| Re: [ide-dev] Eclipse Languages Symposium/IDE Project? |
|
Thanks John, you're definitely correct about the overhead. And, yes, the Eclipse project is probably the rightful home for common language frameworks along with platform debug, resources, ltk, etc. An e4 type thing might be the right way to go.
BTW, frankly, generalizing JDT may not be the right approach. We started the CDT by cloning JDT and there's almost nothing left of that. For a lot of *DT's it's the wrong place to start. I was into JSDT lately and it's just truly bizarre how much cloning
is in there (JavaPackageCompletionProcessor?). I'm not sure how they keep things straight. But that's IMHO and I have to think there are better approaches.
Doug.
From: John Arthorne <John_Arthorne@xxxxxxxxxx>
Reply-To: Discussions about the IDE <ide-dev@xxxxxxxxxxx> Date: Monday, 26 August, 2013 4:42 PM To: Discussions about the IDE <ide-dev@xxxxxxxxxxx> Subject: Re: [ide-dev] Eclipse Languages Symposium/IDE Project? From my experience creating new projects I have almost always ended up regretting it. Creating a new project introduces a perpetual process burden that never seems to go away. You end up with a different set of committers
which creates a barrier to contribution, and someone has to lead and run the project for the rest of its life. If the project is indeed an entirely new initiative with no natural relationship to any existing project then it makes sense. In this case it sounds
fully within the scope of the Eclipse project, whose primary goal has always been an IDE platform. If someone wants to create and lead a new project, knock yourself out, but I think this work would be most welcome in the Eclipse project. In fact a clone of
JDT was created in e4 years ago the last time the "let's generalize JDT" topic was popular [1, 2, 3]. Everyone who expressed interested at the time was immediately given commit rights, but not a single contribution appeared. It could at least provide a "low
process" home for some early experiments that could later be moved into a new project if that made sense.
[1] http://git.eclipse.org/c/e4/org.eclipse.jdt.core.git/ [2] http://blog.vogella.com/2010/06/21/eclipse-e4-jdt/ [3] https://bugs.eclipse.org/bugs/show_bug.cgi?id=36939 John From: Doug Schaefer <dschaefer@xxxxxxx> To: Discussions about the IDE <ide-dev@xxxxxxxxxxx>, Date: 08/26/2013 12:52 PM Subject: Re: [ide-dev] Eclipse Languages Symposium/IDE Project? Sent by: ide-dev-bounces@xxxxxxxxxxx I meant JDT is tied closely to the Eclipse SDK which is managed by the Eclipse PMC. At any rate, Tools is the likely top level project for all this but we'd need to create a new front-line project. And maybe LTK becomes a component of that. But then that also brings up the point that the Platform has been the home of these types of things in the past. I just think it gets lost there with all the UI framework and RCP type stuff. IMHO. Interested in opinions on that (and be constructive ;). Doug. From: Doug Schaefer <dschaefer@xxxxxxx> Reply-To: Discussions about the IDE <ide-dev@xxxxxxxxxxx> Date: Monday, 26 August, 2013 12:37 PM To: Discussions about the IDE <ide-dev@xxxxxxxxxxx> Subject: Re: [ide-dev] Eclipse Languages Symposium/IDE Project? JDT in Tools has been talked about for a long time, but it's also tied closely to the Eclipse IDE, so it's never happened. Also Tools is too big and the mandate isn't as clear as I thought it was. I'd rather leave that as secondary. And, yeah, LTK is a component. Not sure it's big enough to be a project on it's own. From: Mickael Istria <mistria@xxxxxxxxxx> Reply-To: Discussions about the IDE <ide-dev@xxxxxxxxxxx> Date: Monday, 26 August, 2013 12:31 PM To: "ide-dev@xxxxxxxxxxx" <ide-dev@xxxxxxxxxxx> Subject: Re: [ide-dev] Eclipse Languages Symposium/IDE Project? So, I wonder if we should be considering forming an Eclipse Project around all this. And silly enough, calling it the Eclipse IDE project so we can keep using this mailing list and the web site (which is Coming Soon, BTW). We could cover this cross-language framework, an Eclipse IDE installer that Mike M was thinking of, and anything else IDE related that no one else is doing. Plus form a central gathering post for all the related projects (kinda like CDT is with PTP, TM, and LinuxTools). Or is this too much? IMO, LTK is the right place. But it seems currently to be a Platform Component. It would make sense to move it as a project as part of the "Tools" container (it would also make sense to do the same for JDT by the way). -- Mickael Istria Eclipse developer at JBoss, by Red Hat My blog - My Tweets_______________________________________________ ide-dev mailing list ide-dev@xxxxxxxxxxx https://dev.eclipse.org/mailman/listinfo/ide-dev |