AspectJ Moves From PARC to Eclipse.org

Discussions

News: AspectJ Moves From PARC to Eclipse.org

  1. AspectJ Moves From PARC to Eclipse.org (10 messages)

    PARC has decided to transfer AspectJ to an openly-developed eclipse.org project. This project will include documentation, web site, mailing lists, bug database, and sources for the AspectJ compiler (the primary tool produced by the project).

    The AspectJ team also commented:

    "We see Eclipse as an excellent new home for core AspectJ technology development -- it is an active community of Open Source development and innovation in the Java space. Once development moves to Eclipse.org, others will be able to contribute more easily. Similarly, SourceForge.net is a great home for the smaller parts of the development that will be hosted there."

    It will be really interesting to see how things progress when AspectJ is being developed as an open source project.
    Hopefully it will flourish.

    Read about the AspectJ move to Eclipse.org

    I wonder what the core reasons for this move are? Did PARC see this as the best way to move forward? Were they getting strained for time to focus on the project?

    Threaded Messages (10)

  2. AspectJ Moves From PARC to Eclipse.org[ Go to top ]

    AspectJ has always been an open source project. As eclipse conquers the world, it's nice to see AspectJ tagging along.
  3. Have you tried doing professional server side development on Eclipse? Have you tried creating taglibraries, servlets, filters, listeners, jsp, jstl, and others? Eclipse in my opinion has not much to offer - as of today. I still use NetBeans - even though it has a larger memory footprint, it has much more of the important features needed for rapid professional development.

    IMHO, if you are a java newbie and plan to write Hello World type of classes only - it would be ok to use Eclipse. Eclipse is like notepad on antibiotics.

    But, if you plan to write serious serverside or client side programs - NetBeans has more to offer. Eclipse rule the world? Keep dreaming.
  4. When was the last time you checked out Eclipse? Version 1? It's come a long way very quickly. I've been using it for a year now and would never go back to Netbeans. As you stated, Netbeans has a large memory footprint, but also, Netbeans uses Swing. With SWT, I get native graphics on Windows and Linux. It's faster, looks better and is more responsive.

    I can now code Java, C++ and C# all in the environment which is great, because I'd tired of having to learn new keyboard shortcuts and the likes for every different IDE.

    And with new functionality like AspectJ, I can easily get up to speed on newer technologies.

    The community is getting larger by the day. Won't be long before it is bigger than the Netbeans community. Tons of very cool plugins available. I can start a project with Eclipse, with UML class and sequence diagrams, and carry it all the way thourgh to deployment.

    Not being tied to anyone code repository is great as well. Some projects I use Clearcase, some I use CVS. Can't wait to get my hands on a Subversion plugin.

    I'm programming client apps and web application apps. With the code hotswap feature, I don't have to redeploy my app every time I make a small change. Time saver, which translates into $$$.

    IMO Eclipse rocks!

    Ben
  5. Mmm, I thought this was news on AspectJ, not a "what-is-the greatest-IDE" preachers' forum.

    It is clear that the drive and focus of the PARC team have been crucial to get AOP established on the current level. I don't think a similar focus and quality could have been achieved when starting from scratch in a fully open OS community.

    Moving AspectJ to one of the major open-source development consortiums at this moment in time could be beneficial as a bigger collaborator pool can be reached. And many seem to be interested to get AOP in the "mainstream" of dvp paradigms. AspectJ will certainly be one of the main facilitators for achieving this.

    As far as the choice of IDE/platform:
    No one can deny that Eclipse has made an explosive entry, mainly due to its high-quality offering with lots of attention to tool efficiency, a huge features-set and impressive ease-of-use.
    This is combined with a very transparent and well-controlled project management, which can not be said for all open-source projects.
    This offers at least some guarantees that AspectJ will not quietly fade out or be swamped by non-relevant and never-finishing change cycles.

    On the other hand, it will not help AOP if AspectJ suddenly becomes limited to the Eclipse platform, so the AOP community should hope the other main IDEs will continue to be supported as well.

    To add my penny's worth in the IDE-battle:
    The eclipse platform relies on the availability of plugins for extended/specific functionality, just as the other leading IDE platforms. Many of the things you need for "big-shot enterprise-level professional guru-level" development can already be found in one or more of the existing eclipse plugins (although many still in beta-level).

    But, contrary to e.g. Netbeans, the core platform just "feels right" for me (and for many other developers it seems).

    Even to such a degree that I am willing to temporarily miss some desired functions or use beta-level plugin implementations.

    Give the thing some more months, and anybody will be hard to convince to not seriously consider the eclipse family.

    Whether you want to avoid using a great (free) tool, just because it originates from a vendor you dislike, is of course a personal choice anyone is free to make. :-)

    Cheers

    Erwin
  6. AspectJ looking ahead[ Go to top ]

    Thanks Erwin for getting back on topic. I now use Eclipse, but beyond saying that comparing it to Notepad is laughable I have no desire to conduct IDE wars here (or anywhere else--each to his/her own).

    I think the _concepts_ of AOP are very powerful. I'm coming to think that AOP will eventually replace EJB in areas such as CMT. It's a more elegant and more general way of performing interception.

    I think the real question is whether AspectJ will ever be the most widely used implementation of AOP concepts. The JBoss 3.0 architecture is AOP-ish with its use of standard Java dynamic proxies. Rickard Oberg (the originator of the JBoss interceptor approach), Jon Tirsen (Nanning project on Sourceforge) and others including myself have gotten very good results with AOP without going beyond standard Java. Not as powerful as what AspectJ can do, but far beyond what can be done with standard EJB.

    My own experience of using AspectJ is that:
    - it's _very_ powerful, maybe too powerful. I think it could easily be abused. I can sense AspectJ antipatterns...
    - the syntax, although weird at first, makes sense before long
    - it's hard to sell to a team

    By open sourcing AspectJ, are Parc showing confidence in a great future for it, or getting rid of an incumbrance? Hopefully it will go the same way as Eclipse itself, which has flourished after it's kickstart from IBM.

    My gut feeling is that it would be great if a subset of AspectJ capabilities moved into core Java, although I don't see that happening, as the Java community is pretty conservative.

    Regards,
    Rod Johnson, author of Expert One-on-One J2EE Design and Development
  7. AspectJ looking ahead[ Go to top ]

    It's funny how people misinterpret your message. I had no intention to start an IDE war here (you can stick with notepad, or netbeans, for all I care). I was merely pointing out that AspectJ always had an Open Source license (MPL 1.1). As AspectJ 1.1 is already using the eclipse compiler, and there is a big effort in AJDT to provide first class IDE support for AspectJ, it's a natural progression for AspectJ to become more tightly coupled with the eclipse community. As for the ulterior motive for actually moving to eclipse.org, I guess only the PARC people would know.

    In any case, this can only be a Good Thing. The project instantly gains a larger, more active user community, and a better supported, better organized open souce infrastructure. And if one day eclipse does land on every java developer's desktop, it'll do wonders for AspectJ's widespread adoption.

    As for people who equate eclipse to IBM, I'll just have to say you are wrong and you are missing out something wonderful.
  8. For the ones still using NetBeans: U r all just too much into the Java/Sun ideology. Eclipse is already a way ahead of NetBeans and is progressing with MASSIVE speed. Respect.
  9. I think the bottom line of this move is that many of the people from PARC who have been working on AspectJ are moving on to something else. The emacs and other versions will still be maintained but who is going to do this? Currently I'm seeing IBM stepping up as with so many other Java activities today.
  10. AspectJ Moves From PARC to Eclipse.org[ Go to top ]

    In my opionion eclipse is a far way from being a good server side multi purpose coding environment. I use IDEA 3.0 to this end and I am perfectly happy. Its not open source but in the end I am quite sure that this does not matter much, since in the end, it's not "eclipse" but "IBM conquer the world". I am not so sure that this is what I want..
  11. AspectJ Moves From PARC to Eclipse.org[ Go to top ]

    I can now code Java, C++ and C# all in the environment >which is great, because I'd tired of having to learn new >keyboard shortcuts and the likes for every different IDE.


    Hmmm... What IDE is not supporting the shortcuts from others? Usually it's configurable in IDE-XXX to use shortcuts from IDE-YYY, or even to assign one's own shortcuts at will.

    >Not being tied to anyone code repository is great as well. >Some projects I use Clearcase, some I use CVS.

    Usually, an IDE _does_ have a plug-in for this-or-that versioning system. It may not be for free, but...

    Goran.