Re: [ELPA] New package: repology.el

[Dmitry Gutov]

On 26.01.2021 07:59, Richard Stallman wrote:
> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
>
>    > > I think they are software, because they could be modified within an
>    > > infinite space of possibilities.
>
>    > A picture can be modified within an infinite space of possibilities. A
>    > book can be modified. A mathematical equation can be modified.
>
> That is true, but you've taken the question I raised for myself out of
> its context.  The real question is, "In the context, is there any
> doubt that this is a piece of software?  If not, what else could it
> be?
>
> I think it is clear that a specification of a data structure, meant to
> guide a program's operation on that data, is software.  It is
> comparable to a bunch of struct and enum declarations which is how we
> use C header files to show the structure of other data.

I'm going to say that it's not a program, in the same way as images are 
not part of a program code. Unlike enums that have a role in writing the 
code and its compilation, and thus describe aspects of a program, 
schemas are structured documents which contain information pertaining to 
other documents.

There are different things a program can do with a schema: it can 
determine whether a certain document is valid, or it can output some 
sort of structure describing the elements that are missing in the 
document, or it could even try to generate a sample document based on 
that schema (though it would require a fair amount of supporting code). 
This list is almost certainly incomplete.

So a schema is not a program, nor a part of a particular program.

That said, if we always require that accompanying data (such as images) 
to be distributed under free licenses, the "software or not" question is 
probably moot.