On Sep 3, 2009, at 1:48 PM, Ludovic Dubost wrote:
Vincent Massol a écrit :
> Hi,
>
> In order to prepare what we could in the 2.1 roadmap on the domain of
> architecture features (ie not user-oriented features), here's a list
> of stuff I have in mind for working on myself (and/or with some of
> you):
>
> * Services Bridge (between scripts and components)
needed asap
> * Velocity JSR223 implementation
> * Finish xwiki-url
> * Finish xwiki-action
> * Finish xwiki-localization
needed asap
> * Generic Rendering Markers for Transformations
needed quickly to support emoticons for example, for the footnote
macro and for other use cases we have listed (camelcase, etc).
> * Move XHTML parser from wikimodel to our code
base
needed to fix lots of wysiwyg bugs
> * Rendering: make script/include/useravatar macros
usable
> standalone too
> * Component Manager: introduce Realms
needed for wiki macros to work in multiwikis
> * Move to JSR299 annotations + move to Guice. (I
think this one is
> too
> early though - JSR330 public review is for 15th of Sep)
> * I10n of Renderng module / WYSIWYG module
needed asap for the i10n of the rendering module for ex
* Implement
xwiki-search (rewrite using new archi) (We might want to
fix existing lucene plugin first before tackling this one)
* App Manager (too big to start now - I think beg of next year)
The App Manager might be big, but this is more a reason to start early
than to push back.
It's mostly a question of resources. It requires someone full time on
it and you can't just play with it and go back to something else and
come back to it later. I tried that for some time and it didn't work
(since the domain is too complex to be able to do this).
If it's big and it takes 2 release cycle then we
are in mid 2010.
It's more likely to take 5-6 release cycles minimum. However there
could be some first minimal version in 2-3 cycles indeed.
I'd prefer we start it now to have it in March
2010
Me too. But it's not about having it or not having it. We all want it
ASAP. It's about who fixes the other important blocking architectural
problems (that will block several people if they are not fixed). We
need to close the loose ends we've opened by doing the big rewrite of
rendering/wysiwyg + component architecture first before we can tackle
a large topic again IMO.
Let's continue listing important architectural things that need be
done and we'll decide when we do the roadmap for 2.1 and beyond.
Thanks
-Vincent
Ludovic
> Any other stuff that we'd need to work on from an architecture pov?
>
> Thanks
> -Vincent