Discussions

News: Jeet AOP framework 0.9.7 released

  1. Jeet AOP framework 0.9.7 released (33 messages)

    Jeet is new proxy-based AOP framework.

    Features:

    * intercepting method invocations and applying the advices
    * introductions (mixins)

    Overview:

    - proxy-based (CGLIB2)
    - pure java, no language extensions, no external tools needed
    - simple and intuitive
    - xml driven configuration, but also direct, programmatic usage
    - j2ee friendly, no need to class reloading
    - supports AOP Alliance API
    - JDK 1.5.0 annotations support
    - a lot of fun ;)

    Read more at: http://jeet-aop.com

    Threaded Messages (33)

  2. Jeet AOP framework 0.9.7 released[ Go to top ]

    Congrats guys.
    I am always impressed by this willingness of implementing YAPBAOP (YetAnotherProxyBasedAOP) instead of contributing to some already existing ones.

    When Bob Lee shipped dynAOP I thought it was a good place to see some competition around and have Spring AOP evolve.
    dynAOP was innovative from the configuration point of view.
    Spring AOP is good from its out of the box Spring integration and thus implicit user base.
    Blockbusters (AspectJ, AspectWerkz, JBossAOP) all brings something interesting to the picture.

    Then I have seen joyAOP (how many of you haven't even heard of that one ?).
    And now we have Jeet AOP. What value do you bring to the picture ? Could you give me some selling pitch when I am about to make my mind (I am not, but that s a good exercise !).

    Alex
  3. please don't atack YAPBAOPs ;)[ Go to top ]

    First of all Alex I would like to say that you have missed totaly.
    There is no place where I am saying Jeet is better than the other
    frameworks and that is definetely NOT my goal to be top AOP tool, what
    would be nonsense and is impossible anyway. This framework is very simple,
    never thought to be mentioned together for instance with Spring or AspectWerkz. The vision of competing beetween
    AOP frameworks is not my vision. I have respect for _all_ AOP people, that's a pity that you don't.

    Why another framework? And why thousands of new tools are created around ? Why new O/R mapping tool, we all know that Hibernate rules, new IDE? Why? Could be anything better than Eclipse? etc. etc.

    Jeet is my code, my (first) contribution to OS world and I believe it can
    help many people understand AOP better (as every AOP framework does). Please don't expect too much from
    this announce :) My goal was to create YAPBAOP ;) and use it in production system. I succeeded. And that's all.
  4. please don't atack YAPBAOPs ;)[ Go to top ]

    I apologize if I was not clear enough.

    I don't aim at offending what you have done, but wanted to hear why you and the committers in Jeet AOP decided to go by your own instead of helping out one or more team that do need help (AspectJ, AspectWerkz, JBossAOP and Spring have claimed more than once that they are willing to accept any contributions and they all already have).

    If you bring to the table something really new, then you indeed have two choices: have it grown or contribute it within a project. Probably a mix of both.
    I was just trying to understand what was new in Jeet AOP, that could f.e. be transferred with all due credits to your (one or more) favorite partner project(s) once you get the innovation mature enough.

    I do respect all AOP people out there despite what you may say, and you don't need to write your own YAPBAOP to be part of it ! I think that it has started to be much more valuable to contribute a feature, a patch, a reusable aspect, a good article, tutorial, book for any of the framework available out there and would be pleased to see much more initiatives in that area.
    I am willing to hear about others' feelings on that.

    By the way, which team would you like to join Przemek ?
    Alex
  5. please don't atack YAPBAOPs ;)[ Go to top ]

    Ok, now I know your point of view. I even understand what you mean saying about contributing to AOP and
    I can imagine it's disappointing to see new frameworks instead of people willing to involve in your project (which is very cool btw). You are totally right here, but... it is not so simple as it looks. Almost every project started in the same way, with the diffrence
    about what happend next :) Some of them were succesfull and other failed and disapeared.

    I repeat again: I didn't want to create anything extraordinary, what would be better, diffrent, I didn't plan to invent the
    new way to do aspect programming etc. etc. Please notice that probably about 90% of developers even don't use AOP at all, it is still new subject for the most of them, not at TSS community of course, but for example in Poland (where I live) definetely yes. I either didn't want to meant like this "hey guys this is my super framework! I am coming!" ;),
    but rather to tell to other developers about what I've done, just to comment, maybe help, maybe contribute. Your opinion is very important for me even if you are criticizing it. I appreciate it really. Thanks a lot. Man, we just talk. That's way I have announced it.

