On Tue, Aug 12, 2014 at 5:33 PM, Henning Schlottmann h.schlottmann@gmx.net wrote:
On 12.08.2014 16:57, Magnus Manske wrote:
German Wikipedia had 1.1 billion page views in June [1]. ~300 votes (~2/3 against MediaViewer) do not represent the readers, IMHO.
Claiming to speak for a perceived silent majority will not help you much in this discussion.
I do not make any such claim. All I say is that the 300 (is there a movie plot here?) do not necessarily speak for it, either.
There is a common pattern in the conflicts between WMF and several communities over software developments during the last few years. As I wrote two weeks ago to Rachel:
| Decision making seems to be focused on reader experience, including | winning readers to become authors, but existing authors and their | experience (in both meanings of the word) is ignored. Even by people | like Eric, who once was a prolific author himself
| Authors see themselves as the single most important group in the |Wikimedia universe. Without their content, there would be nothing: No | readers, no fundraising banners, no donations, no employees, no | foundation. On the other hand, WMF seems to see the readers (and | donors) as their main target audience. Of course WMF knows, that all | the projects need content and authors, but in my opinion most of them | fail in appreciating the existing authors and focus too much on | winning readers to become authors, by simplifying the entry.
This is serious. WMF really needs to appreciate the expertise of the author community and accept their experience a important and valid. If authors tell the WMF and particularly the devs, that a particular function is necessary, then the devs really, really need to think.
I do agree with this. Visual Editor (which works much better these days) and MediaViewer are not aimed at the experienced editor. They aim to make the reader more comfortable, and try to ease the first steps into editing. Winning new editors has been deemed a priority, somewhat at the expense of WMF-made support for the power user. This is a judgement call the Foundation has to make.
If the community tells the devs, that a particular idea is a bad one, a feature is too buggy to be rolled out (as default) or is unsuitable for a project at all, this warrants more than just a cursory thought.
A formal RfD must not be taken lightly, overruling it by creating a whole new user class, and crippling the elected admins is inpermissible. WMF has broken trust again and this time in a unprecedented way.
As Erik pointed out, WMF had made it quite clear that they reserve the right to overrule the community in that specific matter, before the Meinungsbild was done. WMF then acted as announced, and refused to be "hacked" out of their own servers. An unfortunate escalation on both sides, but since they never promised to accept the Meinungsbild (quite the opposite!), it was not a breach of trust.
Until this event, I thought the dev process to be broken, not just the communication around devs. But now I believe the conflict runs deeper.
It points out an issue we (community and WMF) should discuss, in a more general sense. What should the decision process be for technical changes? When does the Foundation get precendence, and when should the community have the last word? What weight should small-scale "votes" of editors have? Should random polls be done, and included in such votes? Etc.
The MediaViewer "affair" itself gets blown out of proportion IMO.
Cheers, Magnus