On Mon, Jan 25, 2016 at 9:59 AM, Matthew Flaschen mflaschen@wikimedia.org wrote:
On 01/22/2016 05:03 PM, Rob Lanphier wrote:
The reason I want the rename [in T124255 https://phabricator.wikimedia.org/T124255]: ArchCom is the mechanism we hope to ensure
we build and deploy increasingly excellent software on the Wikimedia
production cluster in a consensus-oriented manner. MediaWiki is at the center of this, but ArchCom's responsibility doesn't end with MediaWiki.
In my opinion, there needs to be a group leading development of MediaWiki itself, focusing on the product and the product roadmap (influenced by who uses it). WMF is a critically important user of MW, but not the only one. [...] I would suggest we might want them to be separate groups. The group in charge of MW's roadmap would not have to care about things like major operations/puppet restructuring, while the WMF cluster group would.
(Note, this is related to the discussions about MediaWiki Foundation, but doesn't need to wait on that).
Logically, I think your long-term vision makes sense. Managing the software we deploy to the WIkimedia cluster is a lot of work, and it deserves focus.
In the short-term, I believe a non-Wikimedia focused subgroup of ArchCom may make sense. The declining MediaWiki use outside of Wikimedia has been a longstanding problem for us, but not the biggest problem. ArchCom's focus is (and should continue to be) the needs of Wikimedia.
The reason why I'm delineating it as a subgroup is not a power grab, but an essential step toward building the trust required for long term viability of a MediaWiki(?) Foundation to be viable. The fact that the people working on the "MediaWiki Foundation" are still(!) using the name "MediaWiki" represents a failure of imagination among all of us. For example, if MW Stake wants to be a viable upstream, there has to be a stake/steak pun buried in there somewhere that could represent a great name for a viable fork of MediaWiki.
Yes, I used the word "fork". I believe Wikimedia Foundation would love it if MediaWiki forked, and we were "forced" to switch to the fork. There are other projects (e.g. gcc, KHTML/WebKit, Inkscape) that were helped by a fork. WMF wouldn't be offended at all by an attempt to create a viable fork, as we know that there is a limit to how much we work we should try to fund off of our current donation model. If we should be pressuring anyone to make non-Wikimedia use of MediaWiki viable, it shouldn't be WMF, someone from Wikia should step up. :-P
As it stands, Wikimedia Foundation is the only trusted upstream for the MediaWiki codebase. I believe WMF should jealously guard the "MediaWiki" trademark, if for no other reason than to force someone to come up with a different name. "MediaWiki" and "Wikimedia" are too similar, and there are not good reasons for us to license that trademark to anyone else. WMF doesn't have a patent on the alphabet; come up with your own damn name ;-)
Naming isn't the only issue for a viable fork, though. There are other things that a viable fork would need for WMF to trust it:
- An upstream repository. This could be anywhere (even Github!). We would need to be able to trust that upstream would collaborate with us to solve our dealbreakers. - Trusted architects with clear vision and leadership - A governance structure that allows WMF to operate as a worthy peer
We have healthy relationships with other upstreams (e.g. Phabricator, Debian, Composer), and though we don't always agree with the choices of our upstream, we strive to collaborate with respect, and we figure out what to do if upstream makes a choice that creates a problem for us.
So: forks welcome! Any takers?
Rob