Let's get on with it! (MVC etc.)

Right... It seems that the MVC semantics debate (or lack of debate) is
holding things up (potentially, anyway). So in the interest of getting
to a release, I've made the following decision: We keep the MVC
mechanism (including the Model mechanism) as Dallas implemented it
(possibly with bugfixes where needed) for release 0.2a0 and document
it as experimental. Also, the attribute semantics (in-place
modification vs. copying etc.) will be documented as undefined. (Or
enough of it will be defined for it to make sense to the common user.)
Of course, the part relating to the Model mechanism will be defined,
but experimental. You get the picture ;)

So, basically, I'll plug some stuff about this into the documentation
for now, and we can do another round of discussion before 0.1a1.
(Isn't procrastination fun? ;)

At least then we get to test the Model mechanism in a functioning
release.

So, as far as I can tell, the test suite stands as it is currently
defined in CVS (i.e. the subset mentioned in nondist/admin/status.txt)
so development should now be focused on finding/fixing bugs, and
making all back-ends pass the test suite.

Also, beefing up the test suite to probe the internals more (and not
just test functionality) would be a Good Thing (tm). I believe Samuele
had some comments about this, regarding Proxy etc. Samuele? Could you
take a look and either implement a test or two, or suggest some that
could/should be implemented?

So, let's rock and roll! ;)

-- 
Magnus Lie Hetland        Practical Python          The Anygui Project
http://hetland.org        http://ppython.com        http://anygui.org


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf