Discussions

News: Microsoft releases Java-to-C# conversion tool

  1. Microsoft releases Java-to-C# conversion tool (81 messages)

    Microsoft on Tuesday released the first test version of a tool for converting Java language code to C#, the company's Java-like programming language for building Web services. The Java language conversion tool, available for download from Microsoft's Web site, is part of the Java User Migration Path to Microsoft .Net (JUMP to .Net) initiative unveiled about a year ago.

    read more @ http://news.zdnet.co.uk/story/0,,t269-s2103779,00.html

    Threaded Messages (81)

  2. From what I've seen of C# this would not take much effort, you could probably even do it in your favorite text editor with a search/replace macro!

    The problem here is that it seems they are only talking about the java syntax, key words, layout, structure etc. The majority of any java class is made up of API calls. This may be J2SE, J2ME or J2EE APIs.

    Are microsoft saying that they will also translate my J2EE API calls to something that will work in .Net and provide the same functionality? I need to know...

    Dave
  3. I'm the Microsoft Product Manager for the Java Language Conversion Assistant.

    Both our product (as stated above, now available for download from MSDN as Beta 1) and the ArtinSoft product, the Java Language Conversion Assistant Enterprise Edition, will take both the Syntax of the language and the class library calls and convert them to C# on the .NET Framework.

    The Microsoft released one will eventually (by RTM) take Java-language source code and JSP pages and convert them to C# and ASP.NET Web Forms.

    The ArtinSoft product (which will be in Beta in 2H 2002) will take the full array of J2EE features (EJB, JMS, etc.) and convert it to .NET.

    There is a specific newgroup supporting this product set up on msnews.microsoft.com for more information.

    That said, I'm happy to answer questions here but I will be offline (getting married) over the next few days so my response time may be spotty :)

    I look forward to your feedback.

    TonyG
  4. I can contiune to develop in Java and then ( if needed)
    run the conversion tools.

    No need to learn C# and .Net.

    wojtek
  5. Wow, this is great! Now I can continue to develop my Java applications knowing that not only will I be able to deploy and run them on any J2EE compliant app server, but that it will also run on MS .Net. This is taking "Write Once, Run Anywhere" to levels that even Sun never dreamed of!!! :)

    Cheers,
    Chris
  6. Making the "WORA" dream come true!!!!!![ Go to top ]

    Chris -

    Good point.

    If we build apps using C# and .Net today, we couldn't change our minds tomorrow and quickly convert to Java and J2EE.

    But if we build apps using Java and J2EE today, we could change our minds tomorrow and (according to MS) quickly convert to C# and .Net.

    Now we have an answer to the nagging question about what technology to use.

    "When in doubt, use Java"

    Sounds like we can satisfy management, ensure the longevity of our applications, and guarantee portability ... only by using J2EE.

    - Jim Bell
  7. Hi Chris,

    >>Wow, this is great! Now I can continue to develop my Java >>applications knowing that not only will I be able to >>deploy and run them on any J2EE compliant app server, but >>that it will also run on MS .Net. This is taking "Write >>Once, Run Anywhere" to levels that even Sun never dreamed >>of!!! :)

    Wow.. this is tooooooooooo good !!!.. I can never imagine can anyone thing in this direction :-)

    Murali Varadarajan

  8. As I read about it, this tool will only do some 90% of automatic conversion, still a lot of hand work to be done in complex projects. Then it is a one-way conversion tool and, continuous maintenance being a fact of life, that 10% could prove unmanageable. Actually I find more intriguing this other product: iNet. It translates MSIL to Java code, and apparently it needs far less hand work, so you can go the other way around: develop in .Net, deploy in Java. Are these fun times or what?
  9. One way to view this is that Java is the only platform. The inverse is that Microsoft intends to make Java a Microsoft platform without having to license Java. These frameworks are just the means by which Sun and Microsoft puppetteers the software development community and thereby controls the fate of the world.
  10. Tony Goodhew,

    I have never seen you post in here before and find it highly coincidental your appearance coincides with sherat's post.

    Is Microsoft sponsoring sharat nellutla's H1B visa?
  11. n n,

    You're absolutely correct that I've never posted here before but I have been reading this site for some time.

    I must thank you because your continual search for conspiracy theories is absolutely hilarious and I appreciate the humor, no matter how unintended, you bring here.

    That said the only co-incidence here is that I announced this product today and I've been ready to post follow-up messages on threads that appear about it.

    If you have questions about the product/technology I'm happy to spend the time answering them, however from now on I'll leave comments on your search for black helicopters and Soviet divisions in salt mines under Detroit to others.


  12. 'I must thank you because your continual search for conspiracy theories is absolutely hilarious and I appreciate the humor, no matter how unintended, you bring here. '(TONY)

    Bill said something along the same line, until his ass was handed to him in court.


    'If you have questions about the product/technology I'm happy to spend the time answering them, however from now on I'll leave comments on your search for black helicopters...' (TONY)

    Thanks for the offer but I only use products with an established track record.

    btw,

    That black resin is for anti-radar purposes.

  13. Hi Tony,

    Tony: "I must thank you because your continual search for conspiracy theories is absolutely hilarious and I appreciate the humor, no matter how unintended, you bring here."