    What is new in Jeet? Simply: nothing (yet :)) . It is very easy to use, its API is nice. No ideology inside :)

    Join other framework? Why not. You know what ? It is funny, because I have never thought about it :)))) Which team? Question is difficult. I played a bit with instrumenting bytecode on the fly and I have to say that I like
    proxy-based approach much more.
  6. Ok, now I know your point of view. I even understand what you mean saying about contributing to AOP and I can imagine it's disappointing to see new frameworks instead of people willing to involve in your project
    I claim independence from any vendor and project. As a plain user (occasionally submit small bugs, bugfixes and suggestions) I am disappointed to see YetAnotherSomething because I would like to work with nice and complete frameworks and libraries, not with half backed promisingly looking ones. We all (as a humanity) will be better off if we will collaborate more.
     Join other framework? Why not. You know what ? It is funny, because I have never thought about it :)))) Which team? Question is difficult.
    Yes it is. Sometimes personality(ies) of a particular project team might discourage participation and that might be a good reason to start a new but the same project.

    PS: I am sooo glad to see anouncements like the following on some sites : Development has stopped and users of YYY project are advised to migrate to ZZZ.

    There is no shame- just a sign of maturity, intelligence and modesty.
    I played a bit with instrumenting bytecode on the fly and I have to say that I like proxy-based approach much more.
    Did you check out AW proxies?
    http://aspectwerkz.codehaus.org/new_features_in_2_0.html#AWProxy:_proxy_awareness_of_aspects
  7. Why YetAnotherXX is wrong[ Go to top ]

    I repeat again: I didn't want to create anything extraordinary, what would be better, diffrent, I didn't plan to invent the new way to do aspect programming etc. etc.
    Sorry, could not resist adding Daniel's allegory of 'lined paper' to the "Why YetAnotherXXX is wrong" line of reasoning.
    http://www.ishmael.com/Education/Writings/technology_and_other_war.shtml

    You are writing along the lines - go sideways, aim higher!
  8. Why YetAnotherXX is wrong[ Go to top ]

    Konstantin, i think you are the one who misses the point here. The sole reason - joy of development.
    Ever seen those people who make small toy elecrtic cars and participate in race ? Or small toy planes ?
    Why do they do that ? Why do they waste so much time and effort on absolutely useless stuff ? Why do not they participate in big Honda, or Toyota projects ?
    Because they want to be designers, they want to do intresting work not boring work. And the only way they can get that - start their own project.

    You are advising to join exisiting OS project like Spring or AspectWerkz. But what he will do there ? Update documentation ? write examples ? hunt bugs ? It's boring. And you never get the credits. Everybody knows Rod Johnson. But nobody knows those who updates Spring documentation or samples. All intresting work is already done.

    You are missing the beauty of OS movement. Doing something intresting yourself, being a designer, creator. Even if you are a designer of a small toy car or plane.
  9. Why YetAnotherXX is wrong[ Go to top ]

    The sole reason - joy of development.
    I do not miss it, just want that development and associated joy was redirected to more useful things.
    You are advising to join exisiting OS project like Spring or AspectWerkz. But what he will do there ? Update documentation ? write examples ? hunt bugs ? It's boring. And you never get the credits. Everybody knows Rod Johnson. But nobody knows those who updates Spring documentation or samples. All intresting work is already done.
    Bingo: causes for ‘reinvention’ lie in psychology, desire to be famous, rather than pure joy.
    Still I suggest cultivating modesty and look for ways to apply AW, HiveMind whatever to create something useful and yes, occasionally hunt bugs, update documentation etc. I do discourage ‘wheel reinvention’ and fail to see how repeating the same thing all over again is joyful.
    Because they want to be designers, they want to do intresting work not boring work. And the only way they can get that - start their own project.
    You are missing the beauty of OS movement. Doing something intresting yourself, being a designer, creator. Even if you are a designer of a small toy car or plane.

