- From: Tobie Langel <tobie@fb.com>
- Date: Thu, 22 Mar 2012 09:45:39 +0000
- To: Charles McCathieNevile <chaals@opera.com>, "public-w3process@w3.org" <public-w3process@w3.org>
On 3/22/12 10:19 AM, "Charles McCathieNevile" <chaals@opera.com> wrote: >There has been some significant resistance to versioning instead >proposing >the "living standard model". These two concerns don't seem contradictory. Quite the opposite, actually. I argue rapid iterative development needs versioning to be effective. And afaik, the development of the HTML5 spec happens on SVN. SemVer adds a lightweight layer on top of versioning which gives indication of the nature of the recent changes and their degree of stability. >> Recommendation status would be akin to SemVer's 1.0. > >So the first W3C HTML5 draft would be, in such a model, HTML 5.0.0? Either that or HTML5 1.0.0 depending on the case. >> Editor's draft would follow a similar path to the pre-launch SemVer >> notation: breaking changes would get minor update (e.g. from 0.2.0 to >> 0.3.0), non-breaking changes would get patch updates. Minor and patch >> updates could be pushed to /tr immediately. The editor could go crazy on >> the bleeding-edge spec without affecting anyone yet the spec in /tr >>would >> be continuously updated. > >This last bit is something that I think is crystallising as the common >statement of a goal. Absolutely. Is there agreement around this or this wishful thinking on my part? And I understand there are IP implications.
Received on Thursday, 22 March 2012 09:46:51 UTC