[rfc-i] Copyright and the Independent Stream

SM sm at resistor.net
Sat Sep 5 14:34:01 PDT 2009


Leslie posted some comments about what copyrights should apply to 
Internet-Drafts.  We can either have a separate discussion about the 
copyrights on I-Ds and RFCs or we can look at them as a whole.  The 
different viewpoints about copyrights, the different streams and who 
is the authority turns the topic into a minefield.

Copying for free redistribution is allowed and encouraged.  That's 
the easy part.  The copyright comes into play when we try to 
determine how to handle derivatives of the work.  There are two types 
of derivatives:

  (i) we reuse the text in another RFC

  (ii) we use the "code" in implementations or in documentation.

Type (i) is not an issue, unless stated otherwise in the RFC, as the 
work is published for the information of the community (documenting 
what gets used in the Internet is always good) and it's traditional 
of a RFC or part of it to be used as a basis for work that may be 
published as a RFC in future.

The entire RFC is the work.  Type (ii) became a problem as we 
attempted to define what's "text" and what's code so that we can 
allow different uses.  The result was two copyrights.  The decision 
to "legalize" the RFC also affected the readability of RFCs as we 
identify what is code and the applicable license.

Copyrights can be ignored and the holder may have to take 
action.  Some organizations who have taken legal action in such cases 
have ended up looking bad.  Community opinion can be a more effective 
tool in rectifying any misuse.

Copyright does not equate with FRAND.  Even if the copyright says 
free to copy, it does not mean that you are free to implement.  It's 
not worth opening that issue.  We ask of the people using parts of 
the work that they acknowledge the source.  This allows the reader to 
review the original work.  It's also a way to give people credit for 
their work.

How many words do we need to say all that in derivative work?  The 
general idea, in my opinion, would be to include an acknowledgement 
that identifies the title of the RFC and its number.  Another 
alternative would be to provide the same information as for the 
References section.

The above would be applicable to the Independent Stream.  There has 
been some questions from the IETF Community about whether the stream 
serves any purpose.  If the IETF wants to retain sole rights to 
critical thought and to the RFC "brand", it has the means to drop 
this stream from the RFC Series.  The fact that we are thinking in 
terms of brand instead of creating an environment which encourages 
the review and implementation of technical proposals speaks for itself.

The Independent Stream has been on hold since 15th February, 
2009.  Is it really that bad to go back to Section 4 of RFC 3978 
until the community can agree on what it wants and the implementation 
details?  I don't think so.


More information about the rfc-interest mailing list