And I will strongly oppose any IETF policy that treats commercial or
proprietary code differently from Open Source code.
Out mantra is "running code". We try to stay out of people's business
models.
The presence of a implementation is a useful measure. The presence of
interoperability between two independent implementations is clearly a
much better measure.
But reading the code is not a good measure of the spec. An implementer
familiar with the discussions in the WG is likely to get right tings
that are under-specified. Such an implementation does not tell us
anything about the quality of the spec. Should we require an
implementation done from the spec without talking to the authors or WG?
That would measure the spec better. But would seem to be an
unreasonable requiremet.
he IETF, to the degree it wants to encourage something, wants to
encourage implementation. It ought to be up to the implementors whether
they see value in an open source, closed source, or some other form of
implementation.
One could argue that it does not matter, since the experiment is not
going to change anything. But I will assume that it matters, and will
change something. If so, it should be "running code" that is the issue.
Yours,
Joel
PS: "Free" for whatever definition of "Free"distinct from Open Source
the suggester meant, would seem to be completely irrelevant to the IETF.
On 12/3/2012 12:06 PM, Marc Petit-Huguenin wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
On 12/03/2012 08:54 AM, Stephen Farrell wrote:
On 12/03/2012 04:41 PM, Marc Petit-Huguenin wrote:
I support this idea, but I think that free software should also be
considered as part of this experiment (free software and open source are
not synonymous). Using the acronym FOSS and defining it as Free or Open
Source Software in the document would achieve this.
Fair point. OTOH, there are folks who want to not require open-source too.
For now, I'd gone with "ideally open-source." As I said to Sam, let's deal
with that, if need be, at IETF LC.
OK, but note that I will oppose for this experiment the use of software that
is not either available in source form or accompanied with a conformance test
suite.
In the meantime, I've posted -01 [1] with a bunch of changes. Do let me
know if I forgot to ack someone and as before all comments (inevitable,
this being a process thing;-) are very welcome.
Thanks, S.
[1] http://tools.ietf.org/html/draft-farrell-ft-01
- --
Marc Petit-Huguenin
Email: marc(_at_)petit-huguenin(_dot_)org
Blog: http://blog.marc.petit-huguenin.org
Profile: http://www.linkedin.com/in/petithug
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=30oW
-----END PGP SIGNATURE-----