesman doesn't suit you very well.."

    Thank God! I don't want to be an elderly anything for quite some time.

    Rolf: "Remember what I said about Diderot and hypocrisy?"

    Who? Diderot? Doesn't he play for the Nuggets or something?

    I spent much more time studying Locke, to whom we owe the pleasure of the cause of most of Diderot's writings. Further, I would suggest, we owe most of what we consider "western democracy" to concepts that he first introduced.

    Rolf: "I have no problem admitting that I am a .NET zealot - why should you?"

    Well, to be honest, I'm not a .NET zealot.

    I think that .NET, while similar to Java in many ways, is inferior technically because it trades elegance for some minor functionality gains. It's a common mistake made by junior engineers, however it is far from fatal. It's just my opinion, though.

    From a business perspective, I think .NET is bad for Microsoft and bad for Microsoft's customers; but if it turns out that it is not so bad for Microsoft, it will probably still be bad for most everyone else. To me, that isn't very responsible for a would-be industry leader, but that's just my opinion too.

    Rolf: "Why don't you come out as the Java zealot you really are"

    I like Java, that much is true. If having such an opinion makes me a zealot, then the world is full of zealots.

    However, I believe that "zealot" should be reserved for someone who cannot rationally espouse his/her views, and instead must resort to blindly following their beliefs. Whether either of us would fit that description is certainly for neither of us to fairly decide. Nonetheless, I should expect that it will not halt your attempts to do so. Please, continue.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  84. ad infinitum[ Go to top ]

    With Microsoft averaging 600K in revenue per employee I would like to hear your definition of "Bad for Microsoft"?
    Also, how can just throw out there that it is bad for Microsoft customers? There are many companies that have successfully deployed .Net apps. Would you have these mom and pops or departmental apps be J2EE? Please Explain.
    Chuck
  85. Microsoft's nightmare[ Go to top ]

    Thats totally true!

    Worse yet, it will take a great deal of convincing to win the big corps. as MS has mid-range rep. and a bad habit of releasing software too soon (for marketing purposes) - that kind of risk is not acceptable with the big boys (after all thats how IBM has survived, being tried and true blue - not exciting/cutting edge but dependable).

    The only product I have seen from MS that I have had near flawless execution has been their bread and butter - the office suites. Although Win 2K is fantastic (I've had my 2K server running for almost a year at a time - which is saying a lot considering NT would need a reboot every month) it still had the normal flux at the begining and XP is a peice of crap (greate concepts - but released way too soon as usual).

    Like you said they have to earn that space and it won't be any cake walk thats for sure!
  86. agreed[ Go to top ]

    Me LikesMoney: "The only product I have seen from MS that I have had near flawless execution has been their bread and butter - the office suites. Although Win 2K is fantastic ..."

    Agreed: W2K + OfficeXP is pretty darn good stuff.
  87. Shitty products![ Go to top ]

    And SQL Server is shit too?

    !

    Regards
    Rolf Tollerud
  88. SQL Server is Shit Too[ Go to top ]

    As an off-topic bit of fun: Rolf's comment was amusing in light of a virus that hit my current client. They are big proponents of an all MS solution. Last friday they got hit by the SQL "Slammer" virus that essentially brought the company to standstill. This is a very large company (at least since their recent purchase of my former client). Not only did the virus infect all of their SQL Server machines, but any software that uses the MSDE. That's a lot of software and a lot of machines. I won't repeat the list of affected software here. No, SQL Server certainly isn't shit, per se, but I would remain unconvinced of MS's viability as a true enterprise partner.

    Cheers
    Ray
  89. when you are in a glass-house..[ Go to top ]

    "I would remain unconvinced of MS's viability as a true enterprise partner"

    Well,

    I remain unconvinced of Sun's viability as an enterprise partner too!
    http://www.internalmemos.com/memos/memodetails.php?memo_id=1321

    The best thing with Windows XP SP1 is that you can disable Sun VM!

    Regards
    Rolf Tollerud
  90. SUN VM ?[ Go to top ]

    Rolf, do you have to point your finger everytime :D
    and Windows XP SP1 have Sun VM ? yay now i can develop apps on it.

    Java Z.. em Fans,
    Adam AC
  91. SUN VM ?[ Go to top ]

    Of course I have Sun VM on my computer and of course I develop on it (but not client apps). The point is that I do not want to use it when I play chess with my son over internet. The MS VM is OK. But when I use the Sun VM (for applets) the computer seem to strain it selves, the cursor is "shivering" and everything generally feels unstable. On the serverside I have no problem.

    Regards
    Rolf Tollerud
  92. Shitty products![ Go to top ]

    Completely off topic of this thread:
    SQL Server is a fantastic product - but like many of MS's core products it tries to be everything to everyone - having 10 ways to do the same thing (far from 'unbreakable' when it comes to security).

    They need to do what they have done for IIS - which is provide a little secure setup screen that lets you enable only the options you need (and all of the options turned off by default).

    Things like O-SQL,and the new XML and web-enabled sql interfaces are neat (if you use them), but they all have very different purposes and should be turned off (i.e. completely disabled) until someone needs them.

    It urks me that everytime MS comes out with another features for SQL server - its enabled right out of the gate - its dangerous (web-direct crap is just crazy - use the frikken tiers, thats what they are there for!!!) - and they always fix it AFTER someone gets hurt.

    Just griping - cause SQL is fantastic, and the interface is surperb (shudder...why can't you make a good interface Oracle...why? please!!!). But when I said only Win2K and Office have come close to being a flawless execution - thats cause SQL server (as good as it is) has never been flawless (as demonstrated by SQL Slammer - oh the downtime!).
  93. A very interesting point about the whole Java movement (which amusingly enough seems more of life choice like permaculture or communisim now a days than a tool) is that they gained success by 'opening up to the world' in order to spread out development costs...not a bad thing.

    They got lots of geeky developers to fix stuff for free and it worked, the Java community has contributed substantually to improved functionality and certainly its acceptance in the developer commmunity.

    But acceptance in the development community doesn't pay the bills. People seem to miss that Sun and IBM (yes evil corporate entities that have only one motive - profit no matter what you say) have done alot to get Java used in the corporate world. No matter how good those wonderful little free Java apps running on their revamped 486 running their free copy of Linux is - those are not the projects that make corporations use Java.

    If a company like Sun doesn't keep holding the flag (i.e. the go bankrupt or sell out their assets to a pure money corp. who cares nothing for 'open source') then Java itself will be in serious danger. The legacy of Java will not 'live on' unscathed as many seem to beleive.

    Furthermore, McNealy singling Sun out as the stupid guy in the whole MS thing may have been the first blow to corporate-Java use as a whole. MS-bill can now concentrate fully on destroying Sun, justifying it by saying 'We don't oppose Java - we love Java, we just don't like McNealy' and everyone (even those in the Java community are agreeing) - 'Ya McNealy deserves it, he was stupid and got way too personal - crush away Bill'.

    And even though it may look like they are only striking one target (Sun), they are attacking corporate-Java acceptance as a whole.

    Will IBM come to Sun's aid - no way (not if they like money, and they do!). While MS is applying pressure to bring down Sun, IBM and the others will jump all over that opportunity to disassociate themselves from the Sun-mess (all the while gaining the last corp. strangle hold on Java and IBM is not quite as 'open' as Sun).
  94. Mr. Purdy: "I personally find it much more likely that .NET (like COM and DCOM before it) will become yet another deprecated and discarded technology from Microsoft within the next five years or so. ..."

    I have been following your "less" knowlegeable replies as regards COM/DCOM in general and C++/ATL in particular. Read DON BOX's latest book on CLR. Have you even written a "hello world" program in ATL before? I suspect. Seriously.
    Who cares if it works only as a windows console executable, as long as it gives performance of C++, damn it!

    Again, folks such as yourself should not claim knowledge about CLR by reading up just on MSDN alone. Read DON BOX chapter 10, "CLR Externals". As far as MS exists, COM/WIN32APIs will never die. CLR is implemented as a family of WIN32/COM DLLs. Tou can write unmanaged code in C++/ATL still. Good luck reading.
  95. .Net not converting Java developers[ Go to top ]

    Purdy is one of those people who would rather argue the merits of a depricated technology, than take the time to learn as technology progresses.

