Re: Upcoming Releases Schedule...



2008/9/20 Jo Rhett <jrhett@xxxxxxxxxxxxxxxxx>:

On Sep 19, 2008, at 7:07 PM, Aragon Gouveia wrote:

To get a business to commit resources to a project there must be an
actual goal.

[1] The FreeBSD project would have to commit resources too. Its community

Of course. This is what the requirements analysis is ;-)

For (a), (b), and (z), this is where you come in. Define the goal. Make
a
plan to get there. Assess the effort involved. Convince your employer
that
(a), (b) and (z) is worth it to him/her and that the result of (z) will
convince the FreeBSD project to commit the resources needed to integrate
it.
If they're happy, start working on (z) and bring it to the FreeBSD project
when you think it's ready.


Of course. If this was something that could be done without working with
the freebsd developers, do you think I would put up with this kind of abuse?
I'd much rather have something I could just go and do ;-)

The issue is that nobody is willing to answer the question: "what resources
are too limited to provide longer support? How can we help?"

This the elephant that everyone ignores. To develop a plan, you need to
know the limitations. Once those are spelled out, you sit down and try to
determine what resources are necessary to achieve a certain goal. Then you
find those resources, etc etc...

Without input from the current release team extending the support schedule
is not possible.

--
Jo Rhett
Net Consonance : consonant endings by net philanthropy, open source and
other randomness


_______________________________________________
freebsd-stable@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscribe@xxxxxxxxxxx"


I have been following this thread for a while now and I've only
recently used FreeBSD for the past 1-2 years, so my experience is not
so great.

But from what I've gathered from reading the mailing lists during the
EOL of 4.11, the limited resources are

From Kernel Space
- Security patches
- Drivers (sometimes back porting a 7.x driver to 6.x isn't possible
because if missing API or limited API calls that are only available to
newer kernels)

Base/Core Software
- Security Patches

Ports (I think the biggest issue in maintaining support)
- Ports maintainers have to keep patches for every RELENG thats
available. Previously they had to maintain compatibility with 3-4
branches (i can't remember), 4.x, 5.x, 6.x, 7.x. and there is alot
(10,000+) ports.
- At the moment there is two (i think) 6.x and 7.x, and possible mid
next year (2009) there will be 3 (8.x).
- Some ports didn't work with 4.x due to missing API calls as well and
had to be marked broken if they tried to compile it on a 4.x. This
will eventually happen when FreeBSD reaches 8 or 9 and 6 will not have
the API calls required.

From what I gather, those are the main points in maintaining extended support.

How long are you expecting support for a RELENG to last, 1, 2, 3
years? 5 years? (comparison, Ubuntu LTS is 3 years, security updates)
Are you after support for a RELENG_X or RELENG_X_Y?
What are you expecting from the support? Security only? Drivers? Ports?

Please don't get me wrong, it is great that you and your company is
willing to put the resources in, but its a huge undertaking
maintaining a release. Maybe those questions might clarify to some
FreeBSD Developers what you're asking from them.

NB: I'm not a FreeBSD developer, but a very happy user that maintains
FreeBSD servers for business clients.

Regards
David N
_______________________________________________
freebsd-stable@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscribe@xxxxxxxxxxx"