W3C home > Mailing lists > Public > public-w3process@w3.org > May 2013

Re: superseding a recommendation

From: Wayne Carr <wayne.carr@linux.intel.com>
Date: Wed, 29 May 2013 13:00:54 -0700
Message-ID: <51A65E76.1070709@linux.intel.com>
To: public-w3process@w3.org

On 5/29/2013 8:17 AM, Charles McCathie Nevile wrote:
> Hi,
> in general, W3C Recommendations live forever as Recommendations. There 
> is a process to rescind them, but there is no process to supersede 
> them. Formally a new version of a spec at W3C doesn't have any way to 
> replace an older one, which doesn't really make sense.
> I think it would be useful to have such a process. I believe there are 
> versioned specifications that don't actually supersede the previous 
> version, such as some CSS modules, where the two versions live happily 
> side-by-side. (What about HTML 3.2? Do people use it? Is that a good 
> idea? Maybe we just need a spec for "HTML for email clients"...)
> If we had such a process, what would we use it on? Should we just be 
> more active about rescinding recommendations?

Just like IETF with Obsoletes and Obsoleted-by, it would be useful to 
have a very visible note at the top of an older spec - superseded by.  
That's just a factual statement, housekeeping, so I'd think this could 
just be done now.  I'm surprised it isn't done, given how easy it is to 
search and find an old superseded spec.

modules vs monolithic specs shouldn't be an issue.  they're different 
specs.  but same spec with a higher version number seems obviously a 
place for it.

rescinding is a bigger deal.  other than the obvious patent issue or 
turns out to be harmful, rescinding old specs seems more trouble than 
it's worth as long as their clearly marked.

> cheers
> Chaals
Received on Wednesday, 29 May 2013 20:01:22 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:35:08 UTC