Discussions

News: Meta Group on New App Market Share: 40% J2EE, 30% .NET by 2004

  1. The Meta Group has recenly concluded that by 2004, Microsoft will have approximately 30 percent of the new enterprise application market, with Java stabilizing at 40 percent and the remaining ones using existing technologies (legacy, CORBA, etc.). By 2005-06, there will be many alternative development frameworks and programming models built atop these two basic infrastructures (Java and .Net).

    Read .Net seen gaining steam in dev projects.

    Threaded Messages (51)

  2. What bullshit !

    These "authorities" on technology don't even know what's going to happen tommorow never mind coming up with actual percentage of mindshare that each will have by 2004.

    I've worked with both and it will be a long time before Java plays catch up with .Net as the article suggests.

    Visual Studio .Net is a "feel good" IDE, it hardly has anything to useful to offer in the real enterprise market, note I used the word "real".
    Many developers who love VS.Net believe themselves to be enterprise developers (especially ex VB hackers), this is just another tool to help in the self-delusion they enjoy so much. There are exceptions.

    Wake up guys !

    There may not even be a Microsoft as we know it by next year depending on the outcome of the current anittrust case.

    p.s. Did anyone notice the huge Microsoft flash ad in the middle of the article ?
  3. "During 2002, the .Net framework will be substantially fleshed out and become a reasonable alternative to Java development. " huh? Become a reasonable alternative? There's more perception than reality here.
  4. Unfortunately, perception IS reality. MS .NET will come... Even if we don't want it. Technicaly, it is just a new programming language, but almost none of the existing issues are resolved.
    Companies that are today using MS C++ and/or MS VB are encouraged by MS to say they are .NETting. And people believe it. Unfortunately. Perception IS reality... MS know what marketing is all about.
  5. Which is why the truth must be preached. Not propaganda but the truth. The FUD (i.e. J2EE is difficult, Java has a big learning curve) has to be dispelled. I just love the IBM commercials. For those moving from VB/ASP to VS.Net and really trying to do OOP, the learning curve is great. For those doing the same old thing the curve is minimal (i.e - Where did this menu item go?). ASP scripts can be run as is in .Net. The Java learning curve can be minimal too.

    BTW. .Net is here now.

    Mark
  6. I don't know how will this happen. .NET is still in its infancy whereas J2EE is going from stength to strength. It has taken quite some years for J2EE to reach the positon where it finds itself today and so many vendors have contributed towards this goal. How can one expect .NET to achieve nearly the same level of market share in nearly half the time. Moreover, .NET is and and will remain a Windows platform technology, we won't even be seeing enterprise level deployments where Unix is the king. Not only that, the kind of collective wisdom that has gone into the making of J2EE, is simply missing from Microsoft world. I think .NET is still going to remain a marketing gimmic for quite some years to come, and by the time it achieves some maturity, J2EE would have been embraced by majority. In nutshell, .NET is destined to find its place in trash.
  7. I dont think so.

    MS is not selling .NET .. not from their published Quarterly results. I doubt people in this tight economic market will want to invest anything "enterprise-wide" on an immature technology then invest in a mature technology like J2EE. No one deploys Enterprise Applications on Windows machines - its like sitting on a rotten Egg!
  8. Hi,

    To an extent I agree with them fellas about .NET acceptability. It's like succumbing to the dark side. It's easy, once you get over the initial guilt!

    So what if J2EE will let you build great applications? Why would I want to expend time and effort writing solid, portable code that will provide great ROI to all my clients when I can get by with much less?

    Now take .NET, I could have been a hobo and learn .NET in a couple of months! Folks, its so simple and easy to use, designed just for those who don't really want to spend time understanding the long-term implications of their code (Good God, you want me to try and understand THAT?!)

    Why would I sweat it out to write really great programs when I could easily patch together little bits of spaghetti code that that I could sneak past the eyes of the unoriented in the guise of quality software in half the time and with half the effort (Phew, a mouthful!)?

    Of course, my system might have a couple of dozen or so critical security bugs. But SO WHAT?

    Sandeep Dath

    PS: Don't you really hate it when these so called "research-oriented firms" publish this sort of stuff, misguiding all the many, many already misguided CIOs out there?
  9. dear fellas,
    those who say that they want to get things done with out putting efforts then imagine the future where there would be no use of you (as a developer )
    people who believe on MS will be the first to be jobless
    ;-P

  10. THIS IS FAR FROM REALITY :)

    The things i can see that this may happen are:
    1. MS offers .NET for free :)
    2. Sun does not live up to the expectation of delivering the
         new J2EE and incorporating rich web services support.
    3. IBM and other Sun allies abandons java and goes for .NET
         :)

    Which as i see it the only possibility is #2.

    Ill give it a 60% - 10% market share.
    The possibility of a 40-30 is definitely lower than 10%.
    Have i been acting like the those "business/tech analysts"? :)
  11. It seems these tech analysts' predictions have the same credibility and accuracy as those of the stock market analysts.
  12. If .NET is banking heavily on the success of webservices, then the prediction of 30% .NET adoption is extremely optimistic.