    Your making fun of Mr. n n does not somehow mitigate the fact that your company's track record in this area is terrible, any more than making fun of a moron would make you intelligent or throwing mud on a pig would make you clean.

    Conspiracy is defined as "an agreement to perform together an illegal, wrongful, or subversive act", and "a plot to carry out some harmful or illegal act". Seems like the court already showed your company to be clearly engaging in conspiracies. The .NET project is easily construable as a logical extension of what was already clearly shown to be a conspiracy.

    In other words, the burden of proof is on you now to show that your company has (in the past week or so I guess?) suddenly stopped acting in a conspiratorial fashion. And you are not doing anything to alleviate anyone's worst opinions. Unfortunately, our dear Mr. n n, although wrapping his comments in metal translucence, appears somehow to be even more lucid than you.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
  14. Cameron, I salute you.
  15. I'll second what Cameron said and add a point about a recent event that might have been dismissed as a "conspiracy theory" had not a magazine bothered to take the time to examine their logs.

    Does a certain web poll about what platform people are planning to use to write web services on ring a bell? Shame... Shame...
  16. John Munsch,

    What was that MS poll issue? I haven't heard about it. Can you provide a link to this story or some brief info on it.

    Appreciate it.

    Gurnam
  17. Hi Gurnam,

    Recent news on Microsoft poll rigging:

    http://news.zdnet.co.uk/story/0,,t269-s2102244,00.html

    Older news, same idea:

    http://www.theregister.co.uk/content/1/13255.html

    Constant news (a bit anti-MS you could say):

    http://mailbox.univie.ac.at/~prillih3/evil-empire/

    For the other POV:

    http://www.microsoft.com/freedomtoinnovate/default.asp

    Peace,

    Cameron Purdy
    Tangosol, Inc.
  18. Cameron,

    As I said, I'm happy to answer questions on the product or technology - If you have one please feel free to ask.

    Thanks,

    TonyG
  19. n n

    Maybe you could save your senseless drivel for someone who cares.

    Is taking the piss at Microsoft so incredibly important that you have nothing useful to say? I don't think I've ever seen you post anything that was insightful rather than inciteful.

    -B
  20. a) Well, since he is the Product Manager for a Java Language Conversion Assistant tool, it would make sense to keep up on java and some java web sites.

    b) Do we assume because of Sherat's name, that he is on a visa.

    c) I assume that since your own name sounds somewhat foreign, that you maybe from another planet, or maybe you are just yellow.

    d) Maybe this is not the best place for Sherat's or Tony's posts, so maybe we can just ignore them .
  21. Posts from n n have consistently brought otherwise interesting discussions on this forum into the schoolyard, with his childish nonsense. If this were a usenet group, I'd be voting for moderation, or at least be looking to put anything from n n in my killfile (until he renamed himself to m m... :).

    It would seem unlikely that any "automatic" conversion from one platform (J2EE) to another (.Net) would be 100%, but if a project really does require this, it might at least get the majority of the boring conversion work done - followed by a lot of hand-crafting and debugging...

  22. Lee,

    You're correct in the statement that automatic conversion will not be 100% - If anyone ever tells you they have a tool that will do 100% conversion then they are probably also lining you up for a bridge sale :)

    We're aiming to do pretty much what you stated - Deliver the bulk of an application (in the 80-90% range) conversion automatically and then make you as productive as possible in the remaining manual conversion through integration with the VS IDE.
  23. Microsoft clears up some J2EE FUD[ Go to top ]

    Tony wrote:
    > Lee,

    > You're correct in the statement that automatic conversion
    > will not be 100% - If anyone ever tells
    > you they have a tool that will do 100% conversion then
    > they are probably also lining you up for a
    > bridge sale :)

    > We're aiming to do pretty much what you stated - Deliver
    > the bulk of an application (in the 80-90%
    > range) conversion automatically and then make you as
    > productive as possible in the remaining manual
    > conversion through integration with the VS IDE.

    I have read reports claiming that if you write an application for one J2EE server, it is extremely difficult to port it to another J2EE server.

    My point is simple: If it is so easy to port any J2EE application to .NET, how can it be any harder to port one J2EE app from on app server to another? The only substantial thing you seem to get from the first approach that is lacking in the second is vendor lock-in as the process in one-way.

    I hope everyone appreciates what Microsoft's JUMP program has done to elimitating this FUD about J2EE portability.
  24. Microsoft clears up some J2EE FUD[ Go to top ]

    First of all, right on Cameron! Tell it to em.

    Second, is it really easy to port ANY(!) J2EE application to .NET?? So, the conversion tool not only convert Java code and code that uses vendor specific APIs, but only the deployment descriptor AND vendor specific deployment descriptor, huh!?

    Wow! That cool. And another thing, how they do it for stateful session beans and entity beans? I thought .net doesn't have these.

  25. Microsoft clears up some J2EE FUD[ Go to top ]

    Richard,

    You miss the point of the tool - There is significant work needed to port applications from one app server to another as well as moving it to .NET. The presence of this tool doesn't change that underlying fact.

    What this tool does however is to do a lot of that work for you in the case of moving to .NET. It is our expectation that eventually moving from WebSphere/WebLogic to .NET will be easier (by using the AiS tool) than moving between app servers (without the tool).

