Re: Idea for backend maintainers: "reversal of roles"

Hellen Claggett <[email protected]>:
>
[snip]
> 
> I'd just like to second this idea! Personally, I'd like to see a
> published anygui API hit version 0.2 final with only a single reference
> backend and then see the various other backends developed to support the
> API after the 'official' release.

Hm. This *is* an interesting idea -- and it might even make the
project more viable. The prolification of back-ends in the
distribution is one of the major issues holding back a release, as
they must all be de-bugged, tested and the like.

I think a natural choice for a "standard" back-end (which would be
included in the Anygui distribution) would be tkgui, since Tkinter is
the most standard-ish toolkit. We could, of course, go with _no_
back-ends in the Anygui distribution, but it seems to me that this
would be taking it too far -- after all, the other back-ends should
have something to compare themselves to.

Before deciding on a policy like this, we should canvas the people
responsible for the various toolkit distributions (wxPython, PyGTK,
PyQt, what-have-you) to hear if they're interested in distribute an
Anygui back-end (perhaps we should call them "adapter modules" or
something?). Actually finishing 0.2 with a solid and _simple_ tkgui
(for example) might help.

Also, I suspect that had better keep the Anygui project a forum for
development of these back-ends. Or, perhaps, we could do "Anygui
certification"? And create a logo for it or something? Then the
various toolkits could get "Anygui certified" if they wanted to get in
on the fun... And we could provide back-ends to those who asked, where
we had the capacity?

And another thing: We'd have to devise some way of detecting Anygui
back-ends. One possibility is to simply have back-ends register so
that we put them in the official wishlist (as now) -- then they
wouldn't have to be installed anywhere in particular. Also, it would
be possible to simply add to the wishlist through existing
mechanisms (such as ANYGUI_WISHLIST, or the upcoming setup function).

Samuele: Would the Jython bunch be interested in distributing (not
necessarily maintaining) javagui.py? In that case we would, at least,
have the two main Python implementations covered... :]

BTW: I've been thinking about generating API docs with epydoc
(epydoc.sf.net). I thought about using nullgui as the back-end with
docstrings etc. for this, but if we had only one back-end that would
be just as natural. I believe that a solid, up-to-date API
documentation is sorely needed... (Not necessarily related to the
issue at hand, of course.)

[snip]
> IMHO :-)

-- 
Magnus Lie Hetland               "Nothing shocks me. I'm a scientist." 
http://hetland.org                                   -- Indiana Jones


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