Thanks for highlighting that work item; although it's specific to Mylyn and not quite the same question.

The reason I raised it here rather than as an RFE is that it's a significant decision to either ignore or to decide to remain separate from other existing open source initiatives with similar goals. Rather than raise an RFE, which for me is a demand for extension/compliance, I would initially rather that those making these decisions provide the arguments that have helped get Jazz to where it is. It could be that there are some very strong and convincing arguments why ECF and Mylyn are completely inappropriate and I would love to hear them.

That said, there is an overlap and I believe that some convergence would be good for the future of the product. I feel this is more clear cut with ECF than Mylyn, but even Mylyn would offer some real good value add to the Jazz offering (specifically the context and the hiding of irrelevant artefacts). Although I appreciate that Jazz has a lot more to offer, for me at least, the similarities are clear.

jlemieux wrote:
"To be clear, Mylyn and Jazz only have a small overlap in functionality, namely in the project planning area. Besides that, they are complimentary and whose objectives are different. Mylyn provides context based filtering and a generic bug system integration. Jazz actually implements a bug repository, scm, planning, and so on. We demo'd a Jazz/Mylyn repository integration at EclipseCon this year and have plans to make this code available for beta 2. Mylyn provides a good UI for those working with multiple different bug systems and want to use the same UI for managing the bug lists from all of them."

Jean-Michel, thanks for the feedback on this. I appreciate that they are not identical and that whilst the overlap is in the task driven nature of working, both TC/Jazz and Mylyn have a reasonable amount of 'value-add'. I guess, since they both have this core similarity, I'd have liked to have seen this shared between the projects; although I appreciate it could be a difficult thing to achieve and wholly dependent upon how the Mylyn plugins are broken up.

On the assumption that Mylyn cannot be used as the basis for the task driven functionality within Jazz, it's good to read that Mylyn integration is being developed, do you have more information on this (e.g. a wiki page)? I guess what you're describing here is a Mylyn connector that allows the Mylyn functionality to connect to the TC/Jazz SCM and Task repository?

Thanks for the feedback on Mylyn.

Anyone working on Collaboration/IM willing to comment on ECF?

Jean-Michel,

I'll say this again, the task list support in Mylyn is essentially a wrapper which allows abstraction from the concrete bug/scm system you are using. This is great when you are using tools which lack the integration and rich client integrations. But in Jazz, it's already integrated and as you've pointed out, because Jazz is process enabled, we can do more.

Whilst you're not wrong in what you're saying, in some respects you're missing the key value of Mylyn - that is the ability to allow you to 'focus' on the task at hand. This is so much more that merely providing links to SCM and bug repositories from within the IDE; it's achieved through information hiding - for example, you only see Java classes relevant to the work item being fixed; when you change context to work on a different work item the filters automatically change - this helps you to focus on the important aspects without being 'blinded' by all many and irrelevant artefacts.

I'm not trying to detract from what Jazz provides as such - merely that Mylyn provides more (and I also appreciate less, especially in the process space) - they are complimentary and hopefully (especially in the long term) not incompatible.

There is also a (very) brief discussion about this in work item 29737.