Open Format: Free Format Definition, a proposal
Almost a couple of years ago I wrote a proposal of definition for “free formats” in an Italian mailing-list about digital formats. The idea was quite simple: rewriting the free software definition substituting “program” with “format specification”.
- The freedom to use the format specification, for any purpose (freedom 0).
.- The freedom to study how the format specification, and adapt it to your needs (freedom 1).
.- The freedom to redistribute copies so you can help your neighbor (freedom 2).
.- The freedom to improve the format specification, and release your improvements to the public, so that the whole community benefits (freedom 3)..
.
Freedom 0
Allow you to write programs based upon such specifications, therefore to license such software with any kind of license, free software ones or open source ones included.Format specification based on patent or patent applications containing claims that would necessarily be infringed by an implementation of that standard are critical. Such patents or patent applications must be released under a scheme that is perpetual, irrevocable and royalty-free.
As an example a format specification based on a patent released as Reasonable and Not Discriminatory (RAND) doesn’t comply with the freedom 0.
On the other hand, the W3C Royalty-Free policy is compliant with such definition. As a matter of fact W3C Royalty-Free policy allow format implementation, even if the FSF considers it harmful because of restrictions that might be posed by the patent holder on the “field of use”.
.
Freedom 1
Contained somehow in Freedom 0, and we might keep it for “historical” reasons.
.
Freedom 2
It requires the format specifications to be freely distribuitable, that is necessary if we want to allow anyone receiving data in a free format able to interpret them. As an example if to obtain format specification you need to sign a Non Disclosure Agreement (NDA) freedom 2 is not respected. Still there is space for Standardization organizations to make money of the certification programs and selling official version of such standards.
.
Freedom 3
It should require a copyleft policy, otherwise someone might embrace and extend the standard to create a proprietary version..
Does it make any sense to you?
Roberto Galoppini, 
Simo are discussing.
Simo 5:18 pm on December 5, 2006 Permalink
It definitely make sense.
I’d chop away Freedom n.1 as it is completely covered by 0/3.
The only problem I see is: how do you enforce compliance with the definition?
The specification itslef can be copyrighted, the trademark can be used to avoid abuse of the name and certification purposes.
But the only way to have a sort of “copyleft” license on a specification I am afraid is through a patent covenant on an unavoidable patent that covers all the specification uses.
Simo.
Roberto Galoppini 12:11 pm on December 6, 2006 Permalink
I believe standardization organizations’ policies about Intellectual Property Rights (IPR) must ensure this.
They might do it through patent covenant or releasing free software in order to create prior art and, if needed, be able to cope with patent litigations. Standardization organizations need resources, therefore certifications should be mandatory by law.
Stefano Maffulli 10:55 am on December 11, 2006 Permalink
I think you should join the discussion list for the Internet Governance Forum (IGF) Dynamic Coalition on Open Technical Standards. http://mailman.ctyme.com/listinfo/openstds See you there 🙂
Roberto Galoppini 9:16 am on December 12, 2006 Permalink
Many thanks Stefano, I just subscribed to that mailing-list.