Don Armstrong wrote:
On Mon, 29 Dec 2003, Jakob Bohm wrote:The main trick is to distinguish between the original full text SRFI ("the document") and the free software (document that excerpts or derives from the document).Sure, but if you take that tack, the prohibition of modification of "the document" becomes, in effect, a no-op. You can take excerpts of the document, modify them, slap some more changes on them, and then call it SRFI foo. According to this theory, it's no longer "the document."
I think this line of reasoning mistakenly blurs the line between technical and social restraint. The license bans modification of the document, and is written in a generic way so as to be applicable to any single SRFI. So when applied to an SRFI, what is "the document"? The given SRFI. The license clearly allows you to derive works, as long as you do not change the SRFI itself. That leaves an unfortunately fuzzy, but still clearly small area which you shouldn't touch -- and which nobody has a reason to touch anyway: conflicts with the original document. There's no good reason to release a new SRFI 86 instead of an SRFI 286 which references and quotes SRFI 86. So what's ruled out by this license? Well, calling a derivative work of SRFI 86 "SRFI 86", or otherwise making something that appears to be a modified SRFI 86 would probably count. I can't see anything else as forbidden.
Is your real problem, Don, the vagueness of the identity problem for documents? When is one document the same as another? When is one a modification of another, and not a separate document?
Not that all of this matters, given the strange grant of permission for derivation. I didn't notice that weirdness in my first replies, sorry.
-Brian