W3C home > Mailing lists > Public > public-w3process@w3.org > December 2016

Re: Dead Ends

From: <chaals@yandex-team.ru>
Date: Tue, 27 Dec 2016 15:33:36 +0100
To: fantasai <fantasai.lists@inkedblade.net>, Michael Champion <michael.champion@microsoft.com>, Jeff Jaffe <jeff@w3.org>, "public-w3process@w3.org" <public-w3process@w3.org>
Message-Id: <96331482849216@webcorp02d.yandex-team.ru>
TL;DR: I'm not sure this is worth doing. There are complications regarding patent licenses. But that just means I won't prioritise making a proposal myself. Anyone can do so. That's why we have an open group :)

27.12.2016, 13:54, "fantasai" <fantasai.lists@inkedblade.net>:
> On 12/27/2016 02:21 AM, chaals@yandex-team.ru wrote:
>>  26.12.2016, 14:02, "fantasai" <fantasai.lists@inkedblade.net>:
>>>  It would be *better* if the Process allowed for WDs and CRs to be
>>>  rescinded. Currently only RECs can be rescinded. :(
>>>
>>>  I guess that's a relatively easy fix; hadn't occurred to me to request it.
>>>  Chaals? :) :)
>>
>>  The process currently suggests Working Groups publish dead things as notes -
>>  empty with the obnoxious notice is the common 'best practice'. Where W3C
>>  closes down work, they are required to do that. So I think the mechanism
>>  is there…
>>
>>  As I noted elsewhere, making the default view of the TR page reflect live
>>  things differently from dead things would be a simple thing to do, doesn't
>>  require any process change, just a bit of work on the page itself.
>
> Either it is better for dead things to have their own status *or* the system
> of republishing as notes is better. I can't see any reason, other than "I
> don't want to make a change", to take a hybrid position on this matter.

"I don't want to make a change" is certainly *part* of the motivation here.

There are good reasons to change various aspects of the process - which is why I've spent the last few years getting that to happen. But the more it changes, the less people realise what it actually says. We continue to see that implementation takes a while to catch up with changes.

Also, I'm the editor of the Process document, not the author or owner. If this group, or subsequently the AC and W3C decide something I don't like, it goes into the Process anyway.

Meanwhile there is another reason I'm not excited about such a change…

> If the system of giving dead things a definitive status is better, then all
> things that are dead, whether they reached REC or not, should be able to
> take that status.
>
> If the system of republishing as notes is better, then let us remove the
> Rescinded status from the Process and treat dead RECs the same as WDs and CRs,
> since we have concluded that that method is better.
>
> In either case a change to the Process is warranted.
>
> Personally, I'm in favor of marking dead things clearly with their own
> status and would prefer that the Process allowed WDs and CRs to transition
> to Rescinded as well as RECs.
>
> What, if any, are the objections to this?

Recommendations impose a significant licensing obligation on various W3C members. Rescinded Recommendations limit that obligation to existing licenses with no more being granted. I don't think W3C has *rescinded* a Recommendation yet.

In pre-Rec status, the situation is different, because there are licensing commitments made to specifications that have reached WD, which will only be realised by moving a specification to Recommendation. 

For the proposed new Process the AB developed a concept of "Obsolete" Recommendations which, like abandoned Working Drafts are considered "unlikely to progress". The concept explicitly recognises that the future, being hard to predict, may turn out to include something we thought was dead, which is why it gets an "in-between" status.

To pick an example, the Web Platform WG is now chartered, at the instigation of at least Google and Yandex, to produce a Recommendation for microdata - which was part of HTML 5, then parked as a Note, and now turns out to have very wide adoption of some parts related to document content, while the browser API is clearly falling out of the limited use it had and should be removed. 

A file system API, packaging, and styling of audio are some other areas where there has been on-and-off work over a number of years.

In conclusion, I'm personally not convinced that it is worth developing a whole new set of process for more clearly rescinding and unrescinding development work, to replace the current situation. Which doesn't mean that it is a bad idea, and I assume we will collectively consider any proposal on its merits. 

cheers

-- 
Charles McCathie Nevile - standards - Yandex
chaals@yandex-team.ru - - - Find more at http://yandex.com
Received on Tuesday, 27 December 2016 14:34:14 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 27 December 2016 14:34:14 UTC