W3C home > Mailing lists > Public > public-html@w3.org > March 2011

Re: The right to (easily) fork

From: Ian Hickson <ian@hixie.ch>
Date: Thu, 10 Mar 2011 20:05:34 +0000 (UTC)
To: Lawrence Rosen <lrosen@rosenlaw.com>
cc: 'HTML WG Public List' <public-html@w3.org>, 'PSIG' <member-psig@w3.org>
Message-ID: <Pine.LNX.4.64.1103101957230.944@ps20323.dreamhostps.com>
On Thu, 10 Mar 2011, Lawrence Rosen wrote:
> 
> I understand what you want. But that wasn't a requirement in any of the 
> use cases presented by the HTML WG to PSIG. Nothing was said about a 
> requirement to make it easy to take a spec out of W3C and develop it 
> elsewhere.

If we number the nine use cases in the requirements e-mail consecutively 
starting from 1:

   http://lists.w3.org/Archives/Public/public-html/2009Feb/0093.html

...then use cases 6 and 7 are the ones that require what I describe.


> It isn't me you have to convince; it is the membership of W3C that voted 
> overwhelmingly in a survey that such a generous permission slip would be 
> harmful to W3C and to standards generally.

Actually, it's the other way around: it's you who has to convince me.

We already have a license that applies to this specification that allows 
essentially anything at all. Forking is already unambiguously allowed. The 
idea of having a license on the W3C spec is just so that we don't need to 
refer people to the WHATWG spec to get this permission. If the W3C's 
license doesn't allow forking, reuse, etc, then we'll have to continue 
referring to the WHATWG one as we have been doing.


> But we've been on a treadmill around here over a new W3C HTML document 
> license. Adding a new use case at this late date which requires that W3C 
> bless the forking of its specifications will, at the very least, delay 
> any resolution.

It's not a new use case. It is the very first use case. It literally 
predates any work on HTML5 whatsoever, because it was an impediment to 
starting the work in the first place. There has been no ambiguity over the 
years that this has been discussed that simple forking of the 
specification is one of the important use cases.


> Obstinacy, even when you're right, can sometimes delay progress.

The problem is that forking isn't allowed. If the solution doesn't allow 
forking, then it's not progress.


(Amongst other requirements was the requirement that the license be 
compatible with the GPL, which "option 3" is not. So even ignoring 
forking, this proposal still doesn't address the requirements of the WG.)

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Thursday, 10 March 2011 20:06:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:25 GMT