    Most pretend to be designers like kids painting by numbers imagine being painters.
    I put myself in a difficult position intentionally: instead of encouraging people to do what they “want”, I try to make them do that they really ‘need’: grow professionally and emotionally, see things in broader perspective, ‘be’ rather than ‘pretend’.
  10. Why YetAnotherXX is wrong[ Go to top ]

    I do not miss it, just want that development and associated joy was redirected to more useful things.

    Joy of development is missed when you "redirect" it forcefully to something developers do not want to do (like writing documentation, or hunting bugs). In this case you have to pay them to do the job.
    So please, if you want them to do something for you - give them something instead - joy of development or money.
    If you cannot, then stop whining.
    Bingo: causes for ‘reinvention’ lie in psychology, desire to be famous, rather than pure joy.

    Why do you oppose those reasons ? being famous is a big part of joy of development.
    I put myself in a difficult position intentionally: instead of encouraging people to do what they “want”,
    LOL. Do you pay them money ? If not then you can do that boring work yourself.
  11. Why YetAnotherXX is wrong[ Go to top ]

    Bingo: causes for ‘reinvention’ lie in psychology, desire to be famous, rather than pure joy. Still I suggest cultivating modesty and look for ways to apply AW, HiveMind whatever to create something useful and yes, occasionally hunt bugs, update documentation etc. I do discourage ‘wheel reinvention’ and fail to see how repeating the same thing all over again is joyful.

    You still know what is the best for us ;) gee, your cultivating modesty and words of wisdom should lead the most of linux developers to despair. 50% of linux OS software imho has poor quality. Why those people don't want to consolidate together and work on one decent project? Why we have tens of mail clients or web browsers? Is creating a new file manager reinventing the wheel?
    Most pretend to be designers like kids painting by numbers imagine being painters. I put myself in a difficult position intentionally: instead of encouraging people to do what they “want”, I try to make them do that they really ‘need’: grow professionally and emotionally, see things in broader perspective, ‘be’ rather than ‘pretend’.

    Let me ask you what usefull does this disscusion give to you personally? Do you really need to comment announcements at TSS? Is it anything new, what you do ? Probably it is the same level of usefullness as creating YetAnotherXX :) You just express yourself, just why you do this I think. So do I, so do many YetAnotherXX makers. Is it desire to be famous? Maybe, but is it wrong? I think it is not.
  12. Why YetAnotherXX is wrong[ Go to top ]

    Joy of development is missed when you "redirect" it forcefully to something developers do not want to do (like writing documentation, or hunting bugs).

    I am in no position to impose my will, nor I want to enforce it, nor I believe that such redirection is enforceable. It should come naturally.
    I simply appeal to intelligence “”

    In this case you have to pay them to do the job.
    Do you seriously believe that bright people can do unwanted job for long?

    Why do you oppose those reasons? being famous is a big part of joy of development.
    I do not oppose, I encourage aim higher and get ‘famous’ by inventing something truly unique and different.
    Could you explain how is in your mind somebody can get famous by “wheel reinvention”?

    Let me ask you what usefull does this disscusion give to you personally?
    I express myself and I hope that somebody will listen to reasons and will contribute efforts to really different ideas and projects. That will lead to better quality of projects and I(and others) will use high quality components and finally will have joy instead of facing constantly rough edges, bugs, less than convenient configurations, missed features etc., those things kill joy very effectively :)

    Also such posts help discovering likeminded people, which might one day create a team.

    As you see my agenda is completely selfish - no secret about it.
    your cultivating modesty and words of wisdom should lead the most of linux developers to despair.
    Hope so.


    PS: big part of OS is about (re)using common parts and in the process bugs get discovered even without explicit 'hunt'.
    PS2: Vagif, I guess you mistake “writing CYA documents in a highly bureaucratic environment” for “writing necessary project documentation”. The latest is not that frustrating and quite useful even for writer him/her-self. Read more in “Agile Modeling: Effective Practices for Extreme Programming and the Unified Process” by Scott W. Ambler http://search.barnesandnoble.com/textbooks/booksearch/isbnInquiry.asp?userid=2X524RZD5A&isbn=0471202827
  13. Why YetAnotherXX is wrong[ Go to top ]

