Discussions

News: Developer.com posts products of the year awards, Java wins

  1. Ed Burns pointed out that developer.com has posted their products of the year 2006 awards, with some interesting results. The framework of the year was JSF, the tool of the year was Eclipse, and the utility of the year was Ant; the Technology of the year was AJAX. What's surprising in all this is that .NET products didn't show up except in specifically Microsoft-oriented categories.

    Other interesting results: Java EE 5.0 won for the JSR of the year, and J2ME won the award for mobile technology.

    Java dominated the awards, with open source products Firefox, OpenOffice, and MySQL (with Google's web services API) rounding out the unrestricted categories (the non-Microsoft-oriented categories). While this might mean that Developer.com's polls aren't broad enough, it might also indicate that Java's much stronger than some of the pundits think.

    Threaded Messages (67)

  2. It is really great for Java. Just two things:

    1)why it is 2006 ? At the bottom of this article:

    >Congratulations to all the people who produced the winning
    >products. Thank you to all the readers who participated in the
    > Development.com Product of the Year 2005 contest from the >nominations through the voting.


    2) as seems to me at least the total number of voters must be announced

    Marina
    http://www.servletsuite.com
  3. One-sided results[ Go to top ]

    It doesn't look convincing for me, despite the fact that it looks like something expected. Kind of suspicious...
    I didn't participated, but would like to see contenders and vote numbers.
  4. Jsf framework of the year? And neither Rails, nor Spring nor hibernate. Great. Sounds serious
  5. Jsf framework of the year? And neither Rails, nor Spring nor hibernate. Great. Sounds serious

    Rails has only just reached version 1.0, and has (as yet) had very little visible impact on mass commercial development. Hibernate is a good framework, but surely it would have been more appropriate to nominate EJB 3.0 persistence, of which Hibernate will be one of the most important implementations. However, EJB 3.0 is not finalized yet.

    JSF on the other hand has stimulated a lot of exciting work, both commercial and open source, and is being used for an increasing amount of commercial development. I realise it is not 'fashionable' to support 'specifications', but there you go.

    I do take your point about Spring though...
  6. While this might mean that Developer.com's polls aren't broad enough, it might also indicate that Java's much stronger than some of the pundits think.
    The result mean nothing other than who reads developer.com and bothers to respond. Its a very self selected group.

    Look, I am a Java/JEE guy. I am very confident that Java is very strong and getting stronger in the enterprise space - the space where I work. I don't care for Microsoft or its development tools. Nevertheless, I know a number of Microsoft moonies and they are just as happy as the java junkies. They also tend to be a bit more insular meaning they are less likely to spend time at non-Microsoft-centric sites like developer.com than we fans of open systems, java and open source.

    The results are meaningless, the conclusions are worthless and publishing these sorts of non-events is stupid - unless you're developer.com's or a vendor's PR firm.
  7. Ed Burns pointed out that

    As seems to me JSF spec lead name is Ed Burns. The same Ed?
  8. >Ed Burns pointed out thatAs seems to me JSF spec lead name is Ed Burns. The same Ed?
    Same Ed, yes. He was understandably happy with JSF's win. :)
  9. Ok, JSF should be happy too ;-)

    P.S. why do they (Sun) will need such kind of stimulation for such a great stuff ...
  10. That is the kind of results that just let us think that answers are dicted by the way question are asked !!!

    I don't know for you guys... but to me there is no interest!

    How can people spend their time doing such polls !

    Stephan
  11. Sounds like .. bullcrap[ Go to top ]

    don't know for you guys... but to me there is no interest!How can people spend their time doing such polls !Stephan
    +1
  12. Sounds like .. bullcrap[ Go to top ]

    don't know for you guys... but to me there is no interest!How can people spend their time doing such polls !Stephan
    +1

    +2
  13. Sounds like .. bullcrap[ Go to top ]

    don't know for you guys... but to me there is no interest!How can people spend their time doing such polls !Stephan
    +1
    +2

    +3
  14. Not credible[ Go to top ]

    It ain't credible... XML Spy wasn't selected as "Best Java IDE"!
  15. Not credible[ Go to top ]

    +1!!
  16. Not credible[ Go to top ]

    It ain't credible... XML Spy wasn't selected as "Best Java IDE"!
    LOL!!!!!!
  17. JSF and JSF and JSF[ Go to top ]

    I have no problems to believe that JSF got most hands on Matt's presentation, its new and its allways nice to get informations about new techniques. But on the other hand, i wonder how many people involved in web programming for several years would chose JSF for their next project. Or better, how many people would chose JSF again after using JSF for the first time.

    I came in contact with JSF at T-Mobile, where i had to use it for about 4 months and sorry, i dont think its anyway sexy. I see its advantages and i could even imagine to use it in a web app project where i have complex table interactions and a lot of components on one page. Here JSF has it strengths. But on the other hand, for typical web frontents, i dont see any benefit at all compared to action frameworks. Ok, you dont have these pretty components on the JSP side but we have custom tag libraries which can be equally powerfull. Controlling HTML only on the basis of renderers is also not very handy for several situations. I also found it very hard to use JavaScript properly on JSF generated pages, perhaps it was because of my low experience with JSF but its definitely unintuitive at best.

