Re: proposed reduction in graphviz dependencies

[Jens Rehsack <rehsack%googlemail.com@localhost>]

Hasso Tepper wrote:
> Greg Troxel wrote:
[...]
>> Objections?  Jens - what do you think?
> 
> I'd prefer separate packages where possible instead of options. It's much 
> more usable for end users using both source and binary packages. Without 
> digging too much into it and looking just a PLIST file it seems to be 
> doable for most of graphviz options.

As I wrote to Steven, fell free. Using options is much less to do for me.
I'll keep such proposals in mind and if I have time, I'll see what I can do.
But to look forward, what's to do else:

- update graphviz to have current release (with fixes and feature
  improvements)
- updating doxygen and import DoxygenFilter
- add support for xulrunner/libxul
- keep p5-* modules up to date, add p5 gnome modules
- Update bootstrap-mk-files to reduce file copying to tune one or
  two options for a specific OS/compiler
- Updating CPAN releases of
  + SQL::Statement
  + Proc::Processtable (with NetBSD- and FreeBSD- support higher
    than 2.0/5.0 - read: kvm-interface)
  + Sys::Filesystem (HPUX-Support)
  + DBD::SysTables (SQL access to a lot of system tables)
  + DBD::Anydata (modernize)
...

Sorry when I'm a bit disquietingly when we're discussing about more or
less options of a unloved package (who wanted to touch it before I took
maintainership?).

We don't talk about real problems here, do we? We're talking about Xaw and
librsvg dependencies on the one hand and swig and related options on the
other hand.

Jens