Re: version identifiers

On Wed, 2009-08-26 at 15:29 -0700, Larry Masinter wrote:
> (resent to public tag) with regard to ACTION-283 / ISSUE-41
> 
> In the meanwhile, here is the Editor’s Draft
> 
> http://larry.masinter.net/tag-versioning.html


I'm having a hard time getting engaged in this document; it
uses an editorial style that I find off-putting; it
starts with "This document...", which is too self-conscious
for my tastes.

It also presents a list of terms and definitions without
motivating the complexity.

At the risk of beating a dead horse, if we're to work on
versioning, I strongly suggest we approach it as a collection
of patterns, i.e. known problems and solutions
e.g.
 - when HTML was started, we didn't know which tags we'd ultimately need
  - therefore: ignore unknown tags
  - but: it doesn't always work; see forms, tables
    (perhaps another pattern is the <noscript> alternative pattern)

 - likewise in HTTP
  - but there the solution is to ignore not just the header name
    but the content of the header too

 - the XSLT versioning strategy

 - the COM uuid strategy (mostly for contrast with web techniques)

oops... time for the telcon...



-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
gpg D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Thursday, 19 November 2009 18:07:21 UTC