    Good luck Purdy. When the plug is pulled on "theserverside.com", you will know that you stuck around the java community a little too long.
  96. .Net not converting Java developers[ Go to top ]

    Purdy:

    Establish some respect for Don Box, and you will be better off.

    Cheers!
    Cole
  97. .Net not converting Java developers[ Go to top ]

    This site crawls. Look...its JSP!
  98. .Net not converting Java developers[ Go to top ]

    Tom,
    How is Java (and/or J2EE) depricated technology? Please explain. Please tell us how you know Java to be in maintenance mode - presumably you know all that's going on with the language and with J2EE.

    Thanks
    Ray
  99. .Net not converting Java developers[ Go to top ]

    I say that Java is in maintenance mode because Java is constantly "updated"... Noboby steps back and says: "Wo....we need to step back and look at this technology in terms of the big picture. We need to do some major re-architecting."

    Sun's approach is more like..."Hey..web services...I think we should write a Web Services SDK."
  100. Ravi: "I have been following your "less" knowlegeable replies as regards COM/DCOM in general"

    I am a Microsoft MVP for COM/OLE2. If you don't like my level of COM/DCOM/etc. knowledge, make fun of Microsoft for making me an MVP. I was building software to make dynamic COM stubs on the fly in Windows (both 16 and 32 bit) back years before Java existed. I know why I choose not to focus on Microsoft Windows proprietary tools and technologies ... you see, I've worked on both, and I developed a preference.

    Ravi: "Read DON BOX's latest book on CLR."

    Why? Why not PowerBuilder unleashed?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  101. Cameron,

    Learn from folks like Don Box. In this country, you should not disrespect the folks who have earned to be celebrity, in niche areas like COM/CLR. I certainly think you happen to have MVP stamp for the sake of it, and have started to beleive that soon we will have 80% .net programmers with no clue about the design of CLR. When I asked you to refer to Don's book on CLR, you say read PB unleashed. You are out of your mind on this friday, and still, I say to you, happy hollidays. You need to change your email writting habit a bit.

    Regards,
    /Ravi
  102. Ravi: "Learn from folks like Don Box. In this country, you should not disrespect the folks who have earned to be celebrity, in niche areas like COM/CLR."

    I don't know Don. I cann't disrespect him, because I don't know who he even is, but I don't have time to read about the .NET product right now either. The comment about "why not read a PowerBuild Unleashed book instead" was intended not out of disrespect to PowerBuilder but rather because .NET and PowerBuilder are competitors: Proprietary development tools and runtimes. That market is not very interesting to me right now.

    Ravi: "You are out of your mind on this friday, and still, I say to you, happy hollidays."

    Happy holidays to you too, Ravi.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  103. .Net not converting Java developers[ Go to top ]

    I would like to make a general statement based upon my personal experience regarding java versus .net. Below are the experiences of a "real world" programmer, developing a "real world" application for nearly 10,000 "real world" users... Benchmarks aside..

    I will not claim to be an expert in either realm as I am not. I am an average programmer with average experience. I do not know the documented performance statistics for each facet of java and .net and I usually don't seek out benchmarks to see what is best. I simply look for what I like the best and what seems to work for me. I will share my experience while experimenting with both technologies to build the same project...

    I was recently given the task of building an online calendar / group scheduling application for our company Intranet, (very large company, over 10,000 users). This app would need to connect to either Oracle or SQL server, which ever I deemed would be the more suitable DB platform to house the calendar data, and the calendar would hook into that database. It would essentially serve as the front end to display event information for the company.

    When I started coding the application I had relatively little knowledge of either technology (.net and Java, most of my experience was in VB 6 and .asp). I figured that because of this I would be able to keep an open mind and give either technology a fair shot at the final build and deployment throughout the enterprise.

    I started coding in Java... I used many books and online sources to aid me and I did make progress; good progress. What I started to find was that I felt as if I was writing an awful lot of code for what I felt was nothing more than a calendar interface, simply used to display data... the calendar would also need an interface that would allow administrators to add, view all current, modify, delete, and search for events based on supplied criteria.. Anyway, by the time that I got to nearly 2000 lines of code (I was a little over halfway finished) and two months of time spent, I decided to stop there and try .net instead.

    I did the same thing that I did with Java.. because I was "green" behind the ears on both technologies I picked up my full compliment of books and gathered my Internet sources (just as I did when I started with Java) and got to work... Within two weeks of working on the project part time, and only 1500 lines of code for the entire application!!, (keep in mind that I could only spend a couple of hours each day on this project) I had my first build (in ASP.NET) ready and available for the QA department. Those who QA'ed the app stated that they loved the speed of the application and simplicity of navigation. I was surprised as well, especially at how much better the calendar performed over its Java counterpart. It was so fast and "clean" whereas the Java version seemed slow and "clunky". .net was simply "visually" faster.

    From then on I was sold! I have since rewritten the calendar into an application that is even faster than the first build, (hard for me to believe at first) and with more features that before. I found that producing applications of any kind with Visual Studio 7 and .net technologies has proven to be much less time consuming than with any other IDE that I have worked with. I enjoy the fact that I can indeed debug "right out of the box" over not really being able to debug very effectively with my Java IDE, (debugging was a hassle and it didn't work very well with my Java IDEs (I tried two of them) this was a major setback, in my book).

    In conclusion, I made the choice to code mostly in Microsoft .net technologies, (any chance I get, for the most part) simply because the code that I can produce over Java is much faster, cleaner, and more efficient. I also concluded that while working with Java, everything that I wanted to do took at least twice as long to do, versus working with .net.

    Just my 2c

    Chad
  104. .Net not converting Java developers[ Go to top ]

    Why would the most powerful software company in the world spend $700 million to create something inferior?

    The only people who believe in Java are the ones who spent so much time ramping up on it.

    I believe that Chadwick's experience is typical. The next time I need to write code to run in my toaster, I will think of Java, ladies! Don't worry!
  105. .Net not converting Java developers[ Go to top ]

    <quote> Why would the most powerful software company in the world spend $700 million to create something inferior?</quote>

    Dunno, but they've been doing it with Windows for years now.
  106. .Net not converting Java developers[ Go to top ]

    Tom Cole, rhymes with Troll...
  107. .Net not converting Java developers[ Go to top ]

    Chad,

    Your comments are interesting. Do your think the results would have been different had you first tried coding the app in .net, then tried java? In other words, did the learning curve help you significantly the second time around?
  108. .Net not converting Java developers[ Go to top ]

    Here are the two competing philosophies:

    Sun:
    "Java Everywhere" -Run Java on everything, and it all talks together. Great, but Java doesnt talk to anything else very easily. Java is about attempting to dominate the market, even more than Microsoft. Before anyone barks, yes, Sun supports Web Services....but do you think that Sun liked adopting SOAP (Don Box/Microsoft) as a standard? Definitely not, because SOAP undermines "Java everywhere".

    Microsoft .NET
    "You use your platform, we'll use ours. We promise to talk nicely."-Microsoft knows that you Java losers are never going to give up promoting 'the movement'. Therefore, .NET has the most mature implementation of SOAP/XML Web Services on the market. Microsoft leads the industry to finally establish these standards. The approach assumes that environments will remain diverse, and that integration is the key.

    All CORBA "kingpins" keep quiet. CORBA is for people who are incredibly bored...it is DEFINITELY going away.
  109. Hey Chad,
    Just curious, did you find any documents of value from Microsoft directly? I know they have a best practices group that is supposed to help with getting "green" developers quickly up to speed
    Chuck

    > I did the same thing that I did with Java.. because I was "green" behind the ears on both technologies I picked up my full compliment of books and gathered my Internet sources (just as I did when I started with Java) and got to work... Within two weeks of working on the project part time, and only 1500 lines of code for the entire application!!, (keep in mind that I could only spend a couple of hours each day on this project) I had my first build (in ASP.NET) ready and available for the QA department. Those who QA'ed the app stated that they loved the speed of the application and simplicity of navigation. I was surprised as well, especially at how much better the calendar performed over its Java counterpart. It was so fast and "clean" whereas the Java version seemed slow and "clunky". .net was simply "visually" faster.
    >
    > Just my 2c
    >
    > Chad
  110. The forums here are wonderfully entertaining (I happened upon this site looking for design patterns).

    It amazes me that we (developers) still argue about how is better, what language is best, which company is best, blaa, blaa.
    I find it funny - yet completely understandable.

