On Fri, Sep 5, 2014 at 11:41 PM, Pine W <wiki.p...@gmail.com> wrote:
> Something that that would be useful is a video demonstration of Flow in
> action.

That could be handy, Pine. But sometimes you can't demonstrate all
benefits yet, because they don't even exist in the implementation yet
-- only in the foundations of an architecture. And sometimes you can
show the idea of a benefit, but people will look at the current
implementation and hate some aspect of it, and fail to imagine a
version of it that would be beneficial.

To give an example of the latter, Flow already has the ability to give
thread-level notifications of responses. But the first implementation
of notifications was very spammy (generated too many notifications),
so people who looked at it said "I don't see the benefit! It's just
spammy!". Love em or hate em, sites like Facebook spend years tweaking
the algorithm for every one of their features (newsfeed,
notifications, etc.) to get the best results from their perspective.
It takes time to get this right -- and the initial reaction may often
be "No, this sucks" because, well, it's not quite right yet.

Fundamentally, I'd ask people to relax a bit regarding Flow. Nobody's
planning to push this one out radically. Today we saw some on-wiki
drama because a new test page was turned on. For something like the
en.wp Teahouse, I'd want the hosts to be fully on-board before
converting it over (and the rest of the community to not oppose it).
If that's not doable, we can focus on other use cases first. It's
early days and this one's a long haul -- just like VE. But we
shouldn't shy away from a problem just because it's hard.

Erik
-- 
Erik Möller
VP of Engineering and Product Development, Wikimedia Foundation

_______________________________________________
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 
<mailto:wikimedia-l-requ...@lists.wikimedia.org?subject=unsubscribe>

Reply via email to