    Do you seriously believe that bright people can do unwanted job for long?
    No, that's why they start new projects instead of joining existing ones.
    It is from those countless failures such gems as Hibernate or Spring are born. Only those who do nothing never fail.
    But bright people do not start genius projects without any prior (most of the time bad) experiense.
    How do you expect someone who only participated in updating documenattion and samples in existing projects come up with something unique, if they never tried to do anything at all ?
    Do you think those famous projects we all know today, are their first attempt ?
    It is this endless process of "reinventing the weel" is what makes OpenSource sucessfull, because otherwise rational people like you would still be trying hard to to fix existing solutions like EJB.
  14. Even if 100 versions of the same software already exist, sometimes you just have to do it yourself to learn in depth.

    But, I don't understand posting it to TSS the moment you get something that sorta-kinda works. What is that about? Were we supposed to take it seriously and consider it for our projects? No? Then why announce it to the world?
  15. Why YetAnotherXX is wrong[ Go to top ]

    Is creating a new file manager reinventing the wheel?
    Look here: http://wicket.sourceforge.net/Introduction.html

    The page is clearly written in the same spirit that causes me to offer and promote my template.

    IMO that project clearly differentiates itself from others and therefore do not qualify as 'wheel reinvention' even it is YetAnotherXXX.
     
    Answering your question: if someone can come up with something like Wicket introduction and explain why it is necessary to create YetAnotherFileManager rather than improve existing one I will applaud, otherwise consider it being a child exercise.
  16. Why YetAnotherXX is wrong[ Go to top ]

    I cleary know what you are talking about from the very beginning of this thread. And I repeat it again: most of software of any kind is reinventing the whell according to that point of view. Anyway you still haven't made me
    think that one MUST always do only pragmatic things, all the rest is bad. Try to think as an artist not as a craftsman.

    But OK, let us assume it is right. Please guys stand up who feels afflicted by the fact that new yet another proxy based aop framework was born ;) There is something weird about this discussion.... It seems it is a kind of problem that another framework was announced. To me definetely not. Even if reinvented the wheel. I never thought about such a attitude to OS. Little bit surprised. Anyway I apologize everyone I hurt ;)
  17. Just to add my stone to the discussion, I am announcing the release of YAPB AOP 1.2.3.4 for Java.

    It is proxy based, simple, and runs out of the box.
    A real demo is part of the distribution and code source is available.

    http://blogs.codehaus.org/people/avasseur/archives/000934_yapb_aop_1234_released.html

    So who's next in the YAPB AOP arena ?
    Come on guys, lets bring something more valuable to the community.

    Alex
  18. Congrats! I am very impressed really ;) You spent 40 min. just to create AOP frameworks pastishe based on my project.
    It adulates me. You did it maybe also to prove yourself and others that you can make frameworks
    like mine in few minutes. Never doubt you were able to, indeed. Great man, but superfluous. Very 'nice' and very 'kind', for Shen Li from joyAOP too... You see, Konstantin wrote "Yes it is.
    Sometimes personality(ies) of a particular project team might discourage participation and that might be a
    good reason to start a new but the same project". He was damn right writing that.
    I think that bashing others isn't imo the best idea for inviting people to contribution. To me it works just in opposite way. Moreover playing like that (I mean posting such a ironic stuff) is very easy way to turn any value to null value :)

    EOT for me.
  19. interception frameworks and aop[ Go to top ]

    Long time ago i have noticed inside JBoss an interesting mechanism to provide j2ee container functionality: interceptors. I liked the idea. At that point I made a quick association between interceptors and aop. But this was kindda wrong. Reading the theory behind AOP you will understand this. Reading the discussions on JBossAOP first drafts and the aop guys you will see this.
    Now, about needing so many interception frameworks, I am not sure. What I am sure about is that addressing the problem as Konstantin said I will probably stop at the 1st (as most of them are providing exactly the same).

    cheers,
    :pope
  20. YAPB AOP 1.2.3.4 released ! [for real][ Go to top ]

    The same as Przemek, I wrote my own joyaop never aim to compete with the other ones, initially it was just my course project. I don't care how many uses it has, I don't interest in the bechmark, I just think that my framework is perform efficiently enough among the PBAOPs .

    Since writing a basic PBAOP framework is quite easy and aop is in a relative early stage, why not write one to meet your own preference and requirements :) IMO, most people don't really need aop (Yes, they need the system-level services, but the EJB containers and the framework specific interceptors could satisfy them, e.g. xwork, hibernate), and most aop users just need the basic functions. Then many of the rest ones love the joy of development.

    To me, no other OS frameworks meet my preference before. AspectWerkz: I obtain most of my aop kownledges from it, but it's too bothersome to me to startup the AW apps, and it's hard to debug, the exception trace is confused by the enhanced bytecode (Not sure) and no debugger as aspectj... Spring: i have to use it with the spring container, and it's pointcut config is too verbose to me... Dynaop: I don't like beenshell (but the others love it) and it's hard for me to adapt it... All in all, it's just personal taste.
  21. YAPB AOP 1.2.3.4 released ! [for real][ Go to top ]

    Shen, thanks for providing valuable feebdack on why you did JoyAOP. Obviously you had several reasons and it was not just for fun.

    This information is very valuable to DynAOP, Spring AOP, AspectJ and Aspectwerkz guys, since it helps to understand what s wrong in the current picture. And I never said the picture was perfect.
    This is the kind of feedback we are trying to get from Przemek ;-)

    Alex
  22. Alex, of course the reason of doing my own tool wasn't only the joy of development or will to be famous. I am a developer not a scientist. I can hardly experiment in my work. I am paid to have things done (and working stable). You can disagree, but imo every AOP tool is a little bit experimental stuff even mature ones like AspectWerkz (I mean no as a code, but as a whole: system + framework under several circumstances like application server environment). It is not so easy to apply AOP to existing traditional J2EE systems, especially with byte-code manipulation like approach (class loaders for example). This concerns AspectWerkz too. I had to choose the way of minimal interference to the existing system. No system reconfiguration was possible. Proxy-based AOP seemed to me as the only one solution I could use. Spring framework is great considering its features, but I don't use it and its API is huge (too huge for me :)). dynAOP as Shen wrote uses bean shell and to be honest doesn't seem to me as a very mature, featured and dynamically developed framework (althought its code is very clean and smart). Do I have to like it? :) I know about AW proxies now and I am very impressed, but when I started to work on Jeet it did't exist yet. So I think if we decide to use proxy-based AOP the ability of choice is very limited. I chose to start my own tool, just to make things in the way I like and to have control (on code I know). Maybe it wasn't the best decision. Anyway it's just happened :))
  23. Great

    That s absolutely right. Proxy support in AspectWerkz is very recent, and is there because we have seen an increasing interest around proxy based AOP. It is part of AspectWerkz 2.x which is indeed not a final release yet.

    I do buy the arguments you finally gave us. Be sure that I don't consider that there are so called AOP projects and so called dust projects. I just want to see more collaboration out there.
    I know that we are working on proxy AOP to have it better, that JBoss has some stuff around that, that Spring guys are brainstorming as well. Join the train and subscribe to the project list perhaps since it is likely that The YAPB AOP is actually not yet there despite the marketing I have done around this thread ;-)

    Alex
  24. YAPB AOP 1.2.3.4 released ! [for real][ Go to top ]

    Shen, hail YAPBAOP brother! :)
    The same as Przemek, I wrote my own joyaop never aim to compete with the other ones, initially it was just my course project. I don't care how many uses it has, I don't interest in the bechmark, I just think that my framework is perform efficiently enough among the PBAOPs.

    ...and that's what I was trying to explain, but some guys didn't want to listen to me ;)
    Since writing a basic PBAOP framework is quite easy and aop is in a relative early stage, why not write one to meet your own preference and requirements :)

    Forget. There is only _real_ AOP like AspectWerkz, AspectJ, JBossAOP, Spring, dynAOP and their poor, inferior clones as all the rest and probably all that will ever appear anytime and anywhere. Or take the second advice: do your own, but never announce it :)))
    IMO, most people don't really need aop (Yes, they need the system-level services, but the EJB containers and the framework specific interceptors could satisfy them, e.g. xwork, hibernate), and most aop users just need the basic functions. Then many of the rest ones love the joy of development.To me, no other OS frameworks meet my preference before. AspectWerkz: I obtain most of my aop kownledges from it, but it's too bothersome to me to startup the AW apps, and it's hard to debug, the exception trace is confused by the enhanced bytecode (Not sure) and no debugger as aspectj... Spring: i have to use it with the spring container, and it's pointcut config is too verbose to me... Dynaop: I don't like beenshell (but the others love it) and it's hard for me to adapt it... All in all, it's just personal taste.

