On 11/23/2014 09:51 AM, Jeff Jaffe wrote:
[adding PLH]
There may be several different issues that we should all try to clarify.
1. Byte-for-byte copies. Sam's proposal envisions the use of
byte-for-byte copies. But he also states that these copies must meet
the needs of W3C pubrules. These might be in contradiction; I'll
explain the contradiction and I'll explain how I interpreted Sam'e
proposal.
The contradiction is in the document license. WHATWG publishes with
CC-0 and OWFa, and W3C publishes with the W3C Document license and
occasionally CC-BY. How can it possibly be a byte-for-byte copy if some
of the bytes (document license) are different?
My interpretation of Sam's proposal was that he meant that the documents
would be byte-for-byte copies with the exception of the document
license. Sam, is that a correct interpretation?
My original suggestion was byte-for-byte with the exception of links
(e.g., stylesheet) and references. I also invited everybody to suggest
revisions or alternate proposals.
2. Contributions. W3C requires that contributions come from
participants of the Working Group so that we are on strong ground in
declaring RF. If there is text that is proposed from outside the WG, it
is a responsibility of the Chair to make sure that we are not bringing
in text that is encumbered by patents.
In this proposal, I am assuming that the Chairs would need to fulfill
that responsibility even for text coming in from the WHATWG.
In the case of URL, with most/all of the text coming from Sam and Anne -
both of whose companies are part of WebApps - this does not appear to be
a large burden.
I agree that this isn't likely a problem.
3. Source of original document. There are some documents that are
currently under restrictive licenses such as the W3C Document License.
The W3C community (specifically the Member organizations - the AC and
AB) has not approved forkable licenses (with the exception of the
limited CC-BY experiment). This might be what Anne is referring to. If
you take some existing W3C document (e.g. TTML) which is only published
under the W3C document license and use that as the source of a revision,
Anne is correct that it is not trivial to make that available under a
permissive license.
To the best of my knowledge that is not an issue for URL because my
understanding is that we are talking about a document that is not
encumbered by a previous copyright restriction (to be sure I have not
asked for a legal review).
I think Anne is correct that it would be difficult to implement Sam's
proposal for an existing document such as TTML. From a legal point of
view, W3C owns the copyright and has legal authority to make it
available under a different license. However, I doubt that W3C would
feel comfortable doing that without the agreement of the AC/AB - and I
would not be optimistic.
I don't think this is a problem.
4. Ongoing modifications. Here is another potential issue. Let's say
that a major revision of the spec is done in W3C, and W3C publishes a
new draft under a restrictive document license. The next day, WHATWG
wants to publish a copy.
While arguably that could be a violation of W3C's Document license, I
don't think it arises in Sam's proposal. As I understand it, all
modifications happen first in the WHATWG (with its permissive license)
and are then copied into W3C. So there is already a permissive license
associated with any version of the document prior to its coming under
restrictions of the W3C document license.
I don't think this is a problem.
What appears indeed to be a problem is the text in section 7b of the
Member Agreement, and section 2.2 of the Invited Expert and Collaborator
Agreement. I've asked about this on the public-w3process list:
http://lists.w3.org/Archives/Public/public-w3process/2014Nov/0156.html
Jeff
- Sam Ruby
On 11/23/2014 7:12 AM, Sam Ruby wrote:
cc += Jeff, Arnaud
On 11/23/2014 04:05 AM, Anne van Kesteren wrote:
On Tue, Nov 18, 2014 at 9:18 PM, Sam Ruby <ru...@intertwingly.net>
wrote:
Would it be possible to fork https://github.com/whatwg/url into
https://github.com/w3c/, and to give me the necessary access to
update this?
I'm not sure why this escaped me, but the reason this doesn't work,
provided you want to continue to contribute to the WHATWG, is that the
W3C prevents it. I was after this kind of setup in 2012, and Jeff and
Wendy told me in no uncertain terms that it was in violation of the
Member Agreement. I then became a non-Member, but the Invited Expert
and Collaborator Agreement prevents the same thing.
So if we are to go through with this, I would need to something in
writing from the W3C and IBM (the actual Member Agreements are private
and may have differences per company as I understand it) that this is
in fact okay.
I believe that there must have been a miscommunication; especially
given that Jeff has indicated[1][2] that he doesn't see any changes to
the W3C process required to implement what I plan to do[3].
Just to be clear, I've been doing my work on my own machine. Nobody
seems to have a problem with that. I push it to my site
(intertwingly.net under CC0). Nobody has a problem with that. I push
it to github under my name. Nobody has a problem with that.
The W3C is willing to consider the contribution. The W3C has
different ideas on licensing which I'm not thrilled with, but can live
with.
Michael Champion has proposed that we establish a separate
repository[4] for shared work. Nobody has a problem with that.
Hopefully this miscommunication can be resolved before my rewriting of
the parser work[5] is complete.
- Sam Ruby
[1]
http://lists.w3.org/Archives/Public/public-w3process/2014Nov/0148.html
[2]
http://lists.w3.org/Archives/Public/public-w3process/2014Nov/0155.html
[3] http://intertwingly.net/blog/2014/11/20/WHATWG-W3C-Collaboration
[4]
http://lists.w3.org/Archives/Public/public-w3process/2014Nov/0149.html
[5] https://www.w3.org/Bugs/Public/show_bug.cgi?id=25946
