Hoi,
The notion that the MediaWiki software is almost entirely developed for
one organisation is seriously wrong. OmegaWiki, formerly WiktionaryZ has
realised a lot of functionality already and this is just one other
project. There are many more projects that have developed on MediaWiki
and much of this development, like OmegaWiki, is as Free as the WMF
developed functionality is.
As to distributed MediaWiki, as you may remember the Vrije Universiteit
of Amsterdam (Andrew Tannenbaum's department) is actively working on a
distributed Mediawiki and is particularly interested in the distributed
network functionality that is required. This includes things like making
sure the content is near to where it is requested.
The WMF could provide a meeting place for organisations that use
MediaWiki stimulate cooperation. The WMF can provide a developer that
has as his task to mentor new developers, particularly students from
Universities that want to be technically involved in MediaWiki
projects. My expectation is that it will be possible to do some 50
projects in a half year (only the MW support .. not supporting the
content part of the project) this is likely to lead to a retention of
developers of in between five to ten percent and will as a consequence
be a good investment. The organisations that develop MediaWiki may also
need support to build extensions so that their code can be part of the
main MediaWiki code. This can be a paid for service. When it is not the
WMF who does this, another organisation may be willing to provide this
service ...
I am sure there are more things that can be done when MediaWiki has its
organisational part developed it may even generate money for the Foundation.
Thanks,
GerardM
David Gerard schreef:
On 08/01/07, Tim Starling
<tstarling(a)wikimedia.org> wrote:
Nobody can deny the close relationship between
Wikimedia and MediaWiki.
Some might worry about Wikimedia taking control of MediaWiki,
Well, that's pretty much the de facto condition - it's software
written and actively developed almost entirely for one organisation,
that just happens to be open source and useful to lots of other
people.
OTOH, it's only good sense that getting outside users involved in
development will make for a better and more robust application.
Actively recruiting outside developers doing interesting stuff will
mean the weird and wonderful ideas can go in as extensions rather than
being separate forks.
(Remember the chitchat about "distributed MediaWiki"? Seen
http://www.wikileaks.org/ ? I've already written a note to one of the
devs strongly suggesting they participate in the mainstream of
MediaWiki development.)
[Mind you, I was sure this would be the case with Slash, which is of
industrial robustness ... but an exercise in pain to sysadmin. I want
to hurt it. Real bad. In the face.]
So the question becomes: What would or could Wikimedia being the
official (not just de facto) organisational umbrella do to or for
MediaWiki? Assuming things aren't actually broken right now, which it
appears from this thread is the case.
+ Legal backup in case of attacks from the querulous? Lots of
prominent free software has a legal backup these days. Though most of
the legal concern for Wikimedia is our content.
- Too many eggs in one basket
+ Express task of recruiting outside devs
- ... who then have to be herded by current devs
Please add more.
which is
amusing considering the amount of influence MediaWiki developers have over
Wikimedia affairs. They have as much to fear from us as we do from them ;)
The cure for en:wp admin politics is to point out the devs have all
the REAL power ;-D
- d.