    (1) Recent surveys show a minority of webservices being deployed using .NET .
    (2) Webservices is based on the idea of providing better interoperability. Unfortunately, syntax without semantics is a completely vacuous argument.

    If .NET is banking on a perception that microsoft programmers are less costly. Sure, maybe VB6 programmers will continue to be cheap, unfortunately supply and demand rules. Java benefited from many experienced C++ programmers adopting the language. I doubt you'll have many Java programmers migrating to C#, the value add doesn't justify the effort. So in the end you'll have many inexperienced .NET developers,
    and like they say "you get what you pay for".

  13. Well this study from Evan Data regardless of accuracy will definitely make some java folks jump up and down
    http://www.infoworld.com/articles/hn/xml/02/04/24/020424hnevans.xml

    "Microsoft's Visual.Studio .Net is the most popular toolkit in use by developers working on Web services, according to the study, with 64 percent of developers preferring it over rival toolkits from Sun Microsystems Inc., IBM Corp., Oracle Corp., and BEA Systems Inc., among others."

    Whether these studies are accurate or they have correct predictions, I like to make a few comments:

    1. .NET is here to stay that's what I believe, regardless of what people are going to say it's unproven, not scalable etc.... My take is that developers should look at what MS has to offer on the table. If it's doing a better job than Java platform then it's time to reflect and push for the improvement in the java community. If there are weakness in MS platform then keep that in mind and it can be used as a good point to hightlight the advantage of the J2EE platform. Ranting is the worst form of fight and I would not hesitate to say he belongs to the loser category and he deserves to lose the fight eventually.

    2. Self-delusion, without any in-depth understanding of what MS has to offer and believing that java platform is always superior will not help you in the fight. If I am MS I welcome this attitude because I know these people will not spend time analyzing my strengths and weaknesses, I will beat all these folks eventually.
    Sun-Zi said in "The Art of War" :
    " Know your enemy, know yourself, you will win every battles"
    " Know yourself, know nothing about the enemy, you can win and you can lose too"
    "Know nothing about yourself, know nothing about your enemy, you lose all the time".

    3. Don't waste too much time thinking about what are the strenghts that J2EE platform are beating MS now. Think about what are the weaknesses that can be used as an attack on the J2EE platform. Improve on that. Of course you can rebutt what is misinformation but don't spend too much time on that.

    4. Don't be too happy that J2EE platform are leading now,
    We learn from History that people don't learn from history. Remember Netscape (Don't tell me about how MS used their position to beat Netscape, the fact is that they won in the browser race right? period. Born loser finds excuse for the lost battle, real fighter analyze the reasons and prepare for the next one.)

    5. Never underestimate your opponents. Ridicules and verbal abuses may make you feel better now but it will never stop them from improving their products and advance in their attacks. Of course I am not saying you can't do that, what I am saying is that after the laughs and ridicules, be prepared for the fights :-).

    6. I suspect some of the developer here are platform zealots similar to religion. If they are religious zealots, emotion always rules over rationality and it is simply a waste of time talking to these people because right in their heart, they cannot be wrong because it will tantamount to be dead penalty of their believf if it is proven wrong. They have spent their entire time on it and believe it's the saviour of the day for all computing problems. I hope I would not receive a lot of verbal abuses after I have made this statement. (By then I will find out who are those people :-) ) I am partially platform zealots too but I do listen to what people say :-)

    7. Look at the positive side, if the world is so homogeneous, it's will be a boring place. Look at client desktop, it's rule by Winows, I welcome the competition from OS/X and Linux which will keep MS awakes at night :-).

    8. Take no side which I called sitting on the fench. Take the opportunity to learn what both platforms have to offer. Know the strenghts and weaknesses of both platforms and implement what is the best solution for your customers. This is more difficult than most people think because both platfotms have tonnes of information to be absorbed by the developers :-( . I am more along these line of thoughts :-).

    I believe a lot of what I say can be hotly debated. I think at the end of the day we should not lose sight of what we want to acheive, we want to advance our platform (be it J2EE , in these forum for majority of the people, or .NET) to the next stage of evolution to meet the computing problems at hand or the new challenges ahead of us.

    A past success does not warrant a future success. A future success will be dependent on the way we handle our current success, our reactions to challanges, competitions and plot the next course of actions.


    regards,
    Hun Boon
    From Singapore.
  14. Thank you Hun! I agree with you 100%.

    It seems to me that the BS ratio of the posts on this site has grown significantly lately. Look at the JOnAS vs JBoss discussion. BS from start to end until Rickard stopped it.

    Rage against the machine said: "Know your enemy!"

    So the message to all the J2EE fanatics out there is "Learn .NET". Then in stead of ranting, you can contribute to the J2EE world.

    Aslak
  15. Hun,

    This is the wisest post I have read in a long time. That's exactly what I advocate and practice everyday at my own level. I recommend everyone to read your article.