    Think Cola-wars and you see a striking parallel! How many people adamantly feel that Coke is the king, or Pepsi is the king? It is completely subjective and a matter of taste – a taste that is very much skewed on which one you tasted first (or had a good amount of exposure too first).

    If you started with Pepsi, you’d be a Pepsi guy. If you started with Coke, you’d be a Coke guy. I ask you though, who is winning the cola-war though? Last time I checked both Coke and Pepsi are still around and making more money than most other companies in the world (including software). Last time I checked they both held pretty even shares of the market (oh yes it sways this way or that - but neither of them are ever hurting). Do we, the consumer, ever win though? Somebody else (not us) get our money, so we sure didn’t win (although I’m sure that we are content that our favorite soda drink is still around when we are thirsty).

    There are valid arguments by many who post here (not all – the bigots are just being childish).

    What do you need from your development languages? What are your drivers? Business? Rapid development? Cheap development? Robust, war-hardened code? The fastest code in the west? None of the languages available meet all those criteria – each have their purpose and niche.
  111. Sorry, part 2 - didn't know how much I could post at once here.

    Some comments to make people mad at (cause thats what people on this forum seem to be best at - getting mad, he he):

    Statement: "The only JVM that was ever worth a damn was the Microsoft JVM. Most Java evangelist accept the fact that the MS JVM was superior to equivalent Sun JVM of the time."
    Opinion: MS JVM was good (for its time). It made a number of improvements that made our development life simpler.
    Flipside: It had some serious design flaws. We wrote an entire framework using the MS JVM, and after being plagued by weird memory leaks we were told by MS (while in Seattle) to dump it – they knew it had issues and they weren’t going to pursue it cause of the impending legal crap.

    Statement: Java and JVMs are slow.
    Opinion: That was true of the initial ‘free’ releases, and still continues to plague ‘free’ versions today.
    Flipside: How many compnaies are running production apps with the free crap? You get what you pay for. There are many optimized JVMs by real vendors (IBM and Sun) that have great performance (but you have to pay for them).

    Statement: Java applets are the crappiest thing on this planet.
    Opinion: True. Still to this day. Oh the pain.
    Flipside: Active X, while always working much more reliably, had (and has) a lot of difficulties with regards to downloading and registering them, versioning, etc - not fun either.

    Statement: Java IDEs are the worst pain on the planet
    Opinion: True, for the most part, but again you get what you pay for. Most of them are offered as ‘free’ or packaged with other solutions. Therefore the time and effort has not been spent to make the tools user-friendly, or to even spend money on a decent graphic designer.
    Flipside: But they are almost all free.

    Statement: Visual Studio is one of the easiest environments to work with and makes developing stuff sweet.
    Opinion: That’s completely true. MS pays people to make their products – and you get what they pay for. If they know one thing well – it is UI.
    Flipside: Its not free

    Statement: VB and RAD tools promote low-skilled developers, and degrade the level of code in applications.
    Opinion: That is true, to some extent. Many, many developers were introduced to the world of coding through VB because of its ease of use and low barrier to entry. Many poorly architected solutions resulted from the use of VB (in the hands of those poorly educated VB developers).
    Flipside: VB had (up until .net) 60% of the developer market. The ease of application development and cheaper price of those developers introduced way more companies to software solutions than ever before. Does any Java or C++ developer feel their job threatened or devaluated by hoards of VB developers – I seriously doubt it (if you were afraid of them stealing your job, your not a very good Java developer).

    Statement: MS has abandoned their own by changing VB.net to a more advanced development language.
    Opinion: True. They have taken the gamble, that the VB developers that were skilled developers will have all the resources to come up to speed and have given them a push in that direction. The gamble WILL pay off. The ones that can’t cope will drop out, and the ones that were stronger in the base software design skills will simply take a little pain then succeed more than they could have ever done with the existing limitations of COM+ and VB.
    Flipside: MS could lose a huge amount of their developer market share in the process (remember upwards of 60% of the developers were using VB) – that’s a lot to lose.

    Statement: Open source promotes better code, because ‘all of its guts are hung out for everyone to see from day one’.
    Opinion: True. If we could all afford an army of testers for our applications we would reduce the errors drastically (I would say it incremental – the more testers the fewer errors will slip by and the less embarrassed you will be when it goes prod.).
    Flipside: Open source also means giving away intellectual property that use to make money. Often in this industry, you have to utlize every edge you have to survive. Giving away your edge for free is like baking bread and giving it away for people to taste – not keeping any for yourself. People will like you a lot, but you’ll go hungry :)

    By the way, as a follow up to my last post - I like Coke. Mmmm coke.
  112. Incidentally, to actually address the original topic of this discussion (which most replies have not), we have actually recently converted from Java to .Net (C# only) and found it to be a painless experience.

    Why we did it:
    1) We are a vendor-neutral software development company and were supporting both a J2EE and COM+ version of our products/framework. It was getting to be too much work, so we had to chose one.
    2) .Net's easy-to-use IDE (Visual Studio) was the primary deciding factor (VS buries ANYTHING for Java - even the stuff you pay for (IBM, Sun) sadly).
    3) .Net is newer (and by nature of all software products that are the newest - they usually incorporate all the latest/greatest features of its predecessors) - and it does.
    4) Microsoft is one of the strongest software companies out there and they are not going away - ever. There marketing IS the best out there (no one can refute that) and they do much more of the selling than we do (unlike the uphill battle with Java - simply because too many diverse companies in the Java game).

    How we succeeded:
    1) Our primary concern is delivering the best solution to our clients, not what is easiest, not what we have vested interest in (due to a strategic alliance, etc.). Netscape is a perfect example – we dumped them a while back because they simply sucked (and made our life suck). So many more hours were spent trying to get normal web stuff to work in Netscape (like frames), and we always ended up paring the product down just so it work in Netscape (lowest common denominator) we had to dump it.
    2) We love what we do, and still enjoy software development so much because our focus has always been on the core software design skills, methodologies, design patterns, which constantly improve – yet remain so much the same. So we never had any bigots to begin with.
    3) Many of us were C++ developers originally, who converted to Java (cause it was the best at the time and made some great improvements to C++) – so it was easy to convert to the next evolution in the C chain. It is amusing to find so many people adamantly hold the Java-grail since Java is a derivative (subset for the most part) of the C++ (the architect of OO), which was a derivative of C (just an easier way to use machine code)).
    4) We like money, so it was easy for us to make choices that would grow our company, rather than be the starving artist!

    Although we are definitely looking forward to the Mono projects and other ports so we offer other platforms, we have found .Net completely easy to use, fantastic performance, and all tools well worth their cost.

    Try it, you'll like it.
  113. .Net not converting Java developers[ Go to top ]

    <Q>We are a vendor-neutral software ... </Q>

    Were. You are not anymore.

    <Q>
    .Net's easy-to-use IDE (Visual Studio) was the primary deciding factor (VS buries ANYTHING for Java - even the stuff you pay for (IBM, Sun) sadly).
    </Q>

    You can say and believe it. Doesn't make it true. I use both WSAD and VS.Net. I believe WSAD is better. And Carlos has proof otherwise.

    <Q>
    Microsoft is one of the strongest software companies out there and they are not going away - ever
    </Q>

    Now that you have cast your lots with them, you had better hope so. BTW, can borrow your crystal ball. History is littered with companies that were never going to go away.

    <Q>
    Our primary concern is delivering the best solution to our clients, not what is easiest
    </Q>

    Hmm. So making it so they have to buy Windows if they want your product is the best solution?

    <Q>
    We like money, so it was easy for us to make choices that would grow our company, rather than be the starving artist
    </Q>

    True. But you do have to eat in the future too.

    <Q>
    Although we are definitely looking forward to the Mono projects and other ports so we offer other platforms, we have found .Net completely easy to use, fantastic performance, and all tools well worth their cost.
    </Q>

    Ahhh. My 'favorite' thing about the whole .Net thing. I hope you like walking on quicksand and clouds. I wish the Mono project all the luck. But I wouldn't bet anything on .Net being truly viable on anything but Windows. And when and if Mono gets it all together, expect few, if any MS.Net apps (Winforms/ASP.Net) to convert easily to Mono.Net. And if Microsoft feels truly threatened at all by Mono - watch out!
  114. .Net not converting Java developers[ Go to top ]

    Good reply!

    I agree with most of what you have said...

