Hi Sergiu,
On Tue, Apr 7, 2009 at 6:57 PM, Sergiu Dumitriu <sergiu(a)xwiki.com> wrote:
Jean-Vincent Drean wrote:
On Mon, Apr 6, 2009 at 12:35 PM, Vincent Massol
<vincent(a)massol.net>
wrote:
Hi,
* We consider the 2.0 release as a normal release that we release
early September:
- 2.0 M1: 22 June 2009
- 2.0 M2: 13 July 2009
- 2.0 M3: 3 Aug 2009
- 2.0 RC1: 17 Aug 2009
- 2.0 RC2/Final: 31 Aug 2009
+1
* We add a new skin
I'm not sure about this:
1/ We've spent more than a year stabilizing toucan and we still have
issues on some browsers/OS (namely linux).
2/ We lack resources to work on UI related stuff.
3/ XE 1.0 was albatross, it seems logical to me that our next major
release uses toucan.
Don't get me wrong, I'd like that we have a new skin but I think that
if we can spend time on skin development this summer we should focus
on making toucan better before 2.0 (css cleaning, bug fixing, drop
albatross).
WDYT ?
This is not UI related stuff. This is not about a new flashy cool skin,
but about having a usable and reusable skin. Toucan is much to fragile
to fix. The prioritized list of points that need to be addressed is:
- Reusable skin, with not so fragile CSS rules (changing something
breaks something else right now) and easy to customize parts. Building
on top of the Toucan is close to impossible. This is something that our
custom development team would really love, and something our users are
complaining about.
What does it take to improve the skin without breaking
backward-compatibility and making migration nearly impossible for existing
projects? More specifically:
- Would writing a new skin imply the rewrite of some of the velocity
templates or CSS only?
- Would the underlying HTML structure of some pages / elements need to be
changed?
- How can we ensure that the CSS rules we put in a skin can be easily
tweaked & built upon by custom skins?
Basically, what's the best way to build something better both from the
technical (create new skins on top of it) and user-experience (make it easy
for users to perform actions) point of views without dooming existing
projects and sending them through an hazardous upgrade path? Is it necessary
to create a new skin or can we achieve such an objective by carefully
refactoring toucan?
- Usability improvements. While this can be done
incrementally inside
Toucan, its fragility makes it very hard to introduce such changes.
- Bugfixing and ensuring a uniform aspect. Same as above.
The fact that we've spent more than a year on stabilizing a skin that is
still far from being stable says something to me. And that's not that we
should use it more. The scroll problem on Linux is not it's biggest
problem, and definitely not the only one.
Well, in all fairness we never really took the time required to clean it up.
Laurent was pretty busy on other projects and couldn't spend enough time
working on it. We're already re-building a number of things from scratch and
I wouldn't want us to do it if it can be avoided in this case. Maybe we'll
have to, but we need to be sure we exhausted other options first.
Additionally, I'd be really glad to see the skin prototype Marta & you have
been working on to assess whether it could provide a sound basis for a new
skin shall we choose to pursue this path ;-)
Guillaume
--
Sergiu Dumitriu
http://purl.org/net/sergiu/
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs
--
Guillaume Lerouge
Product Manager - XWiki
Skype ID : wikibc
http://guillaumelerouge.com/