[news.eclipse.technology.kepler] Re: Proposed Kepler project plan on wik

[Scott Lewis <slewis@xxxxxxxxxxxxx>]

Hi Henrik,

Henrik Lindberg wrote:
> Hi,
>
> Here are some comments on the posted plan:
>
>
>
> I assume that the functionality covered in milestone 1 is mostly Buckminster 
> technology - it seems like it, as most, if not all of what is listed already 
> works in Buckminster.
>
> Would be nice if that was mentioned in the plan ;). 


Yes, this is an omission.

> Would also have liked to 
> have discussed milestone dates - Buckminster is after all on a release train 
> for Europa.


As a draft, there is definitely room for discussion and modification if 
these milestone dates don't work for all participants.  They were 
intended as straw man in order to begin discussion.

> I don't understand why we should use Mojo from Codehouse. If you think there 
> is something missing in Buckminster, we are very happy to discuss what 
> anyone thinks should be different. I have not looked at Mojo yet, but if it 
> covers functionality that is needed, I think we should discuss that in 
> public before stating that it should be in a certain milestone. Esp. as 
> there are also IP/Licensing issues involved.


Yes, it is just one candidate technology along with others.  It should 
probably not be in the project plan at this point.

> To me, the plan, as it now stands looks like Kepler is planning a rewrite of 
> technology already in Eclipse projects. Which is not a good move, and 
> something we have already agreed not to do. See meeting notes on the wiki: 
> http://wiki.eclipse.org/index.php/11.08.2006_Conference_Call) where we 
> agreed:
>
>
>
> -          Kepler should integrate existing tools rather than recreate them


Yes.  Kepler will absolutely not plan (or do) a rewrite of technology 
already in Eclipse projects.  The draft doesn't state this and so 
creates a wrong impression.


<stuff deleted>
> I think those agreements should be reflected in the project proposal. 


Agreed.


> Now, 
> milestone 1 focuses on building (and in a way that looks like a rewrite that 
> excludes the existing projects).


No, this won't be the case.  Even if someone wanted to (and I don't know 
anyone that does), doing such a thing would not be practical.


> I also wonder why Continuum is first on the list of continuous build systems 
> to support. There is a lot of support for Cruise Control among Eclipse 
> projects.  If we want to be successful among the Eclipse projects, shouldn't 
> we support technology that is actually being used rather than try to force 
> something new?


Yes.  Continuum is simply a candidate technology, available for use by 
Kepler.  That's how it should be stated.

Thanks for comments.  The draft will be updated.

Scott