    We spent a significant period of time reviewing migration technology from many companies around the world before we decided to partner with ArtinSoft. Quite simply the core ArtinSoft technology is much further ahead than any other migration companies products.

    TonyG
  26. Tony

    Can I just stretch this a little? Assuming that .Net also supports Java as a scripting language (syntax, etc. calling .Net APIs) and java can be used anywhere C# can be used. Then the conversion to C# is not necessary, I could just translate the API calls to .Net equivalents.

    Now also assume that I use this translation capability to create a proxy for the J2EE interfaces implemented by .Net API calls and still use the J2EE APIs in my code.

    I am essentially running an application server on .Net.

    Does .Net have equivalent structures for EJB containers, Web Applications, etc. And will the ArtinSoft software also convert deployment descriptors etc to the .Net equivalents?

    Will .Net then be J2EE compliant?

    Best wishes for your wedding.

    Dave

  27. Dave,

    First up thanks for the best wishes - I'm actually only going to be able to answer this one and one more before I have to get to the plane.

    W.r.t the API use, in either case you'd need to convert the underlying calls to the framework. That is really the task that both of these tools do for the user.

    Let's just focus on the ArtinSoft tool for the moment as it will convert EJBs etc.

    If you were to create a layer about the J2EE App Server (ie your own framework) you could use the Conversion Assistant to move that framework to .NET and still preserve your class layout. I won't go into the derivative ideas here but I think you can see what would be possible.

    Our intent (MS & AiS) is that the full J2EE tool should convert all the application information needed to create a functionally equivalent application on the .NET Framework.

    We have no intention of providing a J2EE implementation layer for .NET.

    TonyG
  28. Hi Tony,
        Its nice that u r following up this thread. Could u pl. explain about ur tool,how does it will convert a Java web application into .net web application. Say, i have a webapp
    with some JSP files, Some class files and 'm using some infrastructure like <b> STRUTS </b>. How does ur tool will convert this into a .net application. How much manual
    effort u expect for this. How does this will take care of this kindof infrastracture like STRUTS or some other.
    actually, we don't want to go through whole msdn to get this info. if you provide in breaf... that would be great. then we can discuss, how much success u get with you
    tool..

    cheers,
    Best wishes for your wedding.
    sharat nellutla
  29. Hi,

    Thank you for this posting. Until this point we are in doubt in our project -that what should we use .Net or J2EE . (Though currently we are in J2EE). Now we have no doubt that our choice should always be J2EE (because any time we can convert our code to .Net). And we I have forwarded this link to some of my friends also saying that - Don't worry use J2EE without thinking anything for the future.

    Thanks
    Srijeeb.
  30. Correct me if I am wrong, but....[ Go to top ]

    OK so let me get this straight....

    Number 1: Web Services give a developer the ability to either wrap or integrate .NET components so that they can be invoked by any code-based client that can generate a SOAP-XML document over HTTP, JMS, or RMI. This means Java, C++, VB clients blah blah blah...

    Number 2: This new C# converter gives a developer the ability to transform his Java code into C# for the purposes of integrating this C# code into a end-to-end MS environment. An analogy would be to have a java converter that would transform C# code into Java code for the purposes of integration into a end-to-end Java environment.

    Number 3: Java developers will have very few reasons to ever *need* to convert their code to C# because of Number 1, and C# developers will likewise have to think hard in order to understand why ever they should port their C# code into Java( even if this was possible, which as a previous poster said, is not too difficult using a simple "Search-and-Replace" macro on any decent text editor... ).

    Out of genuine curiousity, can anyone give me a realistic and frequently occuring scenario as to why either "side" would want to do this? It seems to me that MS already made the most important move by becoming part of the WSDL standard --- when MS did this, they automatically put themselves on the same rung of the ladder as everyone else, which was a stretch considering their past reluctance to do so. The fact that they've developed this C# converter sounds like goodwill to me --- which is great I don't mind can't hurt thank you. Just my two cents...
  31. Correct me if I am wrong, but....[ Go to top ]

    I think this tool will probably be used for applications already developed and not new applications. Microsoft had alot of Visual J++ customers who need to convert to .NET.

    If anyone used this tool for new applications, they will slow down productivity. Java and C# syntax is not all that different anyway. Just develop for the target platform.
  32. Tony,

    Thank you very much for being brave enough to post here!

    I have a few questions:

    1. I have as of yet been unable to find the newsgroup you mentioned. The link you gave doesn't seem to work. Could you provide some direction on how to find the newsgroup?

    2. How does this relate to J#.Net? I have evaluated it with VS.Net Beta 2 and was pretty impressed, but J#.Net has not been updated to work with the Visual Studio release version. I was most impressed with J#.Nets ability to convert Java byte code, and I am eagerly awaiting the J# release.

    Thanks for your time.
  33. Hi Alan,

    Thanks for the comment.

    1. The news server is msnews.microsoft.com and the newsgroup is microsoft.public.vsnet.jlca.

    2. It's just another option for developers who wish to move to the .NET platform. We will be releasing an RTM hosted version of Visual J# .NET in Mar on MSDN.

    TonyG
  34. Ridiculous[ Go to top ]

    No one (management or tech) with an ounce of sense or any experience with language translation is ever going to believe you can do what you claim you will do. You are not going to take a huge, CORBA/RMI, EJB, directory enabled, multi-tier application and just convert it to .Net. Most managers I know no longer buy extravagant, self-serving claims from anyone, especially Microsoft.

    How far behind is Bill Gates going to let Microsoft get before he finally starts firing some people? Why just code for Windows when you can code for deployment on any platform with J2EE? Rather than run your conversion tool, why don't I just put my J2EE application right on Windows XP without any change at all? Are a few seldom-used C# bells and whistles and a yet-to-be-proven WebServices architecture that compelling?

    Microsoft may delay a few IT managers by making preposterous claims like yours, but frankly I just don't foresee anyone but entrenched Microsoft shops buying in. I think Microsoft will be lucky to hold onto even the entrenched shops. "Vendor lock-in" has burned way too many people. I doubt Microsoft is going to talk anyone out of Apache or WebLogic or WebSphere or Oracle.

    By not supporting J2EE, Microsoft is talking people out of Windows in the long run.
  35. Ridiculous[ Go to top ]

    I'm quite confident that both RMI and CORBA will convert to .NET remoting. .NET remoting does everything RMI does plus some additional features like custom protocols (xml/soap over http or binary payload over TCP and others as well).

    I agree that interoperability would be very cool, but we DO have web services available if that's a serious concern.

    I guess you'll have to wait for the next release to see exactly how well, or not, "...a huge, CORBA/RMI, EJB, directory enabled, multi-tier application..." will be converted to C#/.NET.
  36. Hi Tony !

    Tell me something. Is this feature of .net going to bring some add on to Java programmers. Is it going to be an "open"
    project towards the Java community. I would like to know since I was a typical Pascal>c++/asm>Visual Studio 6.0>and finally Java programmer and it seems like I might be getting into the .net in about 6 months since my boss is acutally giving me a choice to boost up my vista in .net direction. Can you give me any pointers about what I might be getting into (in a distant future of .net tech.)? Will .net acutally going to be fully compatible with J2EE,...

    send mail after h. moon to: mitja dot dominko at email dot si
    Mitja Dominko.
  37. Well. another "magical tool" from MS, let's check some facts:

    * First, and most important, as a J2EE developer I don't see the point in having a tool that converts my app. into MS only code. I mean, my app. already runs on MS OS!

    * Second, I can take my Linux app, that's written in C, and compile it under MS C compiler, no language conversion here ( It's both C right? ), and still it doesn't work. Why?. Because of the libraries!. So, is MS porting the libraries too?. And if so, will they be J2EE certified?.

    Jeez, things you do with millions in th bank...
  38. Hi,

    yest they did port some lib (JDK, JE22 ?). But as we all know MS by now, they don't tell you the hole truth. I guess, if you use other libs beside the JDK, as everybody does (like JDOM, ORO, ....) you're lost,

    That's the same game they play with ECMA. Telling people C#/.NET is standardized, but of course most of the libs (the crucial part) are not.

    bye
    -stephan
  39. Tony,

    Many thanks for posting and indicating your willingness to answer questions. It is great that you have taken the time to do so. Of course, being a J2EE site most of us have a preference for J2EE. But I'm sure .NET will prove to be a viable architecture and I'm looking forward to learning more about it.

    Regards

    Bruce Blackshaw
  40. This tool seems like "pie in the sky"- especially when you consider that J2EE EJB implementations are specified by the vendor. For example, porting code between JBoss and Websphere is a chore, for pure deployment reasons alone. I can see it working for some of the standard sdk from java, but the vendor specific stuff.. I doubt it.
  41. I can't belive that ppl are still buying this crap. .NET on other platforms and converting Java into this and that ...

    This web site has earned many praises but it every now and then finds something which is not related to what ppl are coming here for, and posts it here in order to get attention and fuel these useless religious wars ... seems like they are getting paid according to how many posts are added after topic that they pick from who-knows-where ... no further comments here ...


  42. I can't belive that ppl are still buying this crap. .NET on other platforms and converting Java into this and that ...

    This web site has earned many praises but it every now and then finds something which is not related to what ppl are coming here for, and posts it here in order to get attention and fuel these useless religious wars ... seems like they are getting paid according to how many posts are added after topic that they pick from who-knows-where ... no further comments here ...


  43. Is C# an official specified language, with standard API's? Are there any non-microsoft centered C# projects that people can point to?

    I'd like to look into C# if it doesn't tie me to the Microsoft platform and their "be all and end all" agenda.

    -Pete
  44. There actually are some non Microsoft .NET stuff out there. For example http://www.newsforge.com/article.pl?sid=02/02/05/1252253&mode=nocomment (iNET, a Java implementation of .NET) and http://www.go-mono.com/ (.NET on Linux and other platforms)

    /Fredrik
  45. This is a great tool. Now no Java developer will need to learn C# or .NET technologies (which doesn't take too long anyway). All we need to do is to design and Java, right? (This assumes that MS creates a fully compliant tool).

    Then, the other question: Who do you expect to change his/her J2EE application and compile it into .NET? Why? Does-it sound reasonable? Not to me! What arguments does .NET have to modify an existing J2EE application? I'm probably missing something here... ;-)

    On the other hand you give me a good argument with this tool. Now I'm able to say "Let's develop it in Java! If we really need it, we can compile it into MSIL".

        Yagiz (http://www.erkans.com)
  46. If you are like me then you were probably thinking "why would MS or AiS would want to do this in the first place?". Doesn't this seem to encourage a default choice of J2EE when making decisions? (since you can always go to .NET if it turns out to be necessary but not vice-versa).

    After thinking for a while, I came up with (in order of importance):

    1) To allow new and legacy J2EE apps to be deployed on .NET servers instead of J2EE servers

    Most money in J2EE products is probably made on the servers (not tools) so MS gets a share of this revenue. MS is banking on .NET servers as a key source of revenue in the future. Good decision by MS here. You utilise all the current knowledge (best practices, design patterns etc) in J2EE, allows developers to use their familiar J2EE tools while developing and then still get the money with .NET servers come deploy time. If .NET servers can be proved superior in cost/performance/security etc then it will make it an attractive idea.

    2) To utilise the 10% that can't be ported

    That is, an average of 5-10% (IMO) of most Java applications will not be able to map nicely to the .NET framework (as Tony has pointed out); this is fair enough. However if this 10% turns out to be something that, when reworked, directly ties you in with .NET (ADO.NET maybe?) and then becomes hard to migrate back to J2EE if necessary then effectively you have lost your benefit of conversion. It then becomes a one-way street from J2EE to .NET once you convert.

    3) To draw upon the currently large Java developer base

    Face it, no matter how good .NET is, you need developers to drive it into industry. Enticing J2EE developers to the .NET framework by utilising (not deprecating) their current investments in time, techonology and applications is a good idea - much better than promoting the "start from scratch with .NET" approach.

    So Tony, I'd be interested to hear your perspective on this is. Am I way off mark with my points above? What was your main motivation for creating such a product?

    Regards,

    Ben
  47. Hi Tony,

    We probably don't care too much about the announcements at this stage. All we need, in the software industry, tools that works as they are supposed to work. I think anybody with a little experience on Internet Technologies learnt not to get too excited with the early announcements.

    We need to know how far this conversion go. How much effort is required to complete the operations? What happens if I use a broad range of different java packages? How is the mapping done to .NET libraries?

    In case of J2EE, the work seems to be huge (if possible). EJB is just one of the technologies covered by J2EE. How far can the mapping go?

    Best wishes for your wedding,
    Kind regards,

        Yagiz (http://www.erkans.com)
  48. Thanks Yagiz,

    For the first beta release of the JLCA we think we'll cover around 85% of the most used classes. Remember that this tool is the Java language and JSP to .NET converter - For your J2EE code (EJBs etc) the AiS tool will have far more relevant coverage.

    But to talk in terms of base %'s of classes covered is somewhat misleading - The reality is that most developers will use a core set of classes and if we can convert those, we're going to be able to do a good job for the bulk of developers. As I said above no converter will do 100% of an application but we're trying to make the process as painless as possible.

    On of the things the Beta 1 tool does is to produce a list of the classes you use (that's it nothing more and it's fully documented in the readme) and asks the developer if they would like to send that back to us. From there we can get a better idea of the classes used and focus our resources on doing the best coverage.

    There have been a number of questions on "How does this work" and it'd probably worth writing up a whitepaper on this for people. I'll get that done and hosted on MSDN (I'd be happy to post it here but I'm not sure that can be done).

    TonyG
  49. "First they ignore you, then they laugh at you, then they fight you, then you win." Mohandas Gandhi

    This is exactly what is happening here. Microsoft is trying
    to wage a futile war, beacuse we are winning.
  50. "When a thought is too weak to be expressed simply, simply drop it." - Marquis de Vauvenargues

    Seems to me, that the independent observers (are there any in here?) will see the Microsoft product manager acting professionally (and openly evangelising), in contrast to a number of the usual suspects with the "Java-rockz" level of emotional argument...
  51. Lee: "Seems to me, that the independent observers (are there any in here?) will see the Microsoft product manager acting professionally (and openly evangelising), in contrast to a number of the usual suspects with the "Java-rockz" level of emotional argument..."

    I'm not sure if we've been following the same conversation, then. At the top of this page, the logo reads "Your J2EE community", so it seems unprofessional for a product manager for a non-J2EE (and conceivably anti-J2EE) product to be evangelising here.

    Nor have I seen Java rockz arguments, although Mr. n n did give us some interesting opinions, but our friend from Microsoft already discounted those for you.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
  52. "At the top of this page, the logo reads "Your J2EE community", so it seems unprofessional for a product manager for a non-J2EE (and conceivably anti-J2EE) product to be evangelising here."

    Unprofessional? Sounds like he's doing he job of trying to sell his product. Personally, I wouldn't see an enormous value in this "conversion" either, so I'm not arguing for it's value. However, many replies don't seem to discuss the product/technology, but rather attack Microsoft and the product manager.

    Seems to be a trend in most of the posts around here lately.
  53. Lee: "many replies don't seem to discuss the product/technology, but rather attack Microsoft and the product manager."