    Common sense.

    Yann
  16. J2EE 90% .NET 10%[ Go to top ]

    Don't even think about enterprise solution on .NET (Microsoft platforms)

    But J2EE is really matured technology and can provide secure stuff with Linux
  17. I am there with you Hun :-)
    I luv java & i like to see it in the market. At the same time i hate M$ monopoly !!!

    Just by screaming or shouting that J2EE will rule the world won't make it happen. I don't want to see only J2EE in the market. Only competition makes product perfect.

    Long live competition & long live J2EE & .Net & (PHP,etc );)



  18. Remember that these research groups are there to make money. It suits them if the marketplace has many competing technologies. This way, they can justify doing comparisons and then sell the data to brain-dead CIO's whose previous job was in HR. The CIO's see that they should hedge their technology bets which improves the bottom line of all technology vendors and keeps the analysts in a job. Its all just one vicious cycle.

    An analyst only analyses because they couldn't do the job that they are analysing (a bit like Management Consultants) ... so take it all with a grain of salt.

  19. I hate to point out the obvious but... there seem to be only 2000 organizations among the "Global 2000" :-) So this prediction cannot be applied as such to the multitude of somewhat smaller companies that are not even part of META's target group. Out of curiosity, does anyone know what the research "methodology" was ? Interviews ? How many companies were involved ? What were the questions ?

    Alain.
  20. Caution please, caution[ Go to top ]


    Once, looking for browser statistics I found an old prediction about browser war future of a very important trends consultancy. Netscape 4 and MSIE 4 were released nearly.

    This study predicted that Netscape 4 will stay ahead around 80% of market share and MSIE the rest, two years later.

    Two years later the percentage was the inverse.

    Conclusion: considere only the reasons about predictions, perhaps they success, not numbers, because they are invented as you and I can invent.

    .Net will have a percentage of market, there is no doubt.

    .Net will beat Java? It is not sure: most of industry is around Java except Microsoft and periphery.

    .Net will if Java/J2EE market is broken technologically: people like have options, select the tool they like, but they don´t like technological taifa kingdoms, in this scenario people select the monopoly option to stay sure they are made the best selection.

    - Sun must support JBoss (certificating unless, or set as reference implementation) to ensure there is no money barriers to enter J2EE world.

    - No J2EE royalties: free implementation, free standard compliant tests and win the best.

    - Only one and opened standard to every subject, not competing or closed standards (Sun please enhance Swing with IBM, IBM please don´t break the market with SWT)

    - You can make money with Java? sell standard based products and consultancy, develop a new compeling and closed technology, make money with it, but give it to the Industry as open source a year or two later (and improve it closed... and again ...) . An example: Cold Fusion was a very compeling technology, ASP beat it because is free, but ASP can´t beat PHP, if Allaire would have open sourced or given for free, perhaps ASP Microsoft server monopoly Windows-centric wouldn't be the same. BEA please give the Cajun for free next year (or two).


  21. The thing about .NET[ Go to top ]

    The thing about .NET is, and why it will have at least a respectable market share in the coming years, is due partly to the rebranding of .NET and partly due to developer's simply learning the new features of the languages they already know. What is the current percentage of VB / VC++ / ASP developers in the marketplace today? There is your current market share of .NET programmers. In two or three years, there will be at least this many developers.

    If we should learn any lessons from Microsoft, it is that they are very good are marketing and they make good user interfaces & developer tools. For many people, this is all that matters. It doesn't matter how much better J2EE is, if you can develop even a crappy application in half the time that it takes to develop a J2EE app, believe me, there are going to be a lot of crappy applications out there.

    J2EE definitely is a much more robust platform, but for managers to accept it (preach all you want about real developers using vi and a JDK), it needs to be easier to use. I am not just talking about developer tools (they are getting better). I am talking about management tools for the server platform and applications as well.

    For my part, I hope that on those points, .NET really gives the Java community something to compete with. Then things will get better. People who just out of hand dismiss the fact that .NET is here to stay are moronic. It's better to deal with reality and make what you do better.
  22. RIGHT ![ Go to top ]

    you are right.. but just a small detail Java is NOT better, it is just a communotar illusion

    what is better depend on peoples

    for exemple everybody will say Adobe Photoshop is from far teh best, whe for me Corel Photopaint 10 is from far better and 1000 times more practicle, subjectivity, ... only subjectivity !

    Many don't like Microsoft and I too, but .NET platfrom is incredible good
  23. Microsoft has lost the Mobile battle[ Go to top ]

    Microsoft lost the Mobile Battle. Estimations predict Microsoft as a clear winner, good product, good position. But now we know that Microsoft has nothing to do in the mobile arena.

