~kanarip

I'd be a big supporter of this if it was only security fixes after 13 months witch it seems to be. I hope it works out.

I would welcome it and participate in such an initiative, as one who maintains a decent fedora deployment at work...

Hopefully you'll find this is more constructive than dismissive...

To me, the issue of long term support is that it is at odds with the 'leading edge' goal of the project. A long term support target - at first look - appears to conflict.

I think another recent discussion on the development list touches on a better way to accomplish the same thing; the elimination of releases.

The issue with Fedora is that just about the time a release gets stable, its end-of-life is reached and an upgrade must be performed. Upgrades have always been problematic, and many choose to (at least periodically) perform a clean install. This is of course the antithesis of long term support.

If instead of having distribution releases, packages were continually updated and perhaps obsoleted by new packages the goal could be achieved. There would never be an upgrade, so the long term support would be an intrinsic attribute of Fedora. Packages would be continually updated, so the leading-edge goal would not be compromised. Development would be a continuous process. From a user's perspective, change would be gradual and incremental.

Installation media would be cut before any given push - a snapshot of the current state, always up-to-date, eliminating the need for re-spins.

I think the antagonistic nature of the recent discussions on the developer list prevented proper understanding of the benefits to Fedora that the elimination of releases would