    Of course i am biased because i used Action based frameworks for a hell long time, but you know the feeling when you are using a new framework and it somehow feels ugly to deal with. Another thing that really bothers me is the lack of industry support regarding add-on components. On the one hand you have this uber-component-framework and on the other side you dont have too many companies offering advanced components for it.

    Marc
  18. JSF and JSF and JSF[ Go to top ]

    As for the components check out myfaces ;-)
    As for the rest, it is a matter of use cases, if you need straight html with navigation basically every framework is suited.
    But this is not the case anymore nowadays, I did not have
    had and application which at least didnt need a decent date picker etc...
    And suggesting to go for custom taglibs is only a valid approach, if you have the code which often you do not.

    Industry support, I am not sure into which stuff you looked, but on the oss side there is at least, Tobago, the myfaces tomahawk extensions and now also parts of ADF, also the jsf:html lib, jsf-comp on sourceforge and jenia4faces.
    Only ADF And Tobago fall into the all or nothing area the rest works straight otu of the box with any JSF implementation and is intermixable.

    As for commercial offerings:
    Studio Creator2 has extensive components,
    ADF Faces from Oracle (the commercial offering is more extensive than the code drop)
    Webgalileo
    Icefaces and others
  19. JSF and JSF and JSF[ Go to top ]

    .Only ADF And Tobago fall into the all or nothing area the rest works straight out of the box with any JSF implementation and is intermixable.

    Oracle are donating around 100 JSF components to the open source MyFaces project, so no-one will be able to say that there is not a rich and high-standard set of JSF components available for anyone.
  20. JSF and JSF and JSF[ Go to top ]

    .Only ADF And Tobago fall into the all or nothing area the rest works straight out of the box with any JSF implementation and is intermixable.
    Oracle are donating around 100 JSF components to the open source MyFaces project, so no-one will be able to say that there is not a rich and high-standard set of JSF components available for anyone.
    They have already, the code drop was last week... check out the myfaces developers list for more info...
    And yes there are plenty of components already even Oracles donation aside, there already is a huge number of components.
    But many people often make the mistake to check the JSF RI, which is just a basic reference implementation of the standard, and then dismiss it as all there is.
  21. JSF and JSF and JSF[ Go to top ]

    And suggesting to go for custom taglibs is only a valid
    >approach, if you have the code which often you do not.
    So JSF components are not custom tags? Or what do you mean here?

    Marina
    http://www.servletsuite.com
  22. JSF framework of the year.. HA!!!!

    Did they mean "Half baked" framework of the year?
  23. JSF winner framework - does it even stand on its own ?
    Who the hell voted . Looks like a lot of influencing.
    Raises serious questions about developer.com and also the ethics of JSF framework guys ? And as usual - no details of votes given anywhere.
    TSS should really consider posting such a item.
    This news item is as good as a tabloid story.
  24. joke of the year from Developer.com[ Go to top ]

    JSF winner framework - does it even stand on its own ?Who the hell voted . Looks like a lot of influencing. Raises serious questions about developer.com and also the ethics of JSF framework guys ? And as usual - no details of votes given anywhere. TSS should really consider posting such a item. This news item is as good as a tabloid story.

    An interesting point of view! I had never thought of various frameworks having different levels of ethics. This establishes a whole new way of looking at APIs and development tools. Perhaps we should add a morality star rating to product reviews.

    As for the questions about developer.com - well, perhaps we should start an urgent campaign. After all, we can't have such websites going around such questions of their readers without supervision, can we? I'm sure they would have no problem with selected TSS members vetoing the questions and auditing the answers, to check for undue influence, or answers they don't like. I mean, look at the other answers they have given - AJAX, Ant, MySQL... obvious signs of corruption and influence there.
  25. joke of the year from Developer.com[ Go to top ]

    I see no conspiracy in this marketing action, but results surprised me. I thougth JSF is a dead framework, but if somebody sells it then probably somebody will buy it too.
  26. joke of the year from Developer.com[ Go to top ]

    I see no conspiracy in this marketing action, but results surprised me. I thougth JSF is a dead framework, but if somebody sells it then probably somebody will buy it too.

    Oh come on Juozas - what possible evidence could you have for thinking that? No matter what you think of the framework itself (and there certainly have been problems), no-one who keeps up with current development in Java could possibly think it was 'dead' or simply a matter of marketing. One of the most successful implementations of JSF is the open-source MyFaces - now a major Apache project. Surely you aren't going to accuse Apache of 'selling it'?
  27. joke of the year from Developer.com[ Go to top ]

    It was not very active and I thought it will be canceled, but I see I was wrong.
  28. JSF rocks![ Go to top ]

    Actually JSF is a truly wonderful platform building other frameworks on top of. Seriously take a look at Seam and then consider whether it is possible to do anything remotely similar using any of the other web frameworks out there. JSF and EJB3 together give me such a wonderful set of primitives for building a truly productive full-stack solution.