    The same could happen with appserver arena to Microsoft.

    saludos
  24. There is something we are overlooking. Microsoft has a rather large tie-in with business users(e.g. iexplorer, outlook, word, excel, etc.). You can expect their thin clients to have HEAVY interoperability with these applications that ONLY work 65% of the time. To make a long story short, we need more applications developed by any other company rather than Microsoft to be used by the business community to continue our growth and not be stunted by Microsoft.



  25. Ok,

    I am still trying to figure out if these analysts are trying to tell me to build my middle tier with COM+ and MTS.

    Blinded by VB.NET I think many anal-ysts forget the importance of the middle tier.

    Greg
  26. There is a lot of confusion about the capabilities of .NET and I am afraid that MS is winning the battle. Somehow when discussing .NET components take a back seat!!

    I was invited by a small B2B company to help them design their software; 100% of their staff works with asp scripts. When I tried explaining the importance of frameworks and design patterns they looked at me as if I was an academic who had lost touch with reality. They are 100% convinced that .NET will solve all their problems. They will continue developing the way they do and the way MS wants them to.

    The biggest hindrance to J2EE adoption is the steep learning curve. If you come to think about it, how many developers can design a truly Object Oriented System? MS has a very big advantage not in terms of technology but in terms to having a large developer base (mostly VB developers) and these folks cannot comprehend complex technologies. So, MS will keep these folks happy and will continue enjoying a major market share. I have gone through a lot of MS literature and have never read any thing about OR mapping or MTS component design patterns?? They even have stored procedures as a part of .NET strategy, so as long as all MS products can seamlessly interface and are irreplaceable they have met their goals and leave the rest to the marketing team with their 1 degree of separation ad’s.
  27. <quote>how many developers can design a truly Object Oriented System?</quote>

    Exactly

    <quote>
    MS has a very big advantage not in terms of technology but in terms to having a large developer base (mostly VB developers) and these folks cannot comprehend complex technologies.
    </quote>

    Again, I would hesitate to lump them all as developers.

    <quote>
    When I tried explaining the importance of frameworks and design patterns they looked at me as if I was an academic who had lost touch with reality.
    </quote>

    The best way you shut them up is to talk ROI - full life cycle cost of the product. Present scenarios where their product will not work or will be *very* expensive to develop. Again ROI. And in some circumstances what they have may be the best for them. They will continue to buy off-the-shelf products and try to quilt themselves a patch work solution. I prefer the mind share of open source.

    -Matt


  28. <quote>
    Again, I would hesitate to lump them all as developers
    </quote>
    Matt,you can't just do that. You know what.The developers numbers ( millions !!! ) are arrived by including all these people.If they are comfortable with the technology give a damn to it.Anyway they are not going to be the loosers !!! some one is always there to buy something at a cheap price ( less price ).

    The bottom line is ,we can't just proclaim that J2EE is a great technology until & unless it is presented in a cool & easy way for use.

    ~murali
  29. I just talked with someone looking for a J2EE expert. Their system is JSP-->stored procs. This is no different then your ASP example. J2EE/Java and .NET and ASP/COM/MTS are really the same things.

    The high horse you're on is really a dusty old swayback nag.
  30. Sartoris,

    They may be the same things. But this is one of those specious, arrogant, jerkoff comments you hear from people who can't be even handed, have an agenda, or are zealots of some kind or whetever. I'm gonna bite, and respond.

    First off, not all of us are using JSP to stored procs. I've seen people like you, time and time again make specious comments like this. So, if one company does jsp to stored procs, then all J2EE developers are morons. What you are implying is obvious. You sound about as even-handed as Bill Gates or some Java zealot. This is the equivalent of how so many people defend racism/religious zealotry etc etc. "I met on java programmer, he was a jerk, they must all all suck."

    The high horse these people are on range from the "dusty old swayback nag" to the bloody Derby winner. Look at a system like JIRA which was the subject of a recent article here. That bad-boy is definitely gunning for the Derby. Very cool architecture, all sorts of other neat stuff.

    Then there is JSP to stored procedures. Maybe that is a nag, or maybe it's just what's needed. Using advanced architectures for many projects is simple "swatting a fly with a mac truck." On the other hand; have you considered that the reason that they are looking for a J2EE expert is because they have realized their limitations? Realizing your own limitiations is often the sign of the best developers and managers.

    My mentor taught me that "a good manager doesn't do everything, he just gets everything done." There is a difference, and maybe that's where these people are. Maybe you need to learn that yourself. You definitely need to learn how to be a little more objective, and a little less specious.

    If MS .NET is the right tool for the job, cool. If its J2EE, also cool. I like to work with cool stuff and get paid. I personally am not the biggest MS fan in the world, but I do try to keep my politics out of my job.

    -Newt
  31. "So, if one company does jsp to stored procs, then all J2EE developers are morons. "

    You are making my point. I was responding to this post:


    "I was invited by a small B2B company to help them design their software; 100% of their staff works with asp scripts. When I tried explaining the importance of frameworks and design patterns they looked at me as if I was an academic who had lost touch with reality. They are 100% convinced that .NET will solve all their problems. They will continue developing the way they do and the way MS wants them to. "

    I read this as "anyone using .NET is a moron".

    Anyone who thinks there's a bit of significant difference between J2EE and .NET is dead wrong. Many in the Java community suffer from delusions of grandeur.

    "If MS .NET is the right tool for the job, cool. If its J2EE, also cool. "

    I agree 100%.

    "I personally am not the biggest MS fan in the world, but I do try to keep my politics out of my job. "

    I didn't say anything polical. I was calling a stupid statement stupid. If a .NET lover claimed that .NET was oh so wonderful and that J2EE was junk I'd say the same thing.
  32. alright, alright, I mis-read your post (I didn't read previous ones all that carefully). I apologize for my outburst.

    I beg forgiveness.

    -Newt
  33. They are NOT the same. They have things that seem to be similar but they, as whole, are not. VS.Net still has nothing like servlets. VS.Net still has nothing like EJBs. It doesn't have Jini.

    Your example just shows that Java programming can be as easy as VS.Net. I'm using both and VS.Net is no easier than Java, with the right IDE. EJB's may be more complicated. But with a good IDE (like WSAD) they aren't that difficult. And there is nothing in VS.Net like it. BTW, COM+ isn't .Net.

    If one compares apples with apples then VS.Net is NO easier than Java. I use both. I've done VB since the early 90's so MS products are not foreign to me. I would much rather do Java in my Java IDE. With OS tools like GLUE and Hibernate I can whip out apps that use webservices and OO persistance.

    What really has changed for MSers? What can they do different? Hardly anyone did VB objects. If they did web developement, they used ASP pages and no custom COM components. Those who did are few an far between. So will the do it in .Net? It isn't any easier. The only thing that is easier is Web Services. Yipee.

    Mark
  34. <quote>VS.Net still has nothing like servlets</quote>
    .NET does have something similar to servlets.
    If you have the framework, check on IHttpHandler ms-help://MS.NETFrameworkSDK/cpref/html/frlrfsystemwebihttphandlerclassprocessrequesttopic.htm . I believe my.msn.com is using IHttpHandler for processing.
    In addition, if you wish to do filtering similar to servlet filtering , it can be done via IHttpModule:
    Refer to http://www.15seconds.com/issue/020417.htm article on this.

    <quote>VS.Net still has nothing like EJBs</quote>
    You can write ServiceComponet under .NET, refer to ms-help://MS.NETFrameworkSDK/cpguidenf/html/cpconwritingservicedcomponents.htm , but you still need COM+ to manage the runtime stuff like, JIA(Just-intime Activation), Transaction management (No support,Supported,New,Required), Declarative Security. I believe Microsoft has the intention to eventually move the COM+ services to the .NET world. Of course there is no equilvalent of EntityBean in the current COM+ world. But Microsoft has plan for similar stuff in a preview technology called Objectspace for .NET, I am not sure when will it be beta-released , refer to http://hosting.msugs.ch/dotnetrox/articles/Art07.html for preliminary detail.

    <quote>If one compares apples with apples then VS.Net is NO easier than Java. I use both. </quote>
    I used VS.Net and WSAD too, personally I still think VS.NET is easier and better IDE than WSAD. If you are doing Windows programming, the environment is very well-integrated, e.g, you can view/modify/update SQL database from the IDE, view NT event log and write event log or add performance monitor counter without leaving the IDE. I think maybe you can write an Eclipse plug-in to do these stuff :-)

    They are some small points like: Wheel-mouse does not work in WSAD, it works in the help window because it's basically a embedded native IE browser. I guess 1.4 will fix this problem.
    The context-sensitive help is very good and the collapsing of region is really a plus point in VS.Net.
    Personally I also think WSAD is a very good and capable IDE for the Java platform.

    <quote>If they did web developement, they used ASP pages and no custom COM components. </quote>
    I think that ASP.NET is going to the new model that most new web developement will be done on the MS platform. With the code-behind programming model and the server control model, it's definitely a very capable platform. Microsoft packaging of web control such as Calendar , Datagrid, Validation etc.. server controls definitely will speed up the web development for their platform. Of course in J2EE the Standard Taglib will be released which will ease the development of JSP web development on the Java platform.

    Microsoft did not submit ASP.NET to ECMA for standardization , so it is proprietary but I know that Mono project is determine to deliver a compatible-version of the ASP.NET on Mono to the Linux community.

    There is an article on Expand .NET beyond Windows by Mark Davis of IDG which I think it's an interesting read:
    http://www.fawcette.com/dotnetmag/2002_05/magazine/columns/strategy/

    Best Regards,
    Hun Boon Teo
  35. You got me on the IHTTpHandler. Gotta wonder how much the will be used though.

    COM+(MTS) is sort of like stateless EJBs. But still quite different. Building good COM+ objects is difficult.

    It is ok for you to believe VS.Net is easier and better than WSAD. I don't think it is. For the things I do and the way I program, WSAD is better. BTW, WSAD does allow you to view the database and write SQL, etc. My mouse wheel works in WSAD. (BTW, it doesn't in VB6.) Context help is just as good if not better in WSAD.

