On Sat, May 26, 2001 at 04:22:37PM +0300, Shlomi Fish wrote:
>
> Well, Freecell Solver 1.6.5-autoconf-rev12 which can be downloaded from
> its homepage contains no code from the Microsoft C Run-Time library.
>
> Note that the new code (which is public domain) is based on this code.
> However, so is the code of PySol and kpat, and if I'm not mistaken the
> source code of Microsoft's ANSI C RTL is distributed under a license that
> allow looking at it and writing similar code. Or else, PySol and kpat
> would have been in trouble too.
This is great. I'll update the Debian packages.
>
> However, to become part of Debian, there are also man pages to be
> composed. I can write a symbolic one for fc-solve (which will refer the
> readers to its --help switch and to README and USAGE), and for the board
> generation programs (I'll just say "board_gen_program [decimal number]"
> echoes the board to the standard output).
An extensive man-page would be nice, but Debian does not place restrictions
on the contents and layout of man-pages, so writing a symbolic man page
will satisfy Debian policy.
>
> Now, the question is whether the freecell_solver_user interface should
> also be documented in a man-page. I'm planning to write a document that
> will describe how to use the library, and I suppose I can write it in
> perlpod from the start. In any case, most users will not be very concerned
> with the FCS programmer's API, as most of them are not going to implement
> their own Freecell implementations. But if Debian rules that we've got to
> have such a thing, then we are not in the position to argue.
Debian policy does not require one to document one's API via a man-page, although
such a document is naturally solicited. Whether you choose to write this
document or not, its lack of existence will not hinder freecell-solver's
acceptance into the Debian distribution.
Best regards, Yotam Rubin
Received on Sat May 26 2001 - 08:39:51 IDT