[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Outgoing section 5.5 and draft-josefsson (Re: San Diego meeting slot)



This note seems to assume that most of the code in RFCs comes from existing sources and is subject to external license restrictions.

At least in my experience that is not the case.  The code consists of
Code written by individuals for the RFC to expalin processing
MIB definitions written for the RFC
XML Schemas written for the RFC
ASN.1 written for the RFC
ABNF written for the RFC

Remember, the RFC is not normally used as an implementation library. We are not trying to provide sample (much less reference) code. We are trying to provide enough information that people will build interoperable implementations.

We have agreed that Open Source, GPL, commercial, and other users need to be able to use the code from RFCs.
Personally, I think that including code with a GPL in the RFC would be counter to that goal.


Yes, there are a few RFCs where commercial code has been included in the RFC. There may be a problem for such cases in the future. But allowing arbitrary restrictions on the code in RFCs is much, much worse.

Yours,
Joel M. Halpern

PS: Remember that authors can still make available anything they want, with any terms and conditions they want. So an author can still have a copy of the same code with GPL conditions. But if the code needs to go in the RFC then the author needs to make it available according to the IETF rules.

At 10:54 AM 10/5/2006, Simon Josefsson wrote:
Brian E Carpenter <brc at zurich.ibm.com> writes:

>>>   However, the IETF believes it is confusing for IETF Contributions
>>>   to contain additional copyright notices and licenses, and wishes
>>>   such material to be external to IETF documents.
>>
>> The above change (if approved etc) would make it harder to include any
>> source code in IETF documents.  Agree, yes/no?
>
> No! it means that if you want to attach a license to source code
> in an RFC, you do so external to the RFC.

Oh, I see.  It sounds like a good idea in theory.

However, it would make it impossible to use code licensed under many
software licenses in IETF documents.  I'm not sure if that was what
you intended?

For example, take the BSD license.
http://www.opensource.org/licenses/bsd-license.php

  Redistribution and use in source and binary forms, with or without
  modification, are permitted provided that the following conditions
  are met:

    * Redistributions of source code must retain the above copyright
      notice, this list of conditions and the following disclaimer.
...

Similar conditions are found in many licenses, including the GPL, the
MIT license, and the Apache license.

I believe we have two choices:

1) Require that code in RFCs are released under the at-publication
   then-current IETF copying conditions.  The IETF license is
   incompatible with most free software licenses, so we'd see a lot
   less source code in RFC with this option.  None of the RFCs with
   code in them today would have been acceptable, as far as I can
   tell.  I consider this a serious problem, and something that would
   be against the goal of the IETF.

2) Accept source code with other licenses.  We place some requirements
   on the licenses, but looking at the currently accepted licenses in
   RFCs, I don't think we can find any lowest common denominator.

If someone can think of other alternatives, that would be very good.

/Simon


_______________________________________________
Ipr-wg mailing list
Ipr-wg at ietf.org
https://www1.ietf.org/mailman/listinfo/ipr-wg