    Frankly, it didn't seem that way to me, but you are correct that "attacking" is pointless. I'd rather see the Java community learn from some of the neat little tricks that Microsoft innovated in the .NET platform. I'm already getting some good ideas in a couple of areas, and I've noted a bit of a trend with the JCP lately in the same direction.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
  54. Cameron: Fair enough. I'm glad you've been around enough to take a less-religious view. I'm sure many of us have used a a number of technology generations, and expect to use many different technologies in the coming years. One of the most interesting parts of our jobs is understanding the various technologies and making informed decisions.

    Meanwhile, "n n" gives another example of childish personal attacks, rather than intelligent discussion, instead of doing his homework (or whatever he does). ;-)

    Lee.
  55. Meanwhile, "n n" gives another example of childish personal attacks, rather than intelligent discussion, instead of doing his homework (or whatever he does). ;-) -Lee

    Lee,

    Glad you asked. I just finished looking at Prudential Security's backend architecture and don't know if to cry or laugh.




  56. "n n",

    Interesting. What group do you work for in Prusec? I also currently work for Prusec. Email me on lee_h_fuller at prusec dot com.

    Lee.
  57. ;^)

    Wouldn't you like to know.

    *grin*

  58. "Wouldn't you like to know."

    Pathetic.
  59. Oh, I see. It's that time of the day.

    http://www.metamucil.com/





  60. "Seems to me, that the independent observers (are there any in here?) will see the Microsoft product manager acting professionally (and openly evangelising), in contrast to a number of the usual suspects with the "Java-rockz" level of emotional argument..." - Lee


    The "usual suspects" had insightful comments that were ignored by your biased eye, instead you resorted to your usual trolling and whining.


    I suggest Microsoft find an astroturf for dotnet and not make up for their short comings on theserverside because www.gotdotnet.com bombed. Microsoft's marketing of bait ware on theserverside will further fuel dissention from java developers when they find out what's behind this thread's spin: college freshman technology.


    "acting professionally" != "behaving professionally"

    I agree Lee, Microsoft is a great act.


    Java-RockZ(evangelising)




  61. Has anybody thought about the licensing issues of converting open source code?
  62. Hi Ben,

    Just have a quick chance to answer some questions before I really disappear :)

    To look at this from a high level you have 2 major drivers in the space. The first and foremost is interoperability. The industry investment in web services should enable all developers, regardless of platform, to integrate applications easily.

    The second driver is the platform decision. We think that .NET is the best platform to build the next generation of internet applications on.

    You can disagree with this and I expect that many will but the motivation for this tool was to simplify the porting process.

    We've seen a number of customers who have migrated their applications by hand to .NET and we wanted to provide them a tool to make this much easier. Effectively this tool lowers the bar to entry to .NET and helps customers regain a lot of the investment they have made in Java during the conversion.

    To be pragmatic in this space, if you don't believe that .NET is better for you then there isn't any need to migrate is there?

    By providing a platform that has native support for XML Web Services, allows you to do more with less (both in performance and developer resource) and decreasing the TCO costs (in scalability, deployment, reliability and robustness), using .NET is a sensible choice.

    TonyG
  63. Hi,

    Having actually used .NET in some small projects I doubt that it will have too much impact. Simply stated, it is a poor copy of JAVA. It may have a few bells and whistles that may have some slight impact on very few people but the underlying theme is that .NET still has the dependecy issues that is typical of MS products. I had to install IIS5, MDAC2.7 only works on some versions of windows, ASP.net only works with IIS. It killed my office installation on install and then again when I removed it! It is super huge well over a gig with the SDK. Unlike Java apps like Resin where you move one file and it's running .NET based apps can be a pain to move from the dev box to the server...

    I went out used .NET and created some stuff - I am now back in the JAVA world and am not looking back. I am always open to new tech and will jump ship to where ever the grass is greener. The most interesting thing I liked about the ASP.net is the WebForms where all the html elements are bound up into objects called controls. In addition the Webforms handles state automatically and allows you to bind control classes that manipulate the page as an object. Sounds great and is, but WebObjects came up with it first and still does it much better than WebForms.


    How does this relate to the conversion tool? The answer is, who would want to convert? Yea, I want to convert my solid, reliable multi-vendor, mulit-platform application into a vendor specific platform that is essentially a Java copy with bugs.

    Yes .NET is a huge leap forward for MS, but still falls short.

    One point in defense of MS, I was doing all this work with the RC1, not the RTM version of the .NET framwork and used VS.net. But my arguments still hold true.

    Stefan
  64. Oops,

    Made a mistake I meant I had to install IE5.5 not IIS5.

    Stefan
  65. "The most interesting thing I liked about the ASP.net is the WebForms where all the html elements are bound up into objects called controls. In addition the Webforms handles state automatically and allows you to bind control classes that manipulate the page as an object."

    We now have something very similar in the JSP camp:

    http://www.dotjonline.com/taglib/form.jsp

  66. Why is everyone talking about some blue sky world where everything is inter changeable ? MS want to migrate people from Java onto .Net not open up a two way street. Look at the name of the initiative 'JUMP to .Net' I didn't see any mention of '...and back again'.

    Is it real or is it a stunt who knows and who cares ? As they say 'you pays your money, you take your choice'
  67. Greetings,

    I would look carefully on this. Most MS products never look as a threat for open source tools/OS at the beginning but then they simply took over all market just because they always started from simple things. Who could say ten years ago that MS Windows 3.1 will grow in one of the most widely used OS (of course in its new version 98/NT/2000/XP). Same thing can happened to Java vs. C# with this new tool (if it will work correctly). Existing of this tool and small price of the deployment platform based on the MS products (W2K, IIS, SQL Server) can made MS platforms most used for "J2EE" applications but in .NET shape and then MS will simple stop supporting translation tool but there will be hundreds (or even thousands) of "J2EE" sites deployed on .NET platform, so they will be forced to use MS .NET because moving back to pure Java will be very expensive.

    Price of the production environment based on MS products (W2K, IIS, SQL Server) less then Solaris with WebLogic and Oracle RDBMS and that is what most companies are looking for. Most non technical people do not see benefits of second solution (better scalability, performances, multiple solutions, etc.) so they go easy way - W2K server.

    Best regards and stay alert,

    Taras
  68. Taras,

      I see your point and have heard this argument before. I don't mean to discount it completely, but in order for this theory to be valid, MS's C#-conversion tool would have to do the following. I am purposely being sarcastic to get to the point:

    1) Feb-Dec, 2002: every single company in the industry ports over their Java apps to C#.

    2) Jan, 2003: MS suddenly announces that they will no longer support this conversion tool. They do this by adding some crazy proprietary stuff to C# that the experts at SUN cannot understand and cannot make Java integrate with. The Fortune 500 fires everybody with the word "Java" in their resume. MS needs to hire about 600 000 new consultants to keep up with demand. Nobody complains or does anything to stop MS. For some reason, nobody cares to enforce free and open standards anymore.

    3) Feb, 2003: SUN goes out of business. So does BEA, IONA, Pramati, TIBCO, SonicMQ, CapeClear, and so forth. This is because nobody wants to buy their products anymore since EVERYBODY who is ANYBODY is writing their apps in C# or some other .NET syntax. IBM is still around; in fact, their stock inexplicably rises by 10% for the month.

    4) Mar, 2003: Bill Gates calls a worldwide press conference. He announces that he has invented communications satellite which will reduce the cost of cell phone usage by 50%. All wireless companies using a JVM or J2ME apps instantly file for Chapter 11.

    5) April, 2003: With every company in the world successfully converted to .NET, those 600 000 new consultants are becoming redundant. Instead of laying them all off and causing a global economic crisis, MS changes the XML spec to use square brackets instead of angle brackets for nodes and Hungarian notation is no longer allowed. The equivalent of the "Year2000" bug is experienced, funding the industry for another year...

    etc etc...

    Of course, this will not happen. Not only will everything "Not MicroSoft" not allow this to occur, but I think that MicroSoft itself is seeing the value with playing on an open field. They have a good product that can do some great things for customers who need it. This converting tool, however used, will not change that. Give it some time and you will see that it is a goodwill gesture to those J2EE developers who ever want to understand .NET.
  69. Thats exactly what I thought. Because the converter and .NET server might be much cheaper(?how much?) than Solaris+WebLogic+Oracle, M$ will kick out WebLogic and Websphere, and becomes a most used application server for J2EE applications.
  70. Also Taras, to address your point about 3.1, which is very relevant:

      Back when Win3.1 was released it was easy for some to see how much potential MS had to swallow the industry. Whether MS meant to or not, that was its destiny. This is because it made more sense to use to a larger group of people than any other operating system at the time despite its shortcomings.

       I guess what I'm saying is that I can remember back in the late 80's early 90s that most people who really made an effort to transition to new technology found it easiest to do so with Windows, whereas if you were to show these same, normal, non-techie people all the capabilities of Vi and a Bash shell, they would sprint away from your computer class shrieking in horror.

       Again, it's all about marketing and getting the right product to the right people. That's all that this C# converter is too. I totally hear what you're saying, but I just can't see that happening anymore -- the industry is too mature now to let these kinds of things spiral out of control like that. And what with new technologies like Web Services, what's the point?
  71. Greetings,

    Ted I see your point. I agree with you that in current time it is hard to compete with big number of open source products plus to fight with technology that highly supported at least by Sun and IBM. But here a few facts: not so long time ago IE took over market of the Internet Browsers, right now Netscape slowly moving out of usage. Secondary I would add that the company where I work is developing Web Apps for customers and in most cases if application deployed in customers' environment it is deployed on MS platform as ASP application. Most small and middle companies have MS OS installed everywhere and they are not going to by extra Linux server plus hire Sysadmin to maintain it.
    I agree that currently MS does not have force to take out Sun's Java from the market. But I think we should keep in mind what we are dealing with.

    Best regards,

    Taras
  72. Hi Taras,

      I understand your point of view as well. However, with regards to the situation of IE vs. NetScape, that case was an attack on a competing product using a proprietary technology ( the Windows o/s ) as a competitive(!) advantage. In a sense, it *may* be incorrect to compare that issue with this one. This issue, when it comes down to it, is an *option* for corporations that may have the need ( and I can't imagine what -- still waiting for scenarios from Tony or anyone else ) to convert Java source code into C# code. So, the C# converter is not an attack per se, although I understand what you are saying: you are saying to be careful because while it may not be apparent right now, this converter could set the stage for a future attack. Which brings us back to the unlikely 4 steps I listed in my post above.
       In other words, if this is some disguised attempt to rid the world of Java developers, it's not going to work. Period. And MS is smart enough to know this, so I am not yet willing to insult that intelligence by suggesting that kind of a conspiracy. Key word: "yet". If somebody can show me otherwise...
  73. Sorry[ Go to top ]

    Sorry for my negative comments.
  74. Sorry[ Go to top ]

    You need to find a more challenging way to spoof, since you're so bored. Just in case it's on someone's mind, there isn't anything to apologize about.

    n n(the real one)

  75. Time to invent a new term:

    BAIT WARE


    Before you know it we are going to have stranded java programmers taking the bait on m$'s dotnut/c# because they can't get anywhere with their supposed java support. It's going to remind me of the old re-runs of Gilligan's Island.
  76. Why would anybody want to build against a .net black box? I would never go back to developing against a closed environment. Being able to open the hood and examine the engine is how you learn and grow and better yet the community benefits from many eyes examining it. It takes cuts to open source your code. Where is your guts M$?

    -Matt
  77. J2EE, SOAP & .NET?[ Go to top ]

    Perhaps the Microsoft Managager could answer this for me. If I have a J2EE based app, using SOAP it can interface seamlessly with a .NET service or visa vera, right? Besides a marketing ploy, what good is this tool? Wouldn't both microsoft AND sun be better suited in setting up a standard set of data structures that the two platforms can share?

    An example, if I have a J2EE app that requests a table of information from any web service, I still need to know how to parse the incoming SOAP message's data.

    As far as I can tell the only REAL difference between WebService, RMI, DCOM, Corba... is that WebServices are XML based as oppossed to having a proprietary protocal. While this is a huge step forward, I don't see how it is an end-all solution.

    It seems from reading the posts is that the MS conversion tool is pretty un-usable (this could fairly be attributed to the fact that jdk 1.1.8 is targeted?). I also know that if I am a product manager, I do not want to rely on a "conversion" tool to get my product to work on another platform.

    It also seems strange that MS would even create such a tool. The argument for making a Java shop into a .NET shop has the same pitfalls as trying to make a MS shop into a Java shop. Both sides tend to be fairly religious about such isuues. It would still seem to me that centering on inter-operability would suite MS more. MS shops run .NET, Java shops run J2EE and everyone communicates just fine. Everyone wins!
  78. J2EE, SOAP & .NET?[ Go to top ]

    I'm no Microsoft Manager, but here are my 2 cents:

    "Wouldn't both microsoft AND sun be better suited in setting up a standard set of data structures that the two platforms can share?"

    This is already starting to happen, see for example this new consortium. Oddly, Sun is not part of it.

    "... the only REAL difference between WebService, RMI, DCOM, Corba... is that WebServices are XML based as oppossed to having a proprietary protocal. While this is a huge step forward, I don't see how it is an end-all solution."

    There are no silver bullets, just steps forward. I agree that true interoperability (no matter if it still requires a lot of hand work), is a huge step forward.

    "It also seems strange that MS would even create such a tool."

    My *very personal* opinion is that they have created this tool for: a) provide a migration path for Visual J++ projects, b) send a signal about the fact that the Java and .Net platforms are similar enough as for automated conversion tools being possible (I'm not saying the signal is right or wrong, I'm just saying is being sent).
  79. J2EE, SOAP & .NET?[ Go to top ]

    Marc,

    Our investment in XML Web Services answer your main point.

    This tool is to assist those customers that have decided that the .NET platform is a better platform to run their applications on and would like to recover some of the investment they have made in Java.

    If you just want to get systems to work together then XML Web Services are the way to go.

    TonyG
  80. Come on... are you all going to trust Microsoft again? Was Microsoft doing a service to Java developers when they pulled RMI from their JVM and hid it some obscure FTP directory? How about when they extended Java into that "J++" beast -- did that help you? Now they've killed app and applet support by default on XP and they are offering you some new Java tools...

    Some java developers seem to be very gullible or their short term memory is just not there.
  81. I'm sure there is somebody out there currently working on a tool that converts .net apps into J2EE apps. This kind of rather cutthroat competition moves the focus away from co-existance and cooperation via webservices where each technology gets to compete on their strengths.
  82. Having the ambition to be THE definitive environment for Server programming, J2EE fall short in the support of different languages.
    Let´s look at the final picture, 2 or 3 years from now and extract some conclusions:
    J2EE will support multiple languages, added tools to ease the development, debug and test. free App. server and tools
    .NET: will be ported to any major platform, low cost, piracy allowed for research and students.
    Both platforms extended to cover the missing functionalities of the other
    Cross platform converters at the JIL<->MSIL level, no recoding needed.
    .NET API for J2EE, J2EE API for .NET, one only virtual application server. increase of Market concurrent HW and SW solutions, better prices..

    Microsoft is good in this scenario, and I can bet this scenario is around in your mind. But basically the success depends on good technical AND good marketing ideas, both of them are important. I think the J2EE people has to learn from the Microsoft success to face the treath.