    Exactly. It is just because imo there is no the one and only (right) AOP vision. If someone claims so, he is totally wrong then.

    Notice that we as a
    YAPBAOP makers are in a little bit more dificult situation, because even being modesty and explaining that our tools are in early stage and that we know they are simple and we often are followers of mentioned projects, we are attacked.
    I would never try to claim that I see things I dislike in
    those products (like Shen mentioned), probably I would be totally bashed :))

    Hey guys we know your AOP experience is deep, we totally respect your projects, but please (at least try to) respect us (just because of fact thar we try to invent something in AOP area), and do not take away from us the ability to create new projects. It seems a little bit sick to me :)
    AOP monopoly ? ;))
  25. YAPB AOP 1.2.3.4 released ! [for real][ Go to top ]

    Since writing a basic PBAOP framework is quite easy and aop is in a relative early stage, why not write one to meet your own preference and requirements :)

    This has been a very interesting thread, from the point of view of someone who has only recently discovered AOP!

    My view on this is that I am a strong supporter of standards (or at least, widely used existing libraries). Even if they are awkward or not a lot of fun to use, they make your code portable and much easier for others to support. If you write your own solution to a common problem, be it ORM or AOP or whatever, then anyone who has to deal with your code has to learn your solution! Sure, it's fun to do, but my opinion is that it's not a professional or useful way to code for anything but personal or experimental projects. I write from personal experience of having to untangle and debug large projects containing various do-it-yourself versions of common functions, often seriously buggy or unfinished.
  26. .. YAPB AOP 1.2.3.4 for Java.It is proxy based, simple, and runs out of the box.

    Alex, I'm having some problems with YAPBAOP in production. Is there a mailing list that I could post a question to?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  27. Why YetAnotherXX is wrong[ Go to top ]

    excellent reply! it is just what I was writing/thinking about.
    thanks Vagif
  28. Nope, Alex missed nothing. Please use the following template on the very first page of your project:
    1. I(we) had a need to do: <Two phrases to describe>
    2. I(we) have tried the following frameworks: <insert names> before starting to develop my own
    3. They are supposed to address my needs but I was not happy with them because: <reasons why those frameworks are not satisfactory>
    4. I(we) decided not to contribute to any of the existing open source frameworks because: <describe why>
    Some explanations here http://kgionline.com/annoying/openSrc.jsp
    Could be anything better than Eclipse? etc. etc.
    It is very possible to be better than Eclipse. But that is not the point, the point: be different! And that may grow to a better solution.
    I am big fan of CGLib myself and successfully use CGLib based ‘AOP’ (http://kgionline.com/articles/aop_1/aop1.jsp ) for a year now in the production but in the end of 2004 it simply does not make any sense not to use one of leading AOP frameworks and come up with YAPBAOP. I do not use my CGLib based approach in new projects anymore and clearly feel better because mature frameworks bring lots of useful bells and whistles, and of course they are better tested.
    Why another framework? And why thousands of new tools are created around ? Why new O/R mapping tool, we all know that Hibernate rules, new IDE? Why?

    To say bluntly: because their authors are probably too stupid to learn by example, or ignorant, or lack modesty, or simply not aware of similar solutions. Other explanations are possible too.