    WSAD better than Vs.net...
    Ah...no. WSAD is better than the rest of the other java crap out there - but thats not hard to beat :)

    Any company can fall...
    He he, your right that 'any' company can fall. If Bill Gates suddenly starts acting as stupid as Scott McNealy - there is a real possibilty of damaging MS (can't see that happening though, as funny as it would be).

    Is making our client use windows offering the best solution ...
    It certainly depends on what 'best' means to you. If 'cheapest' (i.e. free) is best, than no. If the most features, quickest time-to-market, excellent scalability and performance - without being astronimically expensive - than yes, for now, until other platforms are available the windows solutions are the best for our clients.

    In truth, the majority of our Java deployments we had in the past (on average - not all) were the really small clients that wanted everything for free (including our time, he he) - not alot of money to be made on those clients. So we are ok to alienate those clients with windows-only solutions.

    Hope you like walking on quicksand and clouds...
    For sure we do, else why would we be in this business now a days! If we wanted certainty, and training wheels for our mountain bikes, and a safety harness while we climbed mountains, we would leav this business an persue more 'stable' business pursuits - such as something in the entertainment field (they make money in all climates).

    Can MS rear its ugly head and smite Mono and any .netlet?...
    Yup! But will they ever feel threatened by such projects - nope. They will never convince the majority of businesses that currently run unix servers to switch - those clients have unix staff that will fight tooth and nail to keep them, and they don't want to learn other skills, so MS has a vested interest in simply 'enabling' their software (and the software written by third-parties) to run on unix.

    Maybe someday MS will fight every single battle - but I doubt it. They are the king of business - and that dictates fighting the battles they can win (and have a good chance of winning) first - giving themselves much more bargaining power when it comes to tackling the 'tougher' battles.

    But risk has its reward, and reward is what its all about!
  115. .Net not converting Java developers[ Go to top ]

    Mark - you said 'We are a vendor-neutral software ... Were. You are not anymore.'

    Thats becoming true. We used to support Java and COM+, Oracle and SQL Server, IIS and Apache, Unix and Windows, IE and Netscape.

    Now partly due to necessity and partly due to choice we are more of a MS shop.

    The necessity was to dump Netscape support - they were dragging us down, doubling our dev-time (atleast - sometimes causing redesign(down-scaling)). Luckily their usage now a days is soooo low it was an easy choice - they're gone!

    The choice was to choose one, migrating both our COM+ and our Java streams into one - in this case .net (C#).

    Temporarily, due to the .net move, our new solutions is only available on the windows platform as well.

    So in a matter of 6 months we have gone from all those down to just C#, Oracle and SQL Server, IIS, Windows, and IE - so it appears as though we are a MS shop now.

    But the funny thing is, that we are still able to service all of our new business - so, maybe its not all a bad thing :)

    Furthermore, 7 dev. shops (some medium, some large) here in Calgary, Canada (primarily oil and gas, IT, and skiing - just like Colorado - in case anyone wonders what we do here) have made the jump either to 'also' do .net dev or replace Java with .net in the last 6 months too.

    Also of interest, is the fact that all of the telecom companies in Candada, most of the energy companies (power/utility), and a lot of Oil and Gas companies (who used to be exclusively Java/Oracle shops) have made the migration to MS-only soltuions over the last 2 years.

    Probably doesn't hurt to be so close to Seattle - but there is a trend and its looks like MS-sunny-side-up!

    Bruce
  116. They Call Me ... Bruce[ Go to top ]

    Bruce: "The necessity was to dump Netscape support - they were dragging us down, doubling our dev-time"

    What century are you in? Netscape support?

    Bruce: "Also of interest, is the fact that all of the telecom companies in Candada, most of the energy companies (power/utility), and a lot of Oil and Gas companies (who used to be exclusively Java/Oracle shops) have made the migration to MS-only soltuions over the last 2 years."

    Do you realize how ridiculous your assertion is? Do you know what Telcos even do? Do you understand that they like their systems to run and keep running? Rebooting things is not an option. I've yet to see Windows in the core operations section of a Telco.

    Sounds like you got a job upgrading some old COM+ applications to .NET for some Word/Excel integration or something. Good for you. Could you make some predictions on Telcos for us?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  117. They Call Me ... Bruce[ Go to top ]

    You must be kinding. No your facts before your respond.

    Read on and weap. The time is coming....

     Note that while there are many enterprise shops that are having big
    issues getting J2EE apps done/deployed successfully using Java app servers,
    few of them are willing to go on record about such negative experiences for
    obvious reasons. However, here is a SMALL SAMPLING of some more customers
    that have tried the .NET technology, liked it, and are openly talking about
    their successes. Some (not all) were using J2EE previously, and/or plan to
    use both .NET and J2EE together.
     
     Bank of New York
     ----------------
     Retail Funds Management System (RUFUS)
     -250 servers managing 2M investments today, needed to grow to manage 10M
    investments
     -Customers include Merrill Lynch, AXA, Standard Life, Sun Life, Leland
    General, ý
     -Originally chose Java/J2EE
     -Six months of effort transitioned over
     -Now building .NET Smart Client application
     -XML Web services platform support was key
     -Developer productivity clinched the decision
     
     
     NCR Hausbank
     ------------
     Chose .NET over J2EE for XML Web service account management system
     
     -Market leader in property management
     -200,000 rent deposit accounts for 9,000 property managers
     -High volume business with low margins
     -Chose .NET over J2EE for time-to-market
     -Previous solution was Java/Oracle
     -Completely automated account management
     -Accessible by any device via XML Web services
     -Over 5,000 simultaneous users
     -Increasing revenues without increasing staff
     
     UnumProvident
     -------------
     Leading benefits provider switches from Java to .NET; triples
    productivity
     
     -Partners integrate into core business processes
     -New solution manages enrollment process, claims processing, and other
    customer-facing contact
     -XML Web services built on the .NET Framework enable partner base to
    integrate with its core business processes
     -Development with .NET takes one third time of Java
     -Java solution slow and over engineered
     -With Visual Studio .NET they expect to cut development time from 18
    months down to six months
     -Partners benefit
     -Service is much faster and more flexible
     -New partner campaigns can now be implemented in days (used to take
    months)
     
     ýXML Web services provide the ideal means of exposing core business
    processesýand thus our core competenciesýfor real-time access by partners
    over the Internet. And exposing functionality as an XML Web service is as
    simple as writing one or two lines of code, with the .NET Framework handling
    the rest.ý
     -Rick Klausner, UnumProvident
     
     
     Newport News Shipping
     ---------------------
     XML Web services built on .NET enable new business models for shipbuilder
     
     -Faster time-to-market delivers $36M incremental rev.
     -Gains achieved through productivity of .NET platform
     -XML Web services enable new business opportunity
     -Now refit many ships with same Web services infrastructure
     -Worked with Applied Information Science Systems, a .NET consultant, to
    build wireless services
     -Dozens of sub-contractors integrated together despite disparate IT
    systems
     -.NET and the USS Nimitz
     -Newport News runs large projects for refitting aircraft carriers
     -Naptheon, a subsidiary, built XML Web services and applications on .NET
    to make it possible
     
     

    http://msdn.microsoft.com/vstudio/prodinfo/casestudies/newport/default.asp
     
     ACCOR Services
     --------------
     Chose .NET over J2EE for XML Web services implementation
     -Leader in travel, tourism & services
     -Number one hotel operator: 3,500 hotels in 88 countries
     -Also provides services to 12M people in 31 countries
     -Building an XML Web services solution for integration with legacy
    systems, business partners, customers
     -Building out internal framework & business objects on .NET
     -Creating Europeýs first hospitality market e-procurement tool
     -Chose .NET over BEA WebLogic
     -Faster time-to-market through developer productivity
     -Better performance vs. J2EE
     
     
     Canadian Pacific Railway
     ------------------------
     Implementing XML Web services on .NET for integration, interoperability
     $3.65B shipping company
     -14,000 miles of track make it one of the largest railroad systems in
    North America
     -Many disparate backend systems
     -Chose XML Web services for instant access to data & legacy integration
     -Chose .NET for productivity and flexibility
     -Canadian Pacific Railway developers gain productivity of Visual Studio
    .NET, language choice
     
     
     
     Continental Airlines
     --------------------
     Major airline develops mobile travel planning application with .NET
     Fifth largest US airline
     -122 departures daily to 225 destinations
     -Serves more international cities than any other US carrier
     Flight schedules, rebooking on any mobile device
     -Available on cell phones, PDAs via XML Web services
     -Scales to 80,000 simultaneous requests
     -Chose .NET for time-to-market, host system interop
     -Cut deployment time by 90%
     -Integrating host systems
     
     ýTraditionally, it might take weeks or months to update applications in
    response to customer feedback, but with .NET we can work dynamically,
    growing our technology as itýs being used.ý
       -Ferdy Khater, Director of Application Development
     
     
     Marks & Spencer
     ---------------
     .NET helps apprehend suspects every day with Profit Protection for .NET
     -.NET provides fraud detection
     Loss Prevention application provides fraud detection and prevention
    intelligence for 350 stores
     -XML Web services built with C# on the .NET Framework
     -High-performance solution
     -Processes 2 million XML documents every day
     -2 TB of data in SQL Server 2000
     -100% uptime with .NET Framework and BizTalk Server
     -Return on Investment of 415% in first year alone
     
     ýWeýre using Visual Studio .NET because it is the best tool set for
    developing modern high-performance applications.ý Neil Herbert, Marks &
    Spencer
     
     
     ýWe were looking for stable, robust middleware with a powerful
    comfortable development environment ý all from one source. The Microsoft
    .NET architecture convinced us immediately. An additional factor was the
    negative experience we had had with Java/Oracle in creating HVNEU Online (a
    property management service), which meant that it was just not an
    alternative.ý Gert Liebsch, IT Manager
     
     Scandinavian Airlines System
     -----------------------------
     SAS gives customers wireless access to flight info with .NET
     
     -Number one business travel airline in Europe
     -Flew 23 million passengers to 92 destinations in 2000
     -Over 343,400 flights each year
     -Providing wireless access to flight info, rebooking
     -Available on cell phones, PDAs via XML Web services
     -Easily extended their Self-Service Booking System
     -Chose .NET for time-to-market, easy maintenance
     -Implemented support for all devices in one week
     -Easily add new devices as they become available
     
     ýVisual Studio .NET, along with the Mobile Internet Toolkit, allowed us
    to deliver customized pages for myriad devices quickly and
    cost-effectively.ý
       Peter Muller, Deputy Director, IT Group
     
     Zagat.com
     ---------
     XML Web services built on .NET create new opportunities for #1 restaurant
    guide
     
     -XML Web services give Zagat customers instant data access from any
    device
     -New content management system built on .NET
     -Integrates Web site with traditional publishing tools
     -Creates new business opportunities such as customized guides printed on
    demand
     -Turn around much faster with .NET Framework, VS .NET
     -.NET delivers superior TCO
     -Zagat redeploying IT resources from troubleshooting data issues to
    driving new revenue opportunities
     -$500K hardware savings over Sun solution
     
     "With Visual Studio.NET and the .NET Framework, weýve been able do more
    with less programming effort than we ever dreamed ofýitýs almost as if
    Microsoft knew the specifics of the application that weýre developing. Our
    developers are at least twice as productive as they were before.ý
     Steve Forte, Zagat.com CTO
     
     
     Cafepress
     ---------
     Dot Com switches from Java; .NET boosts performance and productivity
     
     -Cafepress provides outsourced ecommerce solution
     -200 thousand storefronts, 50 million hits a month
     -Initial Java implementation too complex, poor performance
     -Evaluated WebLogic, Orion (now Oracle) and .NET
     -.NET was the ýwild cardý ý were sure they would keep Java
     -Rebuilt entire solution on .NET Framework
     -100 times the performance, one fifth the code
     -Opened new business opportunity
     -Custom merchandise enabled offering by .NET
     -Now the fastest growing business (400% YoY)
     
     "The performance increase after moving over to the .NET Framework was
    unbelievable. The machines used to run at about 50-70% utilization and now
    run at about 2-3%. I have never heard of such an increase in performance.ý
     Fred Durham, Chief Executive Officer CafePress.com
     
     
     Merrill Lynch
     -------------
     
     -Business Challenges
     -One number for customers to call
     -No ýdead-airý or ýtemporarily unavailableý
     -Technical Challenges
     -Create a single call-center system
     -Consolidate legacy systems
     -24x7x365 operation, complete fault tolerance
     -Minimum delay to customer input
     -Minimum implementation dependence
     -Flexible configuration
     -Solution
     -.NET XML Web Services backend with host integration
     -XML mobile client to PDAs/cell phones
     -Handles 75 million transactions a day
  118. JRoss[ Go to top ]

    Let me guess, your name is "Ross Ross". If you are not even honest enough to sign your own name, why would anyone bother to believe a word you say? Particularly when it is cut & paste marketing from Microsoft? What is your name? Who is your employer? What is your background? Why are you a zealot?

    As far as lists of "referenceable" accounts go, I've already debunked at least one of those with Greg Leake. How many times are you guys going to list Newport News? It sounds more hollow than Apple touting PhotoShop! Maybe list the one ML app again, or talk about Nasdaq ... it's been at least a week since those were mentioned.

    One application at the "fifth largest airline in the US" and one at a minor web site that already used IIS/ASP doesn't make a market success. No matter how many times you cut & paste it.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  119. JRoss[ Go to top ]

    Well, its pretty clear who you are Cameron - heh, heh.

    Judging by your responses and serious 'Java-bigot' tone - you are quite obviously a sponsor of this website or no one would tolerate your insolence (i.e. you must be paying this site for the right to speak - certainly not the other way around, he he).
  120. insolence[ Go to top ]

    Dear Me L. Money,

    MLM: "Judging by your responses and serious 'Java-bigot' tone"

    Bigot? Nice try. I've spent a ton of time learning and working with .NET. Just because I think it's stupid that Microsoft built .NET in the first place doesn't make me a bigot. It means I have an opinion, and enough courage to voice it and stand behind it.

    Quick questions for you, Mr. Courageous:
    1. What is your name?
    2. Who is your employer?
    3. Why do you hide behind an alias?
    4. What experience do you have with J2EE software? .NET?

    MLM: "you are quite obviously a sponsor of this website"

    No, not yet, (although they do run on our Coherence software), but I think you have a great idea. I'll consider sponsoring it now.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  121. They Call Me ... Camoron[ Go to top ]

    <camoron>
    Do you realize how ridiculous your assertion is? Do you know what Telcos even do? Do you understand that they like their systems to run and keep running? Rebooting things is not an option. I've yet to see Windows in the core operations section of a Telco.
    </camoron>

    He he, your funny (or mildly amusing anyway).

    Yes, I do understand what Telcos do - I have worked in 3 of them over the last 4 years.

    You've yet to see a Windows server running their core systems!?! Ha! I've yet to see any thing other than mainframes in their 'core' systems. If you are going to amusingly suggest that they use unix or linux or any such IBM-compatible nonsense, I am laughing in advance!

    However, none of them like using mainframes - they're just do damn expensive to replace (and too much work), and they can't afford ANY downtime.

    To that end, ALL of their business applications that humans need to interact with (whether that is their customers or their employees) use distributed applications cause they are fast to build, easy to maintain, and give immediate ROI.

    Integration with the mainframes is still a small part - but EAI solutions solve those little pickles.

    So is .net and COM+ and some Java being used for those distributed business applications - yup.

    And besides, smarty, how much new money is being spent on those mainframes vs. how much is being spent on the business applications ... hmmm, I wonder?
  122. OS[ Go to top ]

    Well I've worked in telcos as well, and suprisingly enough ( not ) they are mostly using UNIX based systems. With business logic usually in middleware ( yuk ) like tibco/IM.
  123. Off topic - but Coke actually has a MUCH larger market share than Pepsi. Pepsi is actually small compared to Coke.
  124. Well, its not completely off topic - when you think of the parallel...

    You said Coke has MUCH more market share than Pepsi.
    In North America they aren't that far off (43.7 for Coke and 33.2 for Pepsi last time I looked), but Coke does have a mazssive market in Europe and overseas too.

    Think perception though, and you fall right inline with the Java/.Net debate. We always perceive them as being major players (even if one secretly gains more ground day by day).

    And no matter how much more share one has over the other - neitehr of them will disappear unless they do something stupid.

    Coke AND Pepsi will be around forever.
    Java and .Net will be around forever (forever in terms of IT which may be 5 years now, never 10 like the old days cause things change too fast).

