Wil, the tl;dr here is "Philosophical beliefs aren't an effective underpinning for good software design. Start over."
It's taken me a while to piece together much from the early discussions about Flow and figure out how we got to where we are now. It's my opinion that the root of the problem is that, much as the WMF wants to move toward being a "software" or "tech" organization, it really doesn't have very much history or experience in the kind of ground-up software design and deployment that is conducive to successful implementation. Tech organizations seeking to redevelop a core function normally start by gathering extensive data on the current system, identifying key functions that must be incorporated into the new system, understanding how the current system is used, what its strengths and weaknesses are, and ensuring that even early iterations will at least include the key functions and strengths from the soon-to-be-deprecated system. This baseline background research was never really done before investing in the development of Flow. Instead, Flow very much comes across as software being designed according to philosophical principles rather than function.
The major deficiencies that have long been identified in the current discussion system (and that can be addressed by technology) are all able to be addressed in MediaWiki software or by extensions. Automatic signatures have been done by bots for years; indenting could be added to the editing function gadget and moved to an extension; much work has already been done on graceful resolution of edit conflicts. The ability to watchlist an individual thread or section of a page is more challenging but, I have been told, still possible.
Several of the features identified as "must-do" have turned out not to quite work out. Automatic signature (something that is currently functional on Flow, but is not customizable) turns out to be more of a challenge when users are widely known by a signature line that doesn't match their username, and there is no method by which users can add an "explanatory" note to their signature such as "formerly known as User:Whatever". The "more efficient" indenting has reduced possible indents to three levels, without exception; even in simple discussions, it's pretty clear that hasn't really worked out as it's often difficult to figure out who is responding to which post. "Rigid predictable technical restrictions on who can edit what" has resulted in inability to remove posts that are obviously unsuitable (there's no "undo" or "revert" function), replaced with a "hide" function that can only be applied by certain users that's practically a red flag for people to look-see what the problem edit is. With broader early discussion, some of these would probably have been fleshed out before getting this far.
At the core is whether or not there is value in developing a "discussion system" that is radically divorced from any other interface used by the system. This is a philosophical question, and doesn't actually have that much to do with technology - and this conversation has never really taken place anywhere but by a bunch of guys who are into making cool software and, often as not, have little interest in the kinds of discussions that normally occur on Wikimedia projects. There has certainly never really been a discussion with the broader community about what would better serve in discussions. More importantly, some of the core assumptions and goals upon which Flow has been built[1] have very little to do with technology at all - plenty of research indicates that new users are driven away by the nature of discussions rather than the technological challenges of participation - and the lack of active broad community consultation means that the development team really doesn't know what's working well, what's problematic, and what kind of efficiencies experienced users are looking for. There's absolutely no basis to believe that Flow is in any way likely "to encourage [more] *meaningful conversations* that support collaboration". (I'd love to see what kind of metrics would be used to assess the meaningfulness of conversations!)
And the other key issue is a complete lack of recognition that the more UIs a new user needs to learn to develop competency, the lower the likelihood that they'll actually be able to develop the necessary skills to become fully functioning members of the editing community. The Wikimedia "family" has largely bought in to the necessity to introduce a WYSIWYG editing interface (that would be VisualEditor), and to recognize that wikitext editing needs to remain in existence as well. Adding a third one whose primary purpose will be to talk about the content being created using the other two is counterintuitive at best.
Risker/Anne
[1] https://www.mediawiki.org/wiki/Flow
On 5 September 2014 21:53, Wil Sinclair wllm@wllm.com wrote:
Risker, what do you think might get us all back on track for Flow? Should the WMF consider a reset of the project and proceed only after making specific and enforceable commitments to work with the community? Is a total rewrite in order? Should we go completely tabla rasa on it and revisit whether we need something like this at all?
,Wil
On Fri, Sep 5, 2014 at 10:58 AM, Risker risker.wp@gmail.com wrote:
I think there have been some pretty strong indications over the years
that
the current talk page system needs to be improved. However, there's been little discussion at all about whether Flow is that improvement. I have been following the development for quite a while, and it really looks
like
the system was developed backwards: essential functions for effective discussion that already exist and are used on a daily basis were not included in the initial designs, while the design incorporated plenty of bells and whistles that were considered desirable (although the reasons
for
desirability weren't necessarily universally held or particularly clear). This has resulted in a huge amount of re-engineering to incorporate (some of the) needed functions , and a lot of downplaying of the feedback given because the feedback has conflicted with the "bells and whistles" of the original design. There is also the fact that it would add another completely different user interface to the editing process, which
increases
barriers for existing users but even more so for new users.
In other words, the issues with Flow are so deeply rooted in its core design and philosophy that it may not be possible to come up with a
product
that is actually useful on the projects we have to replace the discussion system we have. It seems that the Flow team has assembled the
ingredients
to make a chocolate cake with the hope that it will be a suitable replacement for vegetable stew.
Risker/Anne
On 5 September 2014 13:29, Wil Sinclair wllm@wllm.com wrote:
This somewhat circuitously brings us back to the subject. We have a chance to rollout Flow the right way. There are some questions that come to mind that might tell us if we're headed for a big win or a bigger debacle:
- Is the WMF working with the community as closely and substantially
as possible to make sure Flow is ready for primetime?
- Is the community preparing itself for a major change, not only in
interface, but to some degree in wiki-philosophy about how discussions are conducted- not to mention the notion that, while wiki software can do almost anything involving asynchronous online communication, it can't do everything as well as other interfaces?
I think Flow will be particularly challenging. I deployed Liquid Threads on another site. I liked the threaded interface, as did others. But overall it was roundly rejected because it was harder to search (I only found out you have to add the namespace to the searchable namespace in LocalSettings.php later), and it invasively took over all discussion pages, among other headache. Problems like these could easily be addressed before a rollout, but they should be addressed as early as possible. It is notable, however, that the more our users used it, the more they seemed to like it.
What can we do to make the Flow rollout as smooth as starting '''now'''?
,Wil
On Fri, Sep 5, 2014 at 9:34 AM, Marc A. Pelletier marc@uberbox.org wrote:
On 09/05/2014 11:12 AM, Yaroslav M. Blanter wrote:
On 25.08.2014 06:07, Marc A. Pelletier wrote: FLOW?
Last I checked, Flow isn't deployed except as experiments in a handful of places, and is still in active deployment.
But you're correct that this would constitute a replacement rather
than
a new method alongside the old. A long, long overdue and desperately needed replacement -- but a replacement nonetheless.
That also explains the very deliberate development and feedback loop.
-- Marc
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-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org <
https://meta.wikimedia.org/wiki/Mailing_lists/GuidelinesWikimedia-l@lists.wi...
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
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-request@lists.wikimedia.org?subject=unsubscribe
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-request@lists.wikimedia.org?subject=unsubscribe