    Why engineers in other industries are REQUIRED to study existing solutions and improve upon them? Why it is not required in IT?
  29. Nope, Alex missed nothing. Please use the following template on the very first page of your project:1. I(we) had a need to do: <Two phrases to describe> 2. I(we) have tried the following frameworks: <insert names> before starting to develop my own 3. They are supposed to address my needs but I was not happy with them because: <reasons why those frameworks are not satisfactory> 4. I(we) decided not to contribute to any of the existing open source frameworks because: <describe why> Some explanations here http://kgionline.com/annoying/openSrc.jsp

    the way you see OS is very restricted imho. I think everyone has his/her right to create what he/she wants. I don't think that more projects will be disastrous to OS world in any area. to me every programming activity even with no material profits is the benefit for programmer. I think OS contribution of any kind is valuable not only contributing to the-only-one-and-politicaly-correct project(s). I see you want to be funny and to
    introduce something yours like theories on the page you mentioned but I disagree with you, no offence :)
    and fortunately there are my rules on my page, no templates Konstantin ;)

    <blockqoute>
    I do not use my CGLib based approach in new projects anymore and clearly feel better because mature frameworks bring lots of useful bells and whistles, and of course they are better tested.
    </blockqoute>

    I agree, but it still imho doesn't mean that creating new frameworks instead of contributing to existing ones is anything bad. Give people the freedom of choice :) Let them choose.
    To say bluntly: because their authors are probably too stupid to learn by example, or ignorant, or lack modesty, or simply not aware of similar solutions. Other explanations are possible too.Why engineers in other industries are REQUIRED to study existing solutions and improve upon them? Why it is not required in IT?

    hm... interesting ;) I agree a bit, but notice that one can do something just for pleasure. I give you the example. There are many good J2EE servers, also free ones like JBoss, Resin etc. why do Geronimo guys waste their energy to reinvent the wheel and create another boring buggy and less-featured server instead of helping JBoss team to improve their product? Are they ignorants? Are they stupid? Answer yourself please ;) I think because they love what they do. You sound bitter to me, really. Please think of it as a kind of pleasure. Some people code because they like, not only to have the job done. Not only to use the better tool.
  30. I disagree with you, no offence :)
    and fortunately there are my rules on my page, no templates Konstantin
    Of course you are welcome to disagree. I do not mind cause controversy. It is just funny: spend tons of energy and create a project ‘by template’ and not to use another template to explain why you have done that :)

    There are many good J2EE servers, also free ones like JBoss, Resin etc. why do Geronimo guys waste their energy to reinvent the wheel and create another boring buggy and less-featured server instead of helping JBoss team to improve their product? Are they ignorants? Are they stupid? Answer yourself please ;)

    IMO they are not stupid obviously :). I suspect strong political motivation and personal conflicts. I consider Jeronimo as a plain waste.
    Give people the freedom of choice :) Let them choose.
    Exactly. Do not like my template, use anything but explain please how your project is different from others. Help a person make decisions that is only intent of my template.

    You sound bitter to me, really.
    I am bitter because wasteful attitude is widespread and causes disappearance of species and leads to collapse of humankind, sorry for the pathetic. But that is why I dismiss usefulness of "contribution of any kind".
    Please think of it as a kind of pleasure. Some people code because they like, not only to have the job done.
    Sure. Count me here.
     Not only to use the better tool.
    Oops. That seems unnatural: It looks like people like to use best tool for the job and are in constant pursuit of it ( hope you have read http://www.pragmaticprogrammer.com/ppbook/index.shtml ) . Of course what different people consider ‘the best’ vary. But that is the point: create a different approach, wheel reinvention is useless.
  31. I see nothing wrong in YetAnotherXXX, if authors want freedom to innovate and to learn something new. OS development is a very good way to learn and to test some ideas and experiments in practice.
  32. AOP for dotnet[ Go to top ]

    Przemek, AOP stage for java is pretty much established. But dotnet is in dare need for such AOP framework now.
    Starting in dotnet would give you some headstart.
    Do you have any considerations about making a dotnet AOP framework ?
  33. Jeet AOP framework 0.9.7 released[ Go to top ]

    As soon as I saw the title, I knew that there was going to be a lynch mob going after the project's owner :)

    First there is the average lyncher who lynches for the sake of lynching.

    Then there is a creator of yet another XYZ framework who released theirs 9th in the list of yet-another-xyz frameworks. 9th is still good, because 15th sucks.

    There will be many duds in the list of yet-another-xyz frameworks but people use just about every single one of them regardless of their merits.
  34. People do keep using em![ Go to top ]

    I killed YAIL (Yet Another Interceptor Library) close on a year ago now, and have signs telling people to grab DynAOP, but people keep downloading it for some reason. Go figure.

    -Brian