    Netscape was an example of doing something stupid. I thought we would always have the crap dealing with both IE and Netscape - but Netscape caused their onw downfall and IE DID tromp them (not usual in IT).
  125. .Net not converting Java developers[ Go to top ]

    "The only JVM that was ever worth a damn was the Microsoft JVM. Most Java evangelist accept the fact that the MS JVM was superior to equivalent Sun JVM of the time. "

    The problem I see in the whole JVM debate is that you are considering WinTel as the only computer system existing in this world.

    You want a native JVM with Swing running in hardware accellerated OpenGL mode? Take a look at Jaguar (i.e. Mac OS X 10.2, which has a nice official version of MS-Office, Explorer and MediaPlayer running on an open source BSD implementation, I am waiting for hackers to port it to linux).

    Remember that Sun writes a JVM for Win32 without having any control on the OS and without being able to integrate it, as it should be, with the OS at a deep level (take threads).

    The fact that .class files are not compiled would be the beauty of it all, as a tailored made JVM for Win32 built from Microsoft would give you something as fast as anything else that's running on the machine. While the .class would be runnable in all other JVM's as well.

    The fact that Java and C are even just comparable in performance tells you that Java is the best choice since it makes things cleaner and easier for everyone involved with source code.
    It's a better level of abstraction, with very little, sometimes none, price to pay.

    JSP's are slow? Who cares, go out and buy a new faster Intel CPU (unless the problem is in the developer not knowing the runtime model of jsp's and servlets, or maybe those resources wasted running a GUI on a server - I will never get it). Faster CPU doesn't do it? Buy a nice workstation from Sun or maybe an HP-UX server or something from IBM with any hardware configuration you'd like to have and simply copy your class files (if you really need to be some milliseconds faster, you'll have the money to pay for it, right?).
    The Win only world doesn't get the concept of OS as infrastructure, not as requirement.
    It's about best man for the job, with Java you choose, with .Net MS chooses for you.
  126. .Net not converting Java developers[ Go to top ]

    Tom Cole: JSP slower than ASP/JSP interpreted

    Hah... I know I'm already late and this has been handled but please get the facts straight before making such trolling false statements....

    jsp page is compiled to an actual Java Servlet-page on first time it's being run, then compiled to bytecode. In effect, yes, the first time you request a jsp-page it will take some time (Unless you set it up to be loaded on server init) - following requests are going to at least compiled, often cached code which is equal to servlet in manners of execution speed.

    As for benchmarking - I've noticed the results tend to change depending on what gets done - how large tasks, whether or not database access is done, and how, etc etc etc - in practise I've run simple tests: clocked a single .jsp versus .asp or .aspx page with simulated loads of hundred concurrent users on a single windows box.

    Guess what? I could see no difference. Yes you can get differences if you clock precisely enough, and if you define the tasks better, if you compete architecture against architecture. But for me it's important to know that they all work equally well to a certain amount on single box - and can be scaled to larger loads.

    Old ASP should have been slower of these three considering it's interpreted, but difference can only been seen on larger systems with heavier loads. On architectural standpoint JSP/servlet is being run on Java VM, as compiled code, and ASP .NET is being run on CLR, as compiled code - so there is no difference on that level.

    My two cents. Keep on entertaining conversation but please get the facts straight.
  127. The community is too stuck up with .NET vs J2EE rivalry when both are complementary. It is just the way MS markets and perception of the rest that has lead to wasteful comparison. Read this article to know how the software cold war era can be demolished and how <b>only best may survive and that might include .NET but not necessarily Microsoft.</b>

    Check out the full article at:
    Future of Computing
  128. Reason none of the Java will ever convert to .Not is that .Not for the evil empire is still and always be a hype and it is geared for monopoly.

    MS == evil_empire
  129. "MS == evil_empire"

     here we go again... an unprofessional attitude. Everybody would appreciate if you gave technical reasons for your "no microsoft" choice. Stop behaving like a neonderthal.
  130. .Net not converting Java developers[ Go to top ]

    No wonder the neonderthal's are extinct. Very difficult to hide from preditors/enemies when you are all lit up and flashing in hot pink - "Here I am" ... "Here I am".
  131. Whenever I see the word "Gartner" in an article, I immediately switch off. Gartner can be pretty much assured to say anything, even if it wildly contradicts what they said last week. It's just a question of who has their ear this week. Marcus J. Ranum put it best, in a post to the firewall-wizards mailing-list:

    "You've got to understand that most of the input into Gartner is from briefings arranged by the marketing departments of companies that are paying them to listen to their briefings. Basically, Garter sits at the apex of the hype food-chain; they consume pure hype and produce little shit-pellets of hype that is[sic] as dense as neutronium."
  132. MS the evil empire is at it again there is article in Infoworld

    that MS has been very jumpy as of the late bacause of all the innovation on Macromedia's part So much so in fact that the MS just wants to buy them perchance to avoid any competition in the desktop and developer frays where MS plays.

    We know what MS wants to do now with .Not hmmm
  133. .Net not converting Java developers[ Go to top ]

    Yea...buddy....Macromedia is a real force. Jrun is a contender to take out .NET!

    Please.
  134. .Net not converting Java developers[ Go to top ]

    Tom: "Yea...buddy....Macromedia is a real force. Jrun is a contender to take out .NET!"

    Look, the "Microsoft is perfect" nonsense is as tiring as the flip-side "Microsoft is all sh*t" tirades. Get off your high horse and try to stop insulting everything that doesn't come out of Microsoft. It just makes you look like a cheap shrill. Microsoft makes some good products; .NET as a technology is fine, but it's too bad it's not Java compatible, and it would have been nice if they had a better enterprise integration strategy than "you can use web services".

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  135. Cameron: ".NET as a technology is fine, but it's too bad it's not Java compatible".

    But it is frustrating to me when I (-want-to-) switch from WLS 6.1 SP3 to WLS 7.0 SP1, i.e. I am not even talking about switching to IBM Websphere 4.0 or JRun.

    It took me half a day to install WLS 7.0 and figure why the hell they changed the deployment model- for better or worse.

    We are a professional J2ee shop. Frankly, our switch from J2doubleE to Microsoft is mainly becuase our end customer want the solution/product yesterday and they care least if the software is java compatible or not. The custome just wants it yesterday- that's it! And it should run as fast as Excel, tell me can you do that in swing! (There is a pun intended here when folks say you need CPU upgrade to run togetherJ ;-))

    We use Excel as front end, because the supply chain client world just loves it, and Microsoft is king there. We use SQL server as backend since it is the only database that supports bulk insert of huge flat files. The best data-warehousing tools run on SQL Server, and our users have started liking including MS's office web components techn ology compared to the earlier html tables. And now for SAP connectivity we are switching from Java to .NET andd/or C++ based "fast, asynchronous and parallel" RFC connectivity.

    So, don't blame Microsoft, especially in apps area, they are the kind, and they will remain that way if J2EE app server vendors still do not come to a unified deployment model.

    One more thing about the total cost of ownership- why should the customer pay 10 grads per CPU to BEA/IBM, if JBOSS tech supp is better and the cost upfront is none!

    Cameron, I liked your approach this time, although I always suspected your knowledge on the Microsoft front despite your MVP claims. I still firmly beleive that you have never used Microsoft technologies beyond your MVP apprenticeship period!
  136. Ravi: "Cameron, I liked your approach this time, although I always suspected your knowledge on the Microsoft front despite your MVP claims. I still firmly beleive that you have never used Microsoft technologies beyond your MVP apprenticeship period!"

    I worked with Windows development since Windows 2.0. I remember "bozoslivehere", although I was never focused on GUI dev. I worked in ASM/C/C++ starting with 16-bit (real and protected modes) then 32-bit using Windows API. Then COM/OLE2/DCOM, both 16- and 32-bit. No MFC either ... among other projects, I worked on building COM interfaces (machine code) on the fly in memory and then loading it as executable (allocdstocsalias undoc'd API). Ugly stuff.

    It was quite a relief for me when Java became available. It must be equally a relief for developers stuck on Windows to have .NET, but it's a shame that Microsoft rolled their own incompatible product when the rest of the industry was already working together with Java. As I've said dozens of times, it means that developers are forced (for no good business reason) to choose between and/or integrate two technologies, which seems terribly pointless. For software companies, it means porting and maintaining two code-lines if they want to support .NET. Frankly, it is disappointing.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  137. Use Java to create a .net application -:)
  138. Mr. Purdy: "...it's a shame that Microsoft rolled their own incompatible product when the rest of the industry was already working together with Java."

