Hi,

        I'm interested in packaging PVM for Debian. First, let me say that
I'm aware that Drake Diedrich <[EMAIL PROTECTED]> is working on
it, but I've had the intention to package "something big" for Debian for
couple of months (I've packaged small stuff locally) and I think this is
be a good candidate since I need it for my work and I'm getting tired of
"woops, I forgot to do THAT on THIS machine" kind of problems. Probably
Drake can do a better job, but I'd like to try anyway. What I'm trying to
say here is that I don't intend to make my packaging the "official" one, I
just want to give it a shot. 

I've read the relevant documentation (debian/doc/package-developer) and I
have a couple of questions.

* Directory structure

This thing wants to be installed with a tree structure like this: 

        $PVM_ROOT
        +- bin/LINUX
        +- console/LINUX
        +- include
        +- documentation and stuff like that (tests, examples)

That breaks the FSSTND. I can try to convince it to forget about the LINUX
part (it's required for NFS hosted installations on a mixed OS
enviroment), i.e., that means reading and fixing the code with each
release... not funny. I checked some packages that are problematic on this
respect, like Acroread, I've found that they get installed in
/usr/lib/program/whatever_the_program_wants. Is that OK with the FSSTND? I
may be misinterpreting the document, but IMHO, I don't think it is. 
/usr/lib says libraries, not entire programs.

* Where do I have to develop things?

The machine I use to compile and test pvm has bo; I cann't convert it to
hamm, it's out of the question (I'd like to, but cann't). I have access to
a least three computers running hamm. If I want to make the package
official, I have to build it using hamm, don't I?

* PVM has a very loose license (basically it says "you can use it provided
that you keep this copyright appears in all copies ..."). Should a get
something more specific or is it enough? (For an official package, that
is)

TIA,


                Marcelo E Magallon <[EMAIL PROTECTED]>
                Astrophysics Lab -- University of Costa Rica


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Reply via email to