On Tue, Aug 25, 2015 at 5:42 PM, Romaine Wiki <romaine.wiki@gmail.com> wrote:

Abouth the sections of Phase 1 & 2 I am concerned. The way it is suggested to be set up now is that the proposals with the most votes win. First of all, large wikis & communities have more people to submit proposals and more people to vote. So it seems that this is becoming a popularity contest, which is the wrong direction.

I actually suspect that the proposals that will get the most votes will be ideas that benefit the most projects. If an idea would only benefit French Wikipedia, for example, I doubt it would get as many votes as a tool that is usable by every Wikipedia. That said, I do agree that some projects are probably going to be at a disadvantage. For example, I'm not expecting WikiSpecies to make a strong showing. To some degree, this makes sense, as we want to focus our limited resources where they will have the largest impact. We also want to make sure, however, that we are not ignoring vital sectors of the community. We haven't fleshed out plans for it yet, but we have some very tentative plans to do focused technical workshops with specialized community groups next year. For example, we might hold a workshop for GLAM volunteers or Wiktionary volunteers or Recent Change Patrollers to find out what their specific needs are and how/if we can address them. This idea is dependent on what kind of resources our team has next year and is very tentative, so don't hold me to it just yet. Right now, it's just an idea :)

In the Dutch community it happens a lot that users are against certain software, until they tried, and the other way round happens as well. Basing a voting on what people like can easily become unbalanced because of various factors. Also what matters to one person, does not have to matter to another person (like because he doesn't do anything with it), while it can be essential to another.

These points are all true and this is the reason we will be offering feedback on as many of the proposals as we can before voting begins. Specifically, we will try to assess them for feasibility, impact, and risk, so that voters can make well-informed choices rather than just voting for their favorite gadget. Also, keep in mind that voting isn't the final step for prioritization. After voting is finished, we will be taking the top X requests (probably 10 or 20) and then doing our own analysis of them to determine priority. It's possible some of those requests will be referred to other teams or even rejected if they don't make sense for us to work on.

The process for this survey is still a draft, however, and we are open to significantly revising it. If you have suggestions for a system that might work better than voting and you can show community support for the idea, we can switch to a different process. The current draft is based largely on consultations with the German TCB team (which has done similar wishlist surveys), the WMF Community Engagement team, and a few random community members that we've talked to. I'm sure it isn't perfect, but I doubt any process will be.

Ryan Kaldari
Engineering Manager, Community Tech