    But you still do not talk about any C++/ATL experience, let alone on the subject of cross deployment across diff j2doubleE servers or diff versions of a brand-name server e.g. weblogic 6.1 versus weblogic 7 dot 0...Do you have any practical insight into the evolution in that space Mr. Purdy?

    How do you then say that MS has stop catering the VC++ community? Most of us, who dream about J2doubleE being the "main" industry, should not forget that there are apps still being developed in plain old C/C++, VB/PB and COBOL. The J2EE folks cannot forget that there are LAMP (Linux-Apache-mySql-Perl) folks too that care even less about anything else!

    Mr. Purdy, none of us - neither you nor me - can define what the "rest of industry" is therefore, let the (stock) market decide.

    Let the J2EE folks define a unified deployment model amongst diff app server vendors with or without JCP. Maybe JBOSS folks will, at the rate they are gaining the market share from BEA/SUN/IBM.
    And then if you still can't then may I suggest that you folks learn about what are assemblies in .NET, and if that is too trivial, then come on and get on board with plain old C++/STL/ATL folks around such as myself!

    Regards,
    /Ravi
  139. Cameron: The developers are forced (for no good business reason) to choose between and/or integrate two technologies. For software companies, it means porting and maintaining two code-lines if they want to support .NET. Frankly, it is disappointing.

    Well, there are two distinct worlds- one that supports MSFT technologies, and the other that does not. Both have their pros and cons. Let's face it bravely. As to which one is more disappointing, and will there be more than those 2 worlds- we shall see.

    Please refer to the article to understand why "MSFT and the Justice Department filed briefs Monday 1/6/03, opposing an effort to appeal a federal judge's November approval of their antitrust settlement".
    http://money.cnn.com/services/tickerheadlines/for5/200301061428DOWJONESDJONLINE000421_FORTUNE5.htm
  140. I hate this whole debate which is better. Why? Because the big boss at the top just don't care. They have made huge investments into their IT over the last couple of years. Especially with the Y2k and stuff. Then they had to upgrade to Microsoft 2000. Which still is not 100% complete.... There are something like 200 servers in the server room and some of the hardware had to be replaced with the upgrades. And now they have to fork out even more $ with the new licensing...... Not even to talk about the cost of the new .Net Visual Studio license cost. Its is more than the average programmer earns per month in this country. Then I'm not even talking about the training. All the programmers have just finished all the VB6 courses. They have all applied for their .Net courses now. They have to since the average programmer in South Africa is to stupid to learn from a textbook. And there are many of them. Words like threaded programming are still a myth to most of them. And what makes me want to rub my wrists on the bathtubs edge is that the company just keeps on forking out the money. Yes.... the Boss in the penthouse don't care. He wants to see the end result. and he want to see it in the colors he knows. When he goes home his kid is playing games on the pc..... yes it runs the infamous Microsoft operating system. Then why should it not be used on the servers at work. you can afterall have up to 16 clients connect to the quake server. This way at least the big man can sound intelligent when talking to the nerd who feeds on pizza and drink soda with no social life. The big boss never thinks about which is faster or which had better memory handling. RAM to him is a verb meaning to put more in a can. Who cares if the server crash on a regular base. His son has already shown him the blue screen a couple of times. Doesn't all servers crash. And its so easy to fix.... just press the reboot button. better yet lest schedule all the microsoft server to reboot once a week.... that way we don't have too many memroy leaks. Yes our qualified MCSE's are in the same category as the programmers. The plus side is they are cheap and there are many. The fact that half of them (in south africa) can't even find the reboot button on a machine is irrelevant. All you need to know is how to use a mouse anyway. But I must say I love it.... I get to program using C#. The rest of the programmers don't have a clue what is going on in my code since they program in VB for .Net. So my code is nice and clean and untouched...... If I don't want to do something all I have to do is convince the boss that it's not possible or to time consuming in .net. He heard the line so many times before that he believes me in a snap. The System Analyst has finally decided to also apply for a .Net course. Must be because I kept on telling him that his nice UML class diagrams with bussiness classes and stuff does not quite fit into the .Net framework. He did not have a schema-thingy in it that creates a data what-you-call-it. So until he get back I get to surf and read Java vs .Net forums. Wonder how much the .Net 2003 will differ? I must say I feel a lot better now. When I started writting this message I only saw the bad of Microsoft. I get to write spaghetti code. If questioned I can refer my other programmers to the .Net Petstore implementation. I get to surf the net a lot since the MSDN documentation although better is still filled with some very classic statements. My boss gave me his credit card number so I can buy a new controls since the ones in the SDK is way to primitive. I'm getting a bigger and faster desktop pc since the current is to slow for the IDE. On hot days I get to sit in the server room. I get to annoy my boss by calling him at 3am. The Microsoft food at their seminars are always the best.
  141. ROTFL!

    "And now they have to fork out even more $ with the new licensing...... Not even to talk about the cost of the new .Net Visual Studio license cost."

    Our CIO handed down a dictate 2 weeks ago. No more Microsoft. Linux and Java are now the preferred technologies here. Period. Microsoft pissed him off so bad with how much the new licensing is going to cost him, he's flat-out rebelling.

    Is it feasible? I don't know, but I know we just rejected a vendor's product because it was a .Net application. He wouldn't even look at it. :)

    Since most folks over here can't spell "java," it's going to be interesting and entertaining.
  142. .Net not converting Java developers[ Go to top ]

    The reason Microsoft doesnt jump on the Java bandwagon is because Sun controls it. Make no mistake about that. Java is just as proprietary as Visual Basic, in that respect.

    As far as the CIO who says "no more Microsoft". He will pay for that decision 10-fold in lost productivity. His developers spending 40-50% of their time trying to make that crap simply "work", instead of focusing on the business problems at hand.
  143. .Net not converting Java developers[ Go to top ]

    Tom: "The reason Microsoft doesnt jump on the Java bandwagon is because Sun controls it."

    Perhaps, but perhaps the reason that Microsoft didn't stay on the Java bandwagon is because Microsoft couldn't control it.

    Tom: "As far as the CIO who says "no more Microsoft". He will pay for that decision 10-fold in lost productivity. His developers spending 40-50% of their time trying to make that crap simply "work", instead of focusing on the business problems at hand."

    That doesn't sound like a very objective statement. Actually, it sounds rather vindictive and poisonous. Do you have some stake in lost product sales at Microsoft? Are you a Microsoft employee? If not, why are you so vehement about a company, for which you do not work, choosing software that is not made by Microsoft?

    Regarding the ability of companies to successfully use software made by companies other than Microsoft, I would suggest that it is not quite the impossibility that you are alluding to. There were good software companies around before Microsoft, there are good software companies around now, and there will be good software companies around after Microsoft.

    While I choose to use Microsoft Windows and Microsoft Office and Microsoft Keyboard and Microsoft Mouse, I do so because I think they are good products and, despite the perpetually rising cost of the first two named products, I continue to use those products. If the costs of those products and the pain of doing business with Microsoft has exceeded the threshold for various companies, individuals, governments and countries, then Microsoft will start to lose customers and market share, and that will make room for new and innovative companies and solutions to emerge as viable options. That has already started to happen with Mac OSX, Linux, OpenOffice, ThinkFree, and other products. It happened six years ago with development tools and platforms (Java). It happens, because there will always be some smart, hard-working group of people out there somewhere that is willing to work harder and longer and be more accountable to and respective of their customers in order to win that business. And your derision of someone for not simply conforming to what some company tells you is the norm is shockingly Orwellian.

    I'm going to go finish my Coke now.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  144. "As far as the CIO who says "no more Microsoft". He will pay for that decision 10-fold in lost productivity. His developers spending 40-50% of their time trying to make that crap simply "work", instead of focusing on the business problems at hand."

    You make two assumptions here:

    1. The CIO will be keeping the same set of developers and expecting them to work with Java/Linux, etc. There are plenty of others who can.

    2. The developers are the ones focusing on the business problem.
  145. .Net not converting Java developers[ Go to top ]

    You have to be pretty stupid to share the fact that you are going to drink a coke with a discussion forum.

    Good for you, little man. See you in the funny papers!
  146. .Net not converting Java developers[ Go to top ]

    me: "And your derision of someone for not simply conforming to what some company tells you is the norm is shockingly Orwellian. ... I'm going to go finish my Coke now."

    Tom: "You have to be pretty stupid to share the fact that you are going to drink a coke with a discussion forum."

    Oh, the irony!
  147. .Net not converting Java developers[ Go to top ]

    "And your derision of someone for not simply conforming to what some company tells you is the norm is shockingly Orwellian. ... I'm going to go finish my Coke now."

    Tom: "You have to be pretty stupid to share the fact that you are going to drink a coke with a discussion forum."

    Cameron: "Oh, the irony!"

    Cameron, maybe next time you should include <joke> </joke> tags...
  148. .Net not converting Java developers[ Go to top ]

    Tom: "The reason Microsoft doesnÂ’t jump on the Java bandwagon is because Sun controls it."

    It's too late to lament now that it is two incompatible versions. That Sun is a competitor to Microsoft was not enough to make Microsoft jump off the Java. Sun was not only a competitor, but also the leader in the pack who lanced a massive anti-Microsoft campaign through the world full of half-truths, slander, innuendoes, arrogance, personal attacks (even on BG father!) and lots of down right lies. And the Java/Unix/Oracle world was not only applauding, but full of glee.

    So does anybody after this think it's strange that Microsoft decided to abound Java?

    Sun are only getting what they deserve. (Talk of sawing off the branch they were sitting on..) Imagine where Java would have been to day with all Microsoft resources behind it and not two incompatible versions!

    "Vanity goes before fall"

    "People with that attitude and such language can never be right"

    Try mono! http://www.go-mono.com

    Regards
    Rolf Tollerud
  149. I totally agree with your observation about the MS/Sun Java-battle.

    It’s like a pack of dogs trying to bring down a mammoth. They better not let him live, cause if he turns his attention to them – they are dead! The funny thing about a pack too, is that while they are together they are strong – but when they are alone they are little weaklings – especially against a mammoth.

    I do not envy Sun now. They drew too much attention to themselves and made the battle personal. Bill has the power to do them in if he sways their direction – and I think he has.

    Who will come to Sun's aid though I wonder? Will IBM save them (they have a serious Java investment and are Sun's primary competitor) - I think not.

    Should be interesting to watch.
  150. You are so right. They pissed Bill right off. They deserve it.

    Revenge of the Nerds!!!!
  151. Here is a new twist to the issue. In a perfect world Microsoft and Java both solve mostly the same problems. The approach is a might differ. With Microsoft you have an easy to use environment but you pay a lot of money for this environment. But a idiot can use it. With Java the environment is cheap but you pay more for the developer. He has to be a little more technically minded. So typically your project resouce costs will be dived something like this for Java:

    Hardware 20%
    Software 20%
    Developers 60%

    For the same project using microsoft the final cost will probably be the same but the breakup will be

    Hardware 20%
    Software 60%
    Developers 20%

    Java developers are very expensive in South Africa but they are usually a lot better than the Microsoft developer. Our market is flooded with Microsoft ASP scriplet programmers that does not even know what a function is. This has caused the hourly rate of the average VB developer to drop drastically. Its about R130 per hour. The market sees and understand the Java developer is usually more technically minded and pays for this. The hourly rate for the Java developer is around R250 per hour.

    From a management point of view the decision is easy. Either spend money on software or spend money on people. Which company do you want to work for? Managers looking at risk will chose to spend more money on software. What if the ppl resign? Software don't resign. In the beginning software was a once of payment but with the new licensing schema of Microsoft..... Interesting questions you have to wonder about.
  152. For me, working on mission-critical enterprise systems, it boils down to two issues:

    1. Which technology is established? We don't want to be guinea pigs.

    2. Which is the most platform-independent and will allow me the greatesr flexibility in the future?

    For me, I'd rather have the experienced engineers on those projects. They choose Java. I advise management at my company to trust them. End of story.
  153. Well not sure if any of u ppl saw this but check out these ratings.

    http://www.bitbreather.com/programming_languages.html
    http://www.tiobe.com/tpci.htm

    The reports are compiled on two different methods. The first look at the Jobs availibilty... Yeah, Java wins. The second look at popularity of the language. Well check out the methods of how the reports are compiled.
  154. I don't care[ Go to top ]

    I don't care about
    * Windows vs Linux
    * .NET vs Java
    * That vs this techonology

    I would make rather see the professionalism in the IT industry improve. OOA/D, Design patterns, Test-first development, etc are much more important then any technology. Basically it boils down to getting the job done for the least cost at the highest quality. Choose the right tool for the right job; none of this one-size fits all debates.

    The biggest problem in this industry is communication, in particular requirements analysis. Check out the first lecture from MIT at http://ocw.mit.edu/6/6.170/f01/lecture-notes/index.html

    Learn concepts, not technologies. That is my wish to all on this thread.

    Software Engineer,
    Cameron Zemek
  155. I don't care either![ Go to top ]

    Cameron Zemek: would make rather see the professionalism in the IT industry improve. OOA/D, Design patterns, Test-first development, etc are much more important then any technology.

    Cameron, the TSS folks have "The J2EE Community" behind, and so they are not going to listen to individuals like you and me who care more about professionalism and less about thie-is-better-than-that.

    I find many of these threads less appealing now, since there is no moderation in message boards like these! Looks like TSS's purpose is to hook the community folks on biases rather.

    regards,
    /Ravi
  156. I just started using .NET[ Go to top ]

    I am quite Dissapointed in it. I expected a Java type Event system. You know, you can have several buttons and then a method that serves for each to call, passing this method an instance of an event where you can discern which button was pressed by checking the value of the event...This was not the case. I have to create a different method in the Form of all places to handle each button that is pressed. In addition, I can write directly to things using member variables; I don't need get/set methods like I thought I would. The get/set thing isnt too big a deal, it always seemed a little overkill, but the lack of a centralized event handler is pathetic. This is a step back not forward. If I am seeing problems this early on...what is going to happen if I delve into more complex projects?
  157. Whats the problem?[ Go to top ]

    <Michael Rasmussen>
    but the lack of a centralized event handler is pathetic.
    </Michael Rasmussen >

    Whats the problem here?.. Change the event handler to point to your centralized event handler.
  158. who wants to supports ?[ Go to top ]

    Look at the top enginnering schools from top 5.

    MIT <-- stallman and bernes lee from there
    Stanford
    Carnegie Mellon
    UC Berkeley <-- TCP/IP, BSD folks there
    UIUC <-- mark anderson(netscape), apache http is from there.

    Mostly, they are willing to support unix/linux friendly system environment.

    okay, i admitt it, .NET has some potential market. While i was in the college, we only had small lab for NT machine and huge several Solaris labs all around building. Now, i 've heard they are moving to linux-based system half.

    Since Microsoft lacks of educational power, the only recognizable engineering university they created was i guess univ of washington in seattle feeding with money. I remember MS guys try to advertise .NET with free pizzas and xbox during the information conference. i forgot the name of the speaker, steve something(later, i noticed that guy was some famous figure in the microsoft).
    he was redden as peach after his presentation because of so many opposing comments about .net. guess what. No Sig for microsoft in my college while SigUnix and SigLinux is flourishing.

    I know it could be just for the naive education institute not for the industry. But one cannot underestimate why the education institutes tend to support open-friendly-environment. it is definately related with the market, too in the long term.
    These days, i see young bulls (potentially can be a leader of the industry) are armed with linux and lots of opensources. Where many guys loved to use dos-game in win32.
  159. Still.....[ Go to top ]

    This is a real story.

    I went to a gas store for a gas cylinder.
    I asked the keeper for this. He asked me for my name.
    I said "Sanjeev" ... He started writing it...
    He wrote s a n ..... then he asked me ... is it K now ??
    No, I said it is J.. Then he said.. "Oh ! Sir, u means J ... J for Java "

    Ouch !!! Everybody knows java. I asked him, what he knows about Java.
    He said that he frequently listens about java. People talk about java everywhere.

    Every where is J.... who the hell this N.

    J means Java and N for Nothing.

    We luv Java and We all want Java.

    Whenever I talk about Java, I start feeling hot hot hot...
    so posting it now...
    Sanjeev Dhiman