On Feb 25, 2010, at 3:42 PM, IOhannes m zmoelnig wrote:
On 2010-02-25 20:07, Hans-Christoph Steiner wrote:
I don't see the harm of them going into /usr/include, is that
related to
Debian Policy? I haven't found anything along those lines.
i don't think it's a policy violation.
but
- these headers are _private_. (as a matter of fact, i'm not sure they
should be included at all)
This stuff is so messy in the Pd sources, I think its hard to say that
its a private header. There should be a distinction between private
and public, but there isn't really in practice.
- miller has a _very_ ugly naming scheme, that is non-standard at
best.
the names of the header-files are meaningless and will only cause
confusion for about everybody else.
I can't argue with the ugliness. I think the problems caused by
omitting the header will cause more confusion than including it. I
don't think many people browse /usr/include for headers that to include.
- i think, header-files should actually be packaged in a puredata-dev
package
Yes, I agree. I have no objections to that idea whatsoever. I think
that should happen in conjunction with the creation of a pd-common
package that includes pdsend, pdreceive, the directory structure for
installing, and the /etc/init.d/puredata /etc/default/puredata scripts.
I think it is still worth it to accept this version of the 'puredata'
package as a stepping stone towards the puredata-dev. idea.
.hc
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
