Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [wtp-dev] Eclipse 3.1.1 & WTP 1.5

Hi David, thanks for the in depth response..

We currently are targetting the July timeframe for our 2.0 release, and trying to minimize the amount of headaches upgrading several integration points entails (not to mention we'd like to avoid forcing our users yet another eclipse upgrade) ...

Our main reason for needing to move forward to WTP 1.5 will be the resolution of this bug:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=111545

Of course, we will enjoy the new bells and whistles too =)

That being said, has there been an agreed upon release where Webtools will be from "that point on" backwards compatible? Will webtools 2.0 be backwards compatible with Eclipse 3.2? etc...

Your comment about the server features is interesting, that definately will be one of the major components we are targetting for integration, but we will also be integrating all the JSP/XML/HTML/CSS/_javascript_ editors etc. I'm guessing these editors will pull in the tabbed-property editor changes you were talking about?

Maybe it would be feasible to identify all of the areas of API incompatibility with Eclipse 3.1 and outline them somewheres on the webtools project page so that integrators / consumers can just build in their own bridges without having to find out the hard way?

Thanks guys..

On 1/28/06, David M Williams < david_williams@xxxxxxxxxx > wrote:

Aside from the navigator, that's been mentioned already, there's other things in the 3.2 stream that we in WTP will need or want to move to . Tabbed property pages, improved exclusion filters, and (I think maybe) some subtle changes in EMF, some improvements in document threading, improvements in undo management. And .. I'm sure even other things I'm not even aware of.

Bluntly put, the problem with "minimizing the dependancy" and maintaining compatibility is that it takes extra work. Investigation if even feasible, if so implementations of "bridges", not to mention it would at least doubling the testing effort.

Because of that extra cost, I don't think we (in the core of WTP) are motivated to pursue.  We realize this may cause some to investigate the trade-off of staying on 3.1/wtp 1.0 or migrating to 3.2/wtp 1.5 .. but .. I dont' think there's a "business case" for us changing our current plan. Of course, As Arthur implied in another post, others are most welcome to investigate, write bridges, and do the extra testing. We certainly wouldn't "stand in the way" of that sort of effort (that is, we'd investigate bug reports/patches that made it easier for others to do), but there's only so much work we can contain.

Marshall, I would be interested in the motivation and specficis for your question ... I assume the timing or life cycle of jboss ide? Are there technical reasons of not moving forward on to 3.2/wtp 1.5?

BTW, if there are specific component-features you need to be compatible with 3.1, such as the server features, you might want to investigate that specific requirement, since server features are sort of isolated, and likely not impacted as much by the 3.1/3.2 differences. But .. I suspect it would be you that would have to do the investigation ... I have enough trouble getting our server committers to do smoke tests :).

All that said, certainly N-1 or N-2 backwards compatibility is our long term goal and philosopy .. but ... I personally do not thing we can afford that extra work this release.

Thanks for your continued support.



--
Marshall Culpepper
marshall.culpepper@xxxxxxxxx
JBoss Eclipse IDE Lead, JBoss Inc.

Back to the top