    Try figuring out what exceptions are being thrown in .Net. You have to look at the docs. Try loading classes you didn't know about a compile time. .Net is not really that much different than VS6 if you include VJ++ and Web Services toolkit.

    Server-side controls existed in VS6. I still think ASP development will be done the same way. It could have been done differently before and few did it.

    The MS cannot allow .Net to run viable anywhere besides Windows. What do they have that runs on anything besides Windows? So if .Net viably (Message Queing, Transactions, etc) moves beyond Window's what need will there be for MS? Won't happen. It is excellent marketing for MS to make everyone think it will. And THAT is part of the biggest threat to Java. Perception - "VS.Net is better than Java", "Windows and MS programs are cheaper and will get you quicker to market", "VS.Net has 20 'different languages'", ".Net runs on Linux." - not reality. If everyone thinks .Net is better then it doesn't matter if it is. Fact is, it ain't (given a few little unimportant things). Feeling, thinking, believing and liking certain 'features' won't make it true. Some things that are stated as advantages for .Net are things I would consider to be disadvanatges/programming techniques.

    Take away everything not .Net (IIS, Windows, COM+(MTS, MSMQ, etc.), SQL Server, Exchange) and what do you have left? Nothing useful.

    Mark

  36. Yes , in COM+, the only model is stateless component. Building a COM+ component is not difficult but building one with scalable and robust systems in mind required some thoughts and designs. I guess it's the same with EJB :-).

    I wonder what version of WSAD are you using(mine is 4.0.2) , my wheelmouse does not work except in the help windows (becaude it's embedded IE) . Look like I need to check with ex-colleague in IBM :-).

    Checked Exception is a topics that a lot people have different opinions , check the interesting discussion at Bruce Eckel site : http://www.mindview.net/Etc/Discussions/CheckedExceptions

    What is done in the past(ASP) does not mean that it will not be change in the current and future development of ASP.Net. It will be up to MS to educate their developers and programmers.

    Obviously, MS will want everyone to run all applications on their platform, what I am saying is that there is possibility that the open source implementation(Mono or Portable.NET) may produce a compatible-version with MS .NET platform in future which may or may not be endorsed by MS, again past behaviour may say they won't. I guess in 1-2 years time, we should be able to tell the outcome.

    I would say that MS .NET platform can be considered an advantage but it's also can be a disadvantage depending on which angle you look at it. It's tools (VS.NET) and the platforms (.NET framework, COM+/MTS, MSMQ etc) are very well-integrated and work well with each other. But you lose portabiliy and vendor choice. Again this is an area that a lot of people will like to dispute or disagree with me :-).
    This somehow reminds me of a Chinese saying "The river can carry the boat but it too can capsize the boat" :-)

    Regards,
    Hun Boon Teo
  37. One things you guys need to learn is to stop putting people down. Every time someone mentions a VB programmer, it is a negative reference. If I was a VB programmer, I would never want to join a Java/J2EE project, because I would be afraid of how everyone would look down at me.

    Also, some of the best business analysts/programmers may not be the best technical programmers. However, some of the most technical programmers are poor business programmers and have even poorer communication skills. Some customers aren't looking for rocket scientists, but instead average developers who can talk to their business people and convert their business requirements into code. I have worked with many brilliant developers who simply hate gathering business requirements and only want to work with architecture.

    Lastly, VB programmers have made a lot of money in the last 10 years, so I wouldn't call them dumb by any means.

    What we should be doing is trying to persude VB programmers to join us. This is an area that the Java/J2EE community is failing miserably at. We need more tools like BEA's Cajin and SilverStream. J2EE already has great technology. Now we need to make it easier to use.
  38. The biggest hindrance to J2EE adoption is the steep learning curve.


    Absolutely. Turnaround time is important to the business and if they have MS software and can cobble together something that works 'now' the concepts of scalability , security etc., are not on their minds. Going J2EE requires them thinking in terms of engineering and infrastructure, not quick gratification. Unfortunately in this world a quick answer is becoming more important than a considered answer.
  39. No it doesn't. If one does just JSPs they are doing J2EE. A J2EE app can be quickly 'cobbled' together too.

    J2EE is no more difficult than .Net. Java can be more difficult but if a .Net project is done like a typical J2EE project (or at least as close as possible) it will be just as difficult.

    Many of those moving to .Net are struggling if they are in any way doing OOP - which is not required if one uses an OOL.

    I agree that most are more concerned with getting projects up and running and less with the long term. But a Java project can be done just as fast.

    Mark
  40. So if 1+1=2 then 2+2=3?

    If these groups can see the future then what are the winning numbers for The Big Game lottery? That is more important than WAGging as to how much market share Java or .Net will have.

    Is this the world or just the US? From what I can tell MS is really losing it everywhere but in the US.

    Mark
  41. I think that Microsoft has been and always will be it's own greatest enemy (apart from the DoJ of course). Microsft wants you to believe the hype and then, when things are firmed up and people can get their hands on it, they wish people would back off from it. It happened with Windows 95, it is happening with Web Services now, and it is also going to heppen to .NET (IMHO).

    Please see http://www.theregister.co.uk/content/23/24976.html

    Yes, I wholeheartedly agree with other posters that MS has a great Marketing department and a fantastic UI team for it's tools - I personally have never found a better IDE than J++, but the technical implementations are where MS repeatedly falls short.

    If we are to believe MS, Web Services will work wonders for your enterprise, allowing massive EAI to occur - how exactly? Lets have a little look.....

    For every Java API in J2EE that allows you to do something, you have to build a SOAP spec. SO here we go

    JTA - ? ( I believe that Oracle and HP are dealing with issues of distributed transactions and guaranteed oredering)

    JAAS - ? (The WS-Security Language was only proposed a couple of weeks ago!)

    Even just these two are immediate requirements of any reasonably sized heterogeneous enterprise. It's the old 'square peg in a round whole' scenario.

    I personally would rather not have the issues of building, verifying, and parsing XML SOAP documents to deal with EAI internally. B2B absolutely fine - XML fits beautifully.

    A major benefit that MS has is the comfort and the 'new-lets use it' factors - Many existing developers will rather use XML Web Services to build internal distributed applications rather than more appropriate - in some circumstances - technologies like .NET Remoting, COM+, etc. which have failed to get anywhere near as much 'column inches' as Web Services

    The fact that an 'Analyst Group' can blatantly make predictions like this is unbelievable when .NET 1.0 was released less than two months ago, and many specifications that MS will require in order to turn much of the WS hype into reality, haven't got past the specification stage never mind implementation.

    Also bear in mind that of many of these 2000 companies, those that use Java may be perfectly happy with their Enterprise system, and may therefore evaluate Web Services but have no immediate plans to implement, however companies moving from VS 6.0 to .NET will want to adopt these new technologies ASAP, simply because they plug much needed holes in their existing environments.

    There you go,just my opinion

    Calum
  42. <quote>
    I personally have never found a better IDE than J+
    </quote>

    Obviously never worked with IntelliJ, Eclipse, or NetBeans. IntelliJ is ahead of the curve when it comes to XP-like development.

    -Matt
  43. I use netbeans every day - have done for the last 9 months solid - dropped JBuilder for it, and I think it's fantastic. I have not looked at IntelliJ but will do.

    My message regarding J++, was mainly in the 'smoothness' of the IDE, which, I must admit, netbeans is nearly getting to.

    As far as the options, plugins modules etc are concerned I wasn't really referring to them

    Calum
  44. One of the basic problems that will face MS developers who use these new fancy tools to drag and drop distributed solutions is the fact they are now developing distributed applications not VB UIs and developing secure, robust, scalable distributed applications is difficult - regardless of the tools and depends more on overall architecture and design patterns - areas that will be foreign to the average MS developer.

    Web Services are just a new face on an old problem - distributed computing - and if Microsoft succeeds in distilling the pool of knowledge on how to build secure,robust,scalable distributed systems into its tools, then it deserves all the market share it can get.

    However the key to building large scale .Net systems will be to get the best J2EE developers/Architects over to the dark side. (Why do you think C-sharp resembles java so closely) If they can do this, then you'll see a lot of good .Net systems. If they can't do it, then in 5 years time, the J2EE community will still be laughing at MS.

    Julian.
  45. My $0.02

    For those that insist that Java will destroy.Net or
    .Net will destroy Java - think about this.
    How many BIG airline makers are there in the world (two -
    Boeing and Airbus). How many big and successful Oil companies are there (3-4 really big ones).
    In almost any major industry there is usually at least
    two major players.

    Microsoft may have been able to build a monopoly on the
    desktop (but watch for Linux). But in the server environment? Highly unlikely. Java has too much momentum and also the greatest weapon of all - the Open Source
    community. But M$ has the best marketing of all plus
    all the free stuff that gets you hooked (locked) in.
    Ultimately though M$ has to charge for these things
    in order to be profitable.

    Java and .Net will coexist - the biggest challenge
    will can they interoperate? The freaking point about
    Web Services is interoperability. Let's not focus
    on the war but on interoperating. That way
    we all win regardless of what platform we are on.

    Year-to-year M$ may be up or Java may be up but there
    are a lot of skilled developers out there and too
    much $$$ on the line for the big tech firms to pass up
    so I foresee this see-saw for the next 3-5 years.

    Best,

    -Frank
  46. "Java and .Net will coexist - the biggest challenge
    will can they interoperate? The freaking point about
    Web Services is interoperability. Let's not focus
    on the war but on interoperating. That way
    we all win regardless of what platform we are on. " - Frank



    When will notepad be able to display unix formated ascii files ? It speaks volumes about their ideas on heterogeneous systems.