    I spent some time trying to figure out how to support Tapestry in Seam, but to my disappointment there was simply no good way to do it. JSF and EJB work so well together because they are both POJO-based frameworks (with copious extension and plug-in points). Tapestry is not POJO-based, so I can't have a "Tapesty EJB". It just doesn't work. Pity.

    Meanwhile the other stuff in the space is mostly all coarse-grained action frameworks (so called "Web MVC") like WebWork and Struts. I'm sure this kind of approach still has its place but, I'm sorry, I just can't use this stuff anymore. Too many bad experiences. We need a "true MVC" approach (component-oriented, fine-grained event model) like what JSF has.

    Now, we might support WebWork in Seam. It would be pretty easy to do and if people ask for it, I'll do it. But I simply can't see any advantage that it would provide compared to JSF. JSF is IMO just a stronger model for event-driven applications.

    Frankly, most of the stuff I have read about JSF on blogs etc is fairly ill-informed stuff written by people who know other frameworks quite well but have never taken the time to actually learn JSF properly. For example, I have seen "JSF works only by postbacks" written in many places, and complaints that you can't do redirect-after-post. This is not actually true. (I believed it myself for a while, just because I had read it so many times.)

    Don't write off JSF just yet, it's barely got going :-)
  29. JSF rocks![ Go to top ]

    I prefer action/command frameworks, I think it is less error prone for multithreaded and transactional applications. I have no good arguments, probably it is just a dogma or tradition.
  30. JSF rocks![ Go to top ]

    I have to admit that i am not aware of the oracle JSF donation and i indeed worked with the RI of JSF. So perhaps i used the lowest possible set of options available for JSF, but before the Oracle donation, it wasnt really too much there. I made an extensive research but time goes by and right now its possibly a different situation.

    And BTW there also plenty of good custom taglib components for plain JSP usage like common controls. So people really cant sell me on the "hey i can easily create a nice looking table quickly" argument. CommonControls also has some sort of event paradigm on top of struts (not as powerfull though). Ajax integration was a pain in JSF, not so in plain JSPs.

    But one thing is for sure, i wouldnt code all the stuff by hand in JSP anymore. But right now common controls fits very well and i also have components and component models for lists and things.

    Marc
  31. JSF rocks![ Go to top ]

    I have to admit that i am not aware of the oracle JSF donation and i indeed worked with the RI of JSF. So perhaps i used the lowest possible set of options available for JSF, but before the Oracle donation, it wasnt really too much there. I made an extensive research but time goes by and right now its possibly a different situation.And BTW there also plenty of good custom taglib components for plain JSP usage like common controls. So people really cant sell me on the "hey i can easily create a nice looking table quickly" argument. CommonControls also has some sort of event paradigm on top of struts (not as powerfull though). Ajax integration was a pain in JSF, not so in plain JSPs. But one thing is for sure, i wouldnt code all the stuff by hand in JSP anymore. But right now common controls fits very well and i also have components and component models for lists and things. Marc

    Actually the Oracle code drop just happened, the donation was announced however earlier. As for the options, yes it is true lots of momentum has happened last year. But I had a similar problem as you had around beginning of 2005. I was looking for a good framework, found that there finally was a new standard (JSF) went straight to the RI and basically thought, that does not cut it, just basic controls.
    Then I stumbled upon myfaces, which already had a lot of controls in place. But myfaces back then was not an Apache project so its visibility somewhat was limited.

    As for the taglibs yes, there are taglibs there, also event systems. But the keypoint is, JSF has those mechanisms standardized, it does not bring anything extremely new to the table, it just tries to focus on binding all these efforts which have proven over time to have their merits and advantages into one standard, which then can be used by users without tools and toolvenders and users with tools equally.

    So the main keypoint not really is, that you cannot do it without, it is more a question, why do it without, if you have everything in place neatly and in a standard way.

    As for Ajax, Ajax is a pain in JSF, currently yes. This is partially caused by the fact of having server side component s which are directly connected to the view layer and the rendering phases.
    One key point is that it is rather tough to have the dom tree in sync with server side changes caused by Ajax.
    The other one are the restore and save cycle which can be in case of scoping interrupted by ajax.

    Having the backend side component tree however is vital if you really want to have a rich client ui programming model in a server side rendered context one way or the other.


