Website maintenance volunteers?

[Barton C Massey]

In message <68da43e00712050810o5550d379gaa281f10fcb40252 at mail.gmail.com> you wrote:
> 2007/11/29, Arthur Huillet <arthur.huillet at free.fr>:
> > But I'm unsure we need to change the wiki engine, it
> > seems that it would represent a lot of work for
> > everybody involved, and we certainly want to avoid that.
> 
> Yes, unless sure there are implementers who are able to do a complete
> porting of all information to the new system in a short time, it's
> better to fix the old system than to replace it with something
> completely new. The new systems often do not reach the level
> previously set until a very long time, as the migration paths usually
> involve loss of all kinds of (meta)information.

Actually, our MoinMoin->ikiwiki tool is quite good; it was
developed for a fairly large and uncontrolled wiki
(http://psas.pdx.edu) which had previously been a TWiki, so
it was full of interesting test cases.

That said, I agree that there will be various fixups
afterward.

> Spam is not a huge problem with working automatic BadContent updating,
> provided enough users have DeletePage rights. But captcha could be
> welcome.

I'm not understanding people's spam concerns.  As far as I
know, only X.Org members are currently allowed to edit the
wiki.  Am I wrong here?

> If you stay in Moin, what should be done in any case is to do a custom
> theme file for MoinMoin - the default one is one of the ugliest around
> IMHO. Some examples are available at
> http://moinmoin.wikiwikiweb.de/MoinMoinScreenShots.
> https://wiki.edubuntu.org/Edubuntu is quite neat. I also like hiding
> of some or all controls depending on whether one is logged in or not,
> and adding some extra navigation links in the theme - see eg.
> http://www.vapaasuomi.fi/ and http://wiki.ubuntu-fi.org/UKK (I did
> Moin themes for those) - but it's more suitable for the more end-user
> oriented sites where majority of readers are just that, readers.
> Usually some elements like the "trail" are worth hiding in any case to
> remove clutter, IMO.

If someone wanted to work on a custom MoinMoin theme based
on some better existing theme, it would potentially deal
with some of the usability problems we've seen.  I don't
have the energy to do it, though.

> Image usage problems diminish somewhat if ImageLink macro
> is made usable.

That would be a requirement.

> Moin tweaking is easier than changing the whole system, but there
> should be someone to do that, too, so as always it comes down to which
> people are there that may do enough work. So changing the whole thing
> is definitely an option, too.
> 
> -Timo (I can give the theme py files I've done as an example if
> needed, though there are others lying around, too)

Thanks much for your comments!

The big questions in my mind are: whether the git-nature of
ikiwiki is a big enough win to justify the work of a
conversion; in any case whether upgrading and properly
theming the MoinMoin is less work than moving to ikiwiki and
properly theming that; whether there are other ikiwiki
usability issues that would make it less-loved than MoinMoin
(we have no easy way to go *back*, currently).

Knowledgable opinions on these issues are highly welcome---I
need to start moving on this stuff pronto.

    Bart