www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Emmanuel Lécharny <elecha...@gmail.com>
Subject Re: Apache's Third Party Licensing Policy
Date Wed, 30 Oct 2013 05:34:28 GMT
Le 10/29/13 8:14 PM, Lawrence Rosen a écrit :
> Thank you, thank you, thank you, Emmanuel, for reminding us of my first response to JIRA
>> in your first response in the JIRA, you wrote : 
>> "CC-BY license is for a specification, not software directly"
> That is absolutely correct. 
> As the W3C attorney explained when he described the purpose of the CC-BY experiment,
/specifications/ and /software/ are no longer easily distinguishable at W3C. The debate about
this was particularly pronounced in the HTML Working Group. I wasn't there at that time, but
this is how it was described when we subsequently discussed it in W3C PSIG:
> "This argument is at the core of a long dispute between Larry Masinter and
> Ian Hickson. If a specification is just abstract rules for implementation, one
> "uses" the specification to implement code. Masinter argued for such 
> specifications. But if the pseudo-code and code serves as a specification
> and as a reference code at the same time, using that code or pseudo-code,
> transforming it automatically into code and putting it in a new context is generating
> legal problems. Problems that nobody wants, but that interfere with the entire
> development process and standardization."
> Like it or not, Apache and everyone else is or will be implementing /software/ by copying
and creating derivative works of pseudo-code in W3C /specifications/.  At Apache, we like
to respect W3C's specification licenses. But the creation of derivative works is expressly
prohibited by the W3C Document License.

First of all, I'd like to debunk a myth :

Psudo-code is *not* code. It's basically what we call an algorithm. That
some layers confurse those two terms is understandable. Larry, you can't
make ths confusion, being a layer *and* a coder.

Now, that some specification is associated with some reference
implementation sounds quite usual for us, Java coders and JSR
implementaters. Hopefully, we all know that the RI is not and should not
be included copycat into our own implementation. This is quite clear,
for years.

Therefore, I don't think we have -yet- a problem. We have to be extra
cautious (and I thik we ae for years) that nobody at The ASF does such a
simple mistake...
> in W3C PSIG we discussed various ways to change the W3C Document License to allow implementation
of W3C /specifications/ in FOSS /software/, namely how to let FOSS communities freely create
derivative works of /software/ within /specifications/ without creating derivative works of
the /specifications/ themselves. We found it impossible to define those terms precisely, and
impossibly burdensome to require engineers to mark the content of their /specifications/ so
as to clearly identify the /software/ portions therein.

Again, we aren't creating derivative work of the specification at The ASF.
> The solutions we discussed included other licensing alternatives, such as additions or
changes to the W3C Document License; a separate addendum (the so-called "PSIG License" in
its draft form) for specifications that contain software or pseudo-code; and two versions
of Creative Commons licenses (CC0 and CC-BY) that both expressed the spirit of free W3C licensing
albeit with entirely different attribution philosophies.
> W3C is now back to discussing licensing alternatives, since CC-BY apparently won't cut
it with this community. 
> I would really prefer it if Apache members led that discussion from the FOSS community
perspective, rather than wait until one of our projects decides to implement some future version
of an HTML or other W3C /specification/ in /software/ under some as-yet-unknown experimental

Again, I don't see the issue here. Just because the W3C is fuzzy about
the way to deal with the code they are including in the specification,
should not have any impact on us ATM.

And again, until we have a project starting to implement anything
produced by the W3, that could possibly infringe any unkwon or known
licence, I would suggest that we keep lobbying at the W3C so that they
decide to clarify their position. That does not have to impact any of
our current policy or activity, IMO. I'm still wondering what such a
thing could have started a flury of inflamatory mails...

Emmanuel Lécharny

To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
For additional commands, e-mail: legal-discuss-help@apache.org

View raw message