    But those issues currently are adressed on various corners:

    a) Several component libs already have an entire ajax or ajax like infrastructure in place. IceFaces for instance, also the ADF donation has one in place (although it uses IFrames)

    b) The official JSF standardization gremium is tackeling the problem on the specification side currently

    c) MyFaces Tomahawk currently has some basic ajax controls in place which are not too intrusive (the classical example for autocomplete for instance)

    d) Apache Struts-Shale also is tackeling the entire Ajax issue from the framework side of things as we speak (although I have not looked at the code of this area of Shale yet)

    e) And there is Ajax anywhere, which fits nicely into the entire scope of supporting frameworks (ajax anywhere is a general purpose library working with many web framework technologies, JSF being one of them)
  32. JSF rocks![ Go to top ]

    So the main keypoint not really is, that you cannot do it
    >without, it is more a question, why do it without, if you
    >have everything in place neatly and in a standard way.

    the main problem in the current state is "all or nothing" approach. Try to mix plain JSP and JSF in real applications. That is actually a pain.

    Marina
    http://www.servletsuite.com
  33. JSF rocks![ Go to top ]

    >So the main keypoint not really is, that you cannot do it >without, it is more a question, why do it without, if you >have everything in place neatly and in a standard way.the main problem in the current state is "all or nothing" approach. Try to mix plain JSP and JSF in real applications. That is actually a pain.Marinahttp://www.servletsuite.com

    Not really, there is an issue with the c:forEach, in jstl that will be addressed by JSF 1.2. But you do not need that, you have dataList and dataTable which do the same, besides that JSF and JSP can be mixed quite nicely, after all the default implementation of jsf is done for JSP.

    All you have to take care is that you have to use f:verbatim in some cases (datatable being one of the few) to enclose the jsp/html specific part. (if you are unsure where, use verbatim all the time you do that)


    In 4 of the 5 applications I have written with JSF so far, I had one or two jsp/html fallbacks somewhere in between, and no issues so far. The key is, you have to know what you can do and what you cannot do.
    But JSF 1.2 will remove one of these issues the JSTL c:forEach issue at all, and the verbatim tag is hardly needed unless you want to handcode html, and even then you can use a jsf:html library written by one of the Exadel guys.


    But as I said the keypoint is, that stuff like having to fall back into jsp is very rare, and once you hit such a case you usually start to look for solutions because you already have entirely forgotten that you are in a jsp context anyway (given that you program for jsp not xul, vt220 or another possible rendering tech)
  34. Hibernate is[ Go to top ]

    Listen to none of that! Hibernate is the tool most of us can't do without.JSF is just another alternative
  35. JSF rocks![ Go to top ]

    but before the Oracle donation, it wasnt really too much there. I made an extensive research but time goes by and right now its possibly a different situation.

    The Oracle donation was very recent. All the companies listed in previous posts that provide JSF components were around before that.
    Ajax integration was a pain in JSF, not so in plain JSPs.

    There have been plenty of products, projects and resources providing or decribing this around for some time:

    http://www.icesoft.com/products/icefaces.html
    https://bpcatalog.dev.java.net/ajax/jsf-ajax/
    http://www.jscape.com/webgalileofaces/
    http://www.otrix.com/livedemos/webgrid/automaticPaging/index.faces
    http://ajaxanywhere.sourceforge.net/index.html
    http://blogs.sun.com/tor?entry=creator_and_ajax_the_demo
    But right now common controls fits very well and i also have components and component models for lists and things. Marc

    The point about JSF is that it is a specification that many companies and groups are developing for. A lot of the effort that has traditionally been put into developing interactive web pages is likely to become unecessary, as components providing AJAX integration, and quality rendering in a range of technologies - HTML, XML, SVG, WML, will be available off-the-shelf.
  36. JSF rocks! - JSP[ Go to top ]

    If the JSP aspect bothers you and you want (true) JSF behavior, then you can look at Facelets which feels a lot like JSP, but is built and optimized (faster) for JSF. I really couldn't conceive writing a JSF application in JSP now, especially with the templating functionality and 'composite' tags that Facelets provides. JSF's use of POJOs promotes lots of re-use, but when married with Facelet's re-use in the UI, it's the only way to do JSF apps. Plus, it already has support for Oracle ADF and other renderers such as XULFaces.

    Jacob (un-biased) Hookom :-)
  37. Ship ya Boat.. Masonry anyone???[ Go to top ]

    Gavin,
    His Highness, Lord Howard Lewis Ship will come up with a framework called Masonry. Do you dare challenge his "Tapestry community" with your pity little HIBERNATE community? Be careful with what you wish for or shall it be granted!

    Oh god, may the little aussie WIN and let the mighty AUSSIES lose the next world cup. (All point fingers to Ponting)
  38. JSF rocks![ Go to top ]

    I spent some time trying to figure out how to support Tapestry in Seam, but to my disappointment there was simply no good way to do it.

    So, does the fact that you think you can't support Tapestry for Seam make Tapestry a bad framework? Does Tapestry need Seam to be efficient in use? Or maybe JSF needs it for that reason? I am pretty sure a smart guy like you would perfectly be able to support Tapestry if he wanted. Surely there are ways around the abstract component limitation. It smells to me like there are other reasons for bashing it.
    Frankly, most of the stuff I have read about JSF on blogs etc is fairly ill-informed stuff written by people who know other frameworks quite well but have never taken the time to actually learn JSF properly.

