On 09/01/2016 03:35 PM, C. Scott Ananian wrote:
However, no one was present at the dev summit who used LanguageConverter on their home wiki, and few folks who rely on Content Translation routinely. (Maybe one or two were present, but not enough to have a reasonable discussion about the future of these features.) For better or worse, previous dev summits have had weak representation from those who are not American users of projects-other-than-enwiki.
I think this general problem is partly inevitable.
I'm not saying that all Dev Summit attendees are enwiki users, or will be in the future. But I think that even with scholarships and outreach, it will inevitably be skewed in a few ways:
1. Almost no one will show up, when compared to our broad user base. The people who do come will mostly be a subset of the most enthusiastic power users.
Power users are one of the groups we're building software for, but not the only one.
This can be mitigated with hard work at outreach to representative users who don't typically come to this kind of thing.
But this is also why we need to use other outreach and measurement efforts, like EventLogging, to reach and measure the people that don't attend summits (and may not answer surveys regularly).
2. The people that do come will be skewed towards being native speakers of English.
3. They will be heavily skewed towards being either WMF employees, San Francisco, or able to afford to fly. Scholarships can help with this, but only partly.
So I would be careful about seeing this as a great opportunity to meet an accurate cross-section of our user base. We need to be a little more strategic when thinking about what to build and who we're building for.
I agree with C. Scott there might be a role for specifically inviting key groups of users (e.g. shared hosting) to work with them in person and understand their needs. Though this could be at the Dev Summit, it doesn't necessarily have to be. It could be at Wikimania, or in SF but another time, etc.
Matt