27.04.2015, 11:00, "Kai Scheppe" <k.scheppe@t-online.de>:
 
 
Revitalizing W3C - I think it is about culture and people, not documents or guides.
 
Clearly.

Everybody's gut reaction is to create a process or use a tool of some sort, in the hope that this will change human nature.
It will not.
First you need the culture, then the process or tool to support it.
 
True. Including the fact that we do need a process to support what we do - there is too much work spread over too many people for us to somehow hope that everyone already talked to everyone. We can see even within HTML that this idea (which is wired into WHATWG) doesn't really work.
 
So how is human nature expressed in W3C work?
I think people do not have the patience and the interest to stick with standardization work throughout the entire process.
Therefore they are happy to leave that work to others (W3C) if it does not directly affect them.
 
Well, some people, and there are also issues of having the technical and people skills, money, ability to work with those already involved in something, and so on. But that never changes, so the question is what we do about it.
 
What is the culture at W3C?
W3C and those few really active members, most of whom are on this list, want to "get it right" the first time.
This, however, created a drawn out process that is turning the public away.
 
Yes…
 
I think what is missing at W3C is a slightly different outlook and hands on people work.
More agility, less waterfall.

Suggestions for a solution:
- deliver quicker results by focusing on the most important things first and do that well
 
Agreeing what those are, and what "do them well" means is actually quite difficult. If we agree that the "horizontal review" that W3C provides - accessibility, internationalisation, hopefully improving security review, etc, is a value, then we could operationally define "do it well" as "have the horizontal review folks and implementors happy so far…"
 
- embrace incompleteness in specs and allow for practical usage to show where the journey needs to go (Kaizen is the model here)
- use short intervals to publish new iterations, which should consist mostly of addendums and few corrections
- chairs need to lead much more strongly and, for example, drop issues if work is not being done or debated endlessly, because obviously it is not important enough to obtain a result
 
These three things seem really key to me. W3C can produce simple specs fast, or at least as fast as it can deal with the horizontal review questions, and get them implemented. But too often there is a sense that everything has to go into version 1 - sometimes because there is a lack of trust that problems shunted to a version 2 will be dealt with in the real world, sometimes for other reasons.
 
A culture that aimed to produce versions of specs, somewhere between the original W3C model of "make a spec and then stop" which is an assumption somewhat baked into the process and patent policy, and the "what's the spec today" approach of the "living standard" proponents, would represent a real improvement.
 
Stability matters - for people who are trying to build a business that needs to make contracts based on a known goal, among others. But evolution of technology is also an important reality, and holding specs back artificially to enable easier harmonisation is just a recipe for being left behind by that evolution.
 
- the rules must be few, simple and clear, but truly enforced
 
As few and simple as possible - and no simpler nor fewer.
 
- create a much stronger focus for groups, by providing the role of moderator who has no technical stake in a specification, but solely works with the participants to keep them focused
 
That's what a good chair (or chairing team) does. I've had enough discussions with enough different people over the last few years to think that there is a real problem here and we are not doing a good job of recognising it, let alone fixing it.
 
Chairing a group at W3C should be an honour people strive to be good enough to earn, and be based on chairing ability, rather than belonging to the right industry sector, technical ability in the particular field which is as often a distraction from chairing as it is a help, or the like. But to get there from here we have to take a pretty robust approach to assessing, educating and as needed replacing our chairs.
 
cheers
 
Chaals
 
--
Charles McCathie Nevile - web standards - CTO Office, Yandex
chaals@yandex-team.ru - - - Find more at http://yandex.com