    Maybe part of JSF's 'problem' is the fact that there seem to be a few well-known promotors which show up every thread to come to JSF's defend by saying all the critics about JSF are wrong. And when you consistently do that with an arrogant tone and bash some competitors in the process, changes are the JSF will get a bad name sooner or later.

    I like the direction Seam is heading and I´m checking the sources on a regular basis. One thing I never understood though (sorry for OT) is why it was started while there are two at least partially competing databinding JSRs going. Will we have a battle of the databinding APIs in 2006? :-)
  39. JSF rocks![ Go to top ]

    Maybe part of JSF's 'problem' is the fact that there seem to be a few well-known promotors which show up every thread to come to JSF's defend by saying all the critics about JSF are wrong. And when you consistently do that with an arrogant tone and bash some competitors in the process, changes are the JSF will get a bad name sooner or later.

    I honestly feel it is exactly the other way around. On almost every thread about web framework JSF is regularly rubbished as 'difficult', 'hardly used', 'dependent on tools', 'slow', 'hardly supported', 'only suitable for legacy development'. On this thread we have even had bizzare comments like:
    Raises serious questions about developer.com and also the ethics of JSF framework guys ?

    And claims that JSF is dying!

    You may call simply correcting facts and trying to explain things 'arrogant', but I don't. And if there is any arrogance and bashing it is nothing compared to the anti-JSF stuff I have read; some which I can only put down to some sort of anti-JCP/JSR feeling.
  40. JSF rocks![ Go to top ]

    Eelco, I'm probably the biggest critic of JSF and I was on the EG. There's been tons of times when I've thought, "Hey, I could write a better framework to do this," but I keep on ending up with same fundamentals/underpinnings that JSF includes-- around event management, the phases of a request, the component relationships, etc. The JSF spec is really big, yet it leaves tons of opportunity to do some really great stuff as Gavin attested.

    If you think Seam is just for databinding, then I really don't think you've given it a clear look. I'm pretty sure JBoss could add Seam support to other frameworks, but there aren't near the extension points and customizability (again the fat JSF API).
  41. JSF rocks![ Go to top ]

    BTW Doe's somebody knows how "event management" is better than struts style action mapping ? Is it an attempt to make web application design similar to single user/thread application design ?
  42. JSF rocks![ Go to top ]

    BTW Doe's somebody knows how "event management" is better than struts style action mapping ? Is it an attempt to make web application design similar to single user/thread application design ?

    Yes, pretty much. The new component frameworks are similar to taking each of your action-mappings w/ validators, forms, tags, and throwing them into a single, customized document. From that document, you can produce 'valueChangedEvents', 'actionEvents', 'actionListenerEvents', and even your own customized events.

    -- Jacob
  43. JSF rocks![ Go to top ]

    BTW Doe's somebody knows how "event management" is better than struts style action mapping ? Is it an attempt to make web application design similar to single user/thread application design ?
    Yes, pretty much. The new component frameworks are similar to taking each of your action-mappings w/ validators, forms, tags, and throwing them into a single, customized document. From that document, you can produce 'valueChangedEvents', 'actionEvents', 'actionListenerEvents', and even your own customized events.-- Jacob

    Sorry go give food for flames here. But the action state action lifecycle can be compared to VT220 user interface styles.

    Forms->Events systems are like rich client UIs, there is about 10-15 years of development in between. HTML programming was a step back user interface programmingwise at least 20-30 years depending on the perspective of things.
    Suddenly things appeared which thought to have died out again since Xerox Parc hit the scene of user interfaces.

    A user interfaces has to be seen as a set of objects interconnected and reacting to messages, that kind of system has worked tremendously since day1 of graphical user interfaces, while a form not doing anything->action->next form system died out until the web hit the scene and revived those things agin.

    People who think, that a pure form->action based system is state of the art, should check out the history of computing and how user interface programming outside of the web has evolved over time, and which things have died out and which stood the test of time.
  44. JSF rocks![ Go to top ]

    Some single user MVC frameworks like GEF convert UI events to requests and it makes sence ( controler listents model events, but model is updated by commands ) for Undo/Redo feature. Probably it doe's not make sence to listen model events in web page, but it makes sence to use requests and commands/actions for transaction processing (demarcation,resubmit).
    It looks like JSF listens UI events, is this design motivated ?
  45. JSF rocks![ Go to top ]

    If you think Seam is just for databinding, then I really don't think you've given it a clear look.

    I know Seam encompasses more than just databinding, but from an end-user's perspective - considering the possibly missing parts of other frameworks - databinding is where the candy is. You would expect serious web frameworks to have some answers on state management, dependency injection, etc by now thus in other words if that were to be the main selling points of Seam it seems to me it is selling a framework to overcome shortcommings in another one.

    Anyway, that's a complicated way of saying that I think datbinding (actually binding to the business layer, not necesarily just data objects) is a field where there is a lot to be gained by the average web app developer. It would make sense to seriously consider supporting alternative frameworks.
  46. JSF rocks![ Go to top ]

    I spent some time trying to figure out how to support Tapestry in Seam, but to my disappointment there was simply no good way to do it.
    So, does the fact that you think you can't support Tapestry for Seam make Tapestry a bad framework?

    Um. No. Did I or anyone else here say or imply that Tapestry was bad? I like Tapestry, that's why I wanted Seam to support it.

    The point is that JSF being totally POJO-based is a wonderful feature of JSF that many of the other frameworks just don't have. And it is one of the things that makes JSF a great platform for integrating other component models.
    Does Tapestry need Seam to be efficient in use?

    Deep integration between the we framework and the transactional layer (EJB3) has the potential to massively increase developer productivity. So, IMO, this is an important issue.
    I am pretty sure a smart guy like you would perfectly be able to support Tapestry if he wanted.


    Well, there is "support", and there is "deep integration". I can't do "one kind of stuff" - one of the key principles of Seam - with Tapestry, because a Tapestry component can never be a session bean.
    Surely there are ways around the abstract component limitation.


    Well, I'm not an expert on Tapestry, but I'm not aware of any good workaround.
    It smells to me like there are other reasons for bashing it.

    Who bashed Tapestry?? All I did was point out the strengths of JSF, in response to people who were most certainly bashing JSF. Sheesh. If pointing out an advantage of one approach compared to another counts as "bashing", then we can never have a sensible technical discussion about anything.
    Maybe part of JSF's 'problem' is the fact that there seem to be a few well-known promotors which show up every thread to come to JSF's defend by saying all the critics about JSF are wrong.

    That's terrible. Just terrible. Can you imagine anyone else in the Java community behaving in that way?
    And when you consistently do that with an arrogant tone and bash some competitors in the process

    Truly terrible. Arrogance and competitor bashing are just unheard of and totally out of place on TSS.
  47. JSF rocks![ Go to top ]

    Truly terrible. Arrogance and competitor bashing are just unheard of and totally out of place on TSS.

    Exactly! We live in a perfect world where there no such things as hidden agendas, inflated egos and biased opinions. Comments on a thread like this clearly shows that.
  48. joke of the year from Developer.com[ Go to top ]

    JSF winner framework - does it even stand on its own ?Who the hell voted . Looks like a lot of influencing. Raises serious questions about developer.com and also the ethics of JSF framework guys ? And as usual - no details of votes given anywhere. TSS should really consider posting such a item. This news item is as good as a tabloid story.

    Ethics of the JSF framework guys? That's a bit of a leap, don't you think? As a "JSF framework guy" who knows most of the other people involved with the spec (and I'm on the EG, too), I don't think there was anything unethical going on. Personally, I didn't know anything about this poll.

    And, as for JSF standing on its own, I personally can say that it does (since I get payed for developing with, and training people on JSF), and there are quite a few success stories (see http://www.jsfcentral.com/trenches -- new articles to come).

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Kito D. Mann (kmann at virtua dot com)
    Principal Consultant, Virtua, Inc. (http://www.virtua.com)
    Author, JavaServer Faces in Action
    http://www.JSFCentral.com - JavaServer Faces FAQ, news, and info
  49. joke of the year from Developer.com[ Go to top ]

    JSF winner framework - does it even stand on its own ?Who the hell voted . Looks like a lot of influencing. Raises serious questions about developer.com and also the ethics of JSF framework guys ? And as usual - no details of votes given anywhere. TSS should really consider posting such a item. This news item is as good as a tabloid story.
    It's not that simple.

    At Java in Action last year, Matt Raible made a three hour presentation on Web framework, and despite his personal fondness for WebWork, he was honest and asked the audience which framework they'd like him to talk about.

    Which one do you think won?

    That's right. JSF.

    It wasn't even close.

    WebWork got a couple of hands, Struts about a dozen (note that pretty much everybody in the room knew or used Struts, they just didn't want to hear a presentation about it) and JSF easily received half of the room (at least fifty).

    For all its flaws (real or alleged), JSF is receiving a lot of exposure and sparking a lot of curiosity, and I am betting that it's not going to go anywhere, no matter how good its competition is.

    And yes, it is pretty depressing to be threatened to be stuck with a sub-par standard for Web frameworks... again.

    --
    Cedric
  50. joke of the year from Developer.com[ Go to top ]

    JSF winner framework - does it even stand on its own ?Who the hell voted . Looks like a lot of influencing. Raises serious questions about developer.com and also the ethics of JSF framework guys ? And as usual - no details of votes given anywhere. TSS should really consider posting such a item. This news item is as good as a tabloid story.
    It's not that simple.At Java in Action last year, Matt Raible made a three hour presentation on Web framework, and despite his personal fondness for WebWork, he was honest and asked the audience which framework they'd like him to talk about.Which one do you think won?That's right. JSF.It wasn't even close. WebWork got a couple of hands, Struts about a dozen (note that pretty much everybody in the room knew or used Struts, they just didn't want to hear a presentation about it) and JSF easily received half of the room (at least fifty).For all its flaws (real or alleged), JSF is receiving a lot of exposure and sparking a lot of curiosity, and I am betting that it's not going to go anywhere, no matter how good its competition is.And yes, it is pretty depressing to be threatened to be stuck with a sub-par standard for Web frameworks... again.-- Cedric
    I didnt mean to discredit JSF and the efforts etc all.
    But i dont think its the framework of the year. Its not there yet.
    Regarding people more interested in knowing JSF in a conference - Hey you know what when I go to a conference which I mostly is a paid event - I would also want to kmnow about what JSF is over struts or webwork. Because struts is around for long time, simple and easy to use. That goes for webwork too. JSF is new and so people want to know what it is. It will be great to out of those 50 some who raised their hand that they gave JSF a shot after that conference and really use it.
  51. Struts and easy to use[ Go to top ]

    Sorry to jump the gun here, but I would not call Struts easy to use.
    In fact if you confront someone without any struts background with the choice of following frameworks:

    Struts
    Turbine
    JSF
    Velocity Servlet
    (Those are the ones I am most familiar with)

    he probably will grasp Turbine JSF and Velocity way easier than Struts.
    Turbine due to its implicit mvc model with having Velocity as the controller mapping many things implicetly.
    JSF probably due to its more natural user interface approach and less xml and class clutter than Struts.
    And Velocity Servlet due to its easiness in MVC.

    The reason why everyone thinks Struts is easy to grasp is, because most people grew up sort of with Struts and do not see beyound the boundaries of this framework anymore.
    (Most people I knew who loved Struts were those who never
    had done anything in any other more extensive framework,
    like Tapestry for instance, but have had coded Struts only for years)

    Struts has so many problems beginning with outright confusing naming conventions for the xml entries, ending with basically everything pushed into xml which in many cases should have been in other places and too few value for the lines of code.
    (I often had the feeling that many things which saved one line of code in java ended up in two lines of code in XML)

    I do not want to bash struts here as such, it was groundbreaking for what it has done. But times have changed and it is time for Struts to adopt (which it currently is doing) or to move on. But having the old Struts way day in and day out is like riding a dying horse.
  52. Struts and easy to use[ Go to top ]

    Struts only for years)Struts has so many problems beginning with outright confusing naming conventions for the xml entries

    Of course you don't have to write XML by hand. There are some very nice RADs to write struts xml configuration files... ;-)
    ending with basically everything pushed into xml which in many cases should have been in other places and too few value for the lines of code.

    I don't agree. If struts is used properly you can save a lot of code. As this is true with any other framework of course.

    People in 2006 seem not to have yet understood the core of object oriented programming, that is: generalize the code.
    Programmers only want to write a lot of code, maybe using the good old cut-&-paste solution. This is a wrong habit because similar problems should be generalized and solved only once. Writing a generic solution for a problem and then use it with different parameters saves a lot of code. And this is what Struts helped me in doing since I adopted it.

    I'm not saying that Struts is the best solution ever... I just had to say that I didn't agree with what you wrote... that Struts is difficult to use and the fact that it's always bad to write 2 xml lines instead of 1 java line: in small projects that could be true... in huge projects those 2 xml lines could save 100 java lines. And... the code you don't write... has no errors! ;-)

    Cheers,
    A.
  53. Struts and easy to use[ Go to top ]

    Struts has design flaws like XML overuse, but it looks like JSF inherits all Struts flaws and adds more XML complexity.
  54. Struts and easy to use[ Go to top ]

    Struts has design flaws like XML overuse, but it looks like JSF inherits all Struts flaws and adds more XML complexity.
    At least you should try to use JSF a little before bashing it.
  55. Struts and easy to use[ Go to top ]

    Struts has design flaws like XML overuse, but it looks like JSF inherits all Struts flaws and adds more XML complexity.

    No, it doesn't - you should try it. One of the reasons I like JSF is because it removes so many of what I consider to be flaws in struts - you can use POJOs, and you don't have to inherit from struts classes (at least this was the case last time I used struts); the items you get from a form are converted to reasonable types; validation can easily be customised by the page designer; you get an easy event-based rather than action-based system, and finally, you get a lot less XML complexity - I find a JSF application far simpler to figure out than a Struts one.
  56. Struts and easy to use[ Go to top ]

    Yes, I am not JSF expert, I just read J2EE tutorial and tried to develop trivial prototype. I found a lot of redundancy like "managed beans" and "navigation" in XML. Probably it is the mater of taste, but I am not happy to edit this stuff.
  57. Struts and easy to use[ Go to top ]

    Yes, I am not JSF expert, I just read J2EE tutorial and tried to develop trivial prototype. I found a lot of redundancy like "managed beans" and "navigation" in XML. Probably it is the mater of taste, but I am not happy to edit this stuff.

    I just don't get the redundancy point that you are making with JSF? The fact that it does use drop-in 'components' and 'POJOs', any redundancy found would be your own fault in implementation. I don't know how much more abstract you can get with re-use. On top of that, JSF's proper model/controller separation allows *you* to choose where your business objects reside. If the faces-config doesn't appeal to you, you can have JSF work directly with your Spring beans, or throw XML out the window with Seam.

    -- Jacob
  58. Struts and easy to use[ Go to top ]

    You get it right, there are more things to drop than to use.
  59. Struts and easy to use[ Go to top ]

    You get it right, there are more things to drop than to use.
    Can you be more specific, or are you just spreading FUD?
  60. Struts and easy to use[ Go to top ]

    No, I am not spreading FUD, I am just not a JSF fan. Try to read this thread to find more specific information.
  61. Struts and easy to use[ Go to top ]

    No, I am not spreading FUD, I am just not a JSF fan. Try to read this thread to find more specific information.

    Seriously, if you want to see some cutting edge JSF stuff from a pure functional standpoint (not including the many cool component libraries), check out:

    JBoss Seam (JSF Annotated w/ EJB 3)

    Facelets (Templating/Component UI for JSF)
  62. Struts and easy to use[ Go to top ]

    No, I am not spreading FUD, I am just not a JSF fan. Try to read this thread to find more specific information.
    Neither I am a JSF fan, I'd pick Wicket instead anyday.
    Anyway, since you're just telling me that you don't have any information to base you opinions about JSF and I should look somewhere else, but I can't see anything on this thread to support your assertions, I guess you're just going with the flow as much of the other bashers are.
  63. The core of OOP?[ Go to top ]

    People in 2006 seem not to have yet understood the core of object oriented programming, that is: generalize the code.
    Programmers only want to write a lot of code, maybe using the good old cut-&-paste solution. This is a wrong habit because similar problems should be generalized and solved only once. Writing a generic solution for a problem and then use it with different parameters saves a lot of code. And this is what Struts helped me in doing since I adopted it.

    Well, generalization (at least in the form you described) is not restricted to OOP, but also applies to imperative/procedural and functional programming. One may call it "good programming style", but it is definitely not the "core of OOP". What you described is "only" generalization (in the sense of abstraction) achieved by parameterization.

    Regards,
        Dirk
  64. The core of OOP?[ Go to top ]

    One may call it "good programming style", but it is definitely not the "core of OOP"

    Of course... I expressed myself in a bad way. Sorry.

    I mean... generalization can be used in ASSEMBLY, too. Take some effort and it will work.

    OOP is supposed to make things easier with class hierarchy and patterns like the Command Pattern or the Strategy Pattern.

    Thanks for correcting me,
    A.
  65. Struts and easy to use[ Go to top ]

    People in 2006 seem not to have yet understood the core of object oriented programming

    And as long as they are keep using frameworks like Struts, that wont change much either! Sorry for that sarcistic joke, no flamewar intended ;)
  66. Struts and easy to use[ Go to top ]

    People in 2006 seem not to have yet understood the core of object oriented programming
    And as long as they are keep using frameworks like Struts, that wont change much either! Sorry for that sarcistic joke, no flamewar intended ;)

    Don't worry... I'm not a flamewarrior. ;-)

    What I wanted to say is that it's a matter of fact that programmers still prefere to cut-&-paste code rather than writing generic and configurable classes and libraries as the OO paradigm suggests.

    Even when they're give powerful instruments such as JSF, Struts or anything else (I won't list 100 frameworks here) they still use them at the 10% of their possibilities. So it doesn't make any difference if the use a framework or they just use raw JSP. I'm speaking about web based apps of course.

    Sometimes I read posts like "I've been using JSF/Struts/Cocoon/Barracuda/Tapestry for one week and it sucks!!!". Well... you get the point. I needed 1 year to become fully confident with all the features in Struts and believe me... I'm not a clueless programmer.

    I'm actually building *cutting edge* applications with Struts, so I find difficult to understand people who claims that Struts is dead. It's alive and has still much to say! ;-)

    Cheers,
    Alex
  67. Ant The Utility of The Year...2005?[ Go to top ]

    Admitedly, Ant is a great tool which I can't live without, but that's exactly the point here - if a utility that has been around for 6 years, that has become taken for granted on almost every Java project, and that only made 3 bug fix releases throughout 2005, would be voted the utility of the year, I'm not sure that's really a nod to Ant, or rather mocking the creativity of entire Java community.
  68. Ant The Utility of The Year...2005?[ Go to top ]

    Admitedly, Ant is a great tool which I can't live without, but that's exactly the point here - if a utility that has been around for 6 years, that has become taken for granted on almost every Java project, and that only made 3 bug fix releases throughout 2005, would be voted the utility of the year, I'm not sure that's really a nod to Ant, or rather mocking the creativity of entire Java community.

    Actually point taken on thise one, I´d rather give maven or maven2 this awared, ant is too old for the award now.
    But this is the usual thing with such awards you never will agree to the outcome.