John Halton wrote: >>> OTOH, if it is just a case of making a program that meets the spec., >>> and the program itself is free and does not contain the spec. itself, >>> then I don't see that's a problem. (See the recent discussion here >>> concerning a program that implemented a non-free RFC.) >> Unless you want to distribute the specification itself (in main)! > > Well, quite. But in the case of the RFC, the consensus seemed to be > that there was no problem including the program that implemented it in > main, provided the actual RFC text was removed from the source. Sure, the program merely implements the specification, so the code license rules if the spec is not included. > But in the case of an XML specification, perhaps implementation > inevitably involves including the text of the specification itself, > i.e. the XML code (unlike the RFC, which was just an English text > document). This is where I shrug and go, "But I'm not a programmer, so > I dunno". ;-) Sometimes a great deal of text is included, for example to create a stringprep library most coders would include the large text tables from RFC 3454 (with or without some reformatting). Sometimes a snippet of text is included or a protocol example is included -- think "literate programming". Typically I think there is no need to include the entire specification in a code library if all you are doing is coding to the spec (e.g. you can just point to the canonical location for the spec). But still it would be nice for the spec to have a license that enables a coder to include it if desired (something we are still working on in the Jabber community....) Peter -- Peter Saint-Andre https://stpeter.im/
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature