Re: Everblue Standards :D

> whether the rest of the world should get a chance to comment on what we do at a phase where it is still useful to do so, 
> to let us know when the way we're trying to do things has overlooked some important use case.

I'm afraid we have to disagree on whether "waterfall" is a good characterization of the web standards process.  The notion that comments must come "at a phase where it is still useful" is intrinsically waterfall-ish.  In the "fluid" world, feedback is welcomed even after a feature is deployed, and probably taken more seriously because there is a concrete test failure, inability to perform a use case with assistive technology, inability to use a feature with some language, etc. Also the implementers' have stringent accessibility requirements that are enforced as part of the shipping process; my understanding (indirect, I don't work as an actual product PM anymore) is that our developers get much more timely and actionable A11Y and I18N feedback from our internal experts and procedures than from the W3C community.   

Likewise if an important use case can't be achieved, that feedback is welcome, and will be acted upon, at any point. And few product managers are going to block release of a version until it supports all known use cases; triage is the very essence of effective product management, especially "defer until the next release" means waiting a few weeks rather than a few years to support a use case.

-----Original Message-----
From: Florian Rivoal <florian@rivoal.net>
Date: Monday, June 17, 2019 at 2:06 PM
To: fantasai <fantasai.lists@inkedblade.net>
Cc: Michael Champion <Michael.Champion@microsoft.com>, Philippe Le Hegaret <plh@w3.org>, W3C Process Community Group <public-w3process@w3.org>
Subject: Re: Everblue Standards :D
Resent-From: <public-w3process@w3.org>
Resent-Date: Monday, June 17, 2019 at 2:05 PM

    
    
    > On Jun 17, 2019, at 22:52, fantasai <fantasai.lists@inkedblade.net> wrote:
    >> On Jun 17, 2019, at 21:48, Michael Champion <Michael.Champion@microsoft.com> wrote:
    >> 
    >> The W3C Process as practices today evolved at a time when the https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FWaterfall_model&amp;data=02%7C01%7Cmichael.champion%40microsoft.com%7Cafa60a718aa84301929f08d6f3679cf1%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636964023677935788&amp;sdata=sS545PJTMLqTlX3M8I7cFca645aiCO0Nd6z4BW%2FLczo%3D&amp;reserved=0 was dogma
    > 
    > You are right that the software industry as a whole has moved away from Waterfall 
    
    I dispute the claim the the web platform as a whole was ever under the waterfall model, ever since multiple browsers have existed. For sure, some features were developed spec-first-implement-later in the past. Some are also done this way in the present. In both cases, we then revisit the spec after implementations exist to make the two converge. However, in that same past, there were also features that were implement-first-spec-later. We have got better at writing specs, we have got better at writing implementations, we have got better at iterating on them, but there was no point in the history of the web where everything was waterfalls.
    
    So we need to need our process to get better because all our practices get better, and there's no reason the process should lag behind.
    
    This is completely orthogonal to the question of whether the rest of the world should get a chance to comment on what we do at a phase where it is still useful to do so, to let us know when the way we're trying to do things has overlooked some important use case.
    
    —Florian

Received on Monday, 17 June 2019 22:07:55 UTC