    Microsoft's webservices is another enticement to herd the masses to the slaughter house.

  47. <FLAME
    I think that .NET will have more market share than J2EE.
    Lots of PHB do J2EE/w EJB and .... when it fails they BUY .NET. The key word is they Buy, not deploy. .NET is vaporware, but it sells well.

    No I think the driver for .NET is not that it is good, but that PBE and newbies fail when they develop EJB. EJB is not scalable!
    EJB is there for Sun to sell HW.

    Solution:
    Write J2EE w/o EJB. Roll your own persistance layer (such as my sample at basebeans.com.)
    /FLAME>
  48. I do not wish to comment on these prodictions. Because of these predictions only, most of the venture money went into unrealistic business models (in so called dot com burst).

    But I wish to add one thing. I feel .net and j2EE are going to co exist just like COM/DCOM & CORBA. For mission critical enterprise applications, j2ee will be the dominent platform and small businesses will be using .net platform to expose their web services.

    Anyway, web services enables us to utilize the best of the both worlds (Java & Microsoft) which is good for the industry.

    Krishna Sriramagiri
  49. I'm not so sure about interaction via Web Services. It works if all I send and return are basic objects (integers, Strings, etc). What happens when I use a custom object? It works fine with a Java Client and Java Web Service.

  50. By 2004 means in 1 Year[ Go to top ]

    Microsoft is comming to the realization that .Net Server is not going to be widely available until 2003 ( http://www.infoworld.com/articles/hn/xml/02/04/26/020426hnmistake.xml )

    This prediction about 2004 therefore implies that in the span of just 1 year, 30% of web services are going to be deployed on .Net.

    The company I work for is a Microsoft house ( I am a lone but successful Java rebel ) focused at Government financial management. They are already making their .Net plans and upgrading their VS-6. But I can assure you that it will be at least a year before their developers can use .Net effectively, and another year before anything is deployed - read 2005 or 2006. VB programmers who use Excel & IE as a presentation layer and Foxpro (yuck!), Access, & SQL Server as the data layer are going to be using .Net as the business logic layer and for web services. That is a simple reality we must accept.

    I continue to write better JSP/Servlet code (no EJB yet) ~3X faster than the VB programmers, but the VB inertia and mind share can not be overcome in this Microsoft shop. The CIO point blank told me: "We wanted to hire Java programmers but they were too expensive and unavailable."

    In my mind, the momentum is still with J2EE:
    1) Universities teach Java, not VB. Most VB programmers are born enhancing Microsoft Office
    2) Look at the number of Open Source projects in Java vs. VB.
    3) Major application providers (SAP, SAS, IBM) are exposing their code bases through J2EE now and paying lip service to .Net (while waiting on it)
    4) High performance and 64 Bit computing is available now through specialized JVMs and it could be 2 more years before it is available through .Net
    5) Microsoft has not developed an alternative to JXTA, which may become very important is a increasingly distributed wireless environment
    6) Free, high quality, and open source J2EE IDEs and app servers that will run on many platforms

    Tom Schaefer
    Senior Technical Expert
    Tecolote Research, Inc.
  51. By 2004 means in 1 Year[ Go to top ]

    <quote
    4) High performance and 64 Bit computing is available now through specialized JVMs and it could be 2 more years before it is available through .Net
    /quote>
    I don't think this would be an important factor in the .Net v.s Java battle.
    Microsoft traditionally starts of at the lower end and then moves up to the higher level. Look at the evolution of NT from a cheap server replacing Novell thru to the Advanced DataCenter Server. Or at how SQLServer came in as a cheap replacement to Oracle and then moved on to bigger things.

    Microsoft's biggest pitch with .NET would be to department heads who have limited budgets and have to deliver projects in a short span at time. They don't mind some intern/developer slapping together some screens and getting it out of the door quickly. As long as there are cheap developers/interns available to maintain it they are happy.
    Also thanks to VB/Visual studio they have developed the perception that Microsoft products are cheap and easy to buy and build.
    J2EE's cause wan't helped either with all companies pushing EJB as the only way to go. Having Open source helps J2EE's cause but is still a difficult sell to managers. Managers are used to dealing with companies. If something dosen't work with a product there are people they can call to get it fixed or in the worst case they always have someone to blame.
    I do like Java and I hope competition from .NET will make it better.
  52. By 2004 means in 1 Year[ Go to top ]

    Tom,
      You should let your CIO know that it will not take them any longer to learn C# or VB.Net than it would to learn Java. In fact it might be better that they learn Java because it will minimize the 'Where did this vb function go?' syndrome. By sticking with MS development tools they are leveraging very little and risking much. The fact that Bill Gates said he might have to pull Windows from the market should scare every MSer out there (http://www.computerworld.com/storyba/0,4125,NAV47_STO70604,00.html). (Would he ever do it? He has all the money he needs.)

    BTW, I use both Java and VS.Net(customers get what they want) so I'm not just talking out of my behind.