RE: assimilating OpenBSD
My thougts are as follows:
There are two BSD based systems that I find very very interesting:
FreeBSD
Mac OSX
OpenBSD is great (I use it on a firewall here).
Honestly I wouldn't consider puting anything but that onto our firewalls.
I think that an interesting point is that MacOSX is going to very quickly
gain quite a bit of ground in the unix market. Once people realize how easy
it is to port things such as Gnu WGET, lynx, OpenSSH, etc. it's going to
become a VERY reasonable and optional unix desktop machine.
FreeBSD is gaining quite a bit of ground but lacks a bit in application
managment. (Ports are sort of a pain!) :)
It would be nice to have the option to install the 'Debian OSX Packaging
system' or the 'Debian FreeBSD packaging system..'. The problem I see with
OpenBSD is that it's going to be dificult to attract users and future
package managers and maintainers to the system if the core OS scares them
off. As OpenBSD being more bare-bones, less support on hardware.. etc, will
happen.
What I do propose is that we look at OpenBSD as a target for the package
managment system ports and use definately use it as a potential 'Full Based
debian Distro' and fork the entire project maintaing the core system (kernel
modules, sysv architecture, etc) but rewriting or implimenting debian
installation tools. (which IMHO a -HUGE- project outside the scope of the
4-5 people that are really interested in doing work).
If we take the few people we have and break them off into two groups.. the
packagers for end user applications and maintainers for the ports of the
core debian packing tools for the operating system we should move quickly.
Obviously one has to happen before the other.
Another notch for FreeBSD over OpenBSD as the 'base' of the movement is the
ports directory. It's going to take a -LOT- of time to base the packages
and building of the packages from tar-balls. there are a LOT of tools out
there and a LOT of work has already been done in the port projects. We can
leverage the work that has already been done by re-implimenting the built
port packages into the deb's. This would allow us to get a lot done very
quicly. Maybe on some of the software we will feel that we can do a better
job building it than the BSD team. (I personally doubt that we can).. but
thats not really the scope.
Once we get the debian software ported over we can immediate start thinking
of porting it to the other BSD systems.. I am not sure how different they
are or if they are even (generally) binary compatible or library compatible.
(I suppose if the programs where all linked statically wouldn't be a
problem...) MacOSX seems to go on the 'Static Build' approach and keep the
software not so dependant on libraries, etc. I like that idea a lot as
disk-space vs. price is growing by orders of magnitude (in terms of cost per
megabyte) and unix console applications just aren't getting much bigger.
But now I am starting to tread into water that I would hardly consider
myself expert is really something we don't have to think about for a while.
Cheers
.mark
> > system. I think that we have to do it in a way that will keep
> the DebianBSD
> > in line with the movement of the BSD system we are following.
> Maybe the end
> > result is a agnostic BSD based package managment system.
>
> agnostic? I am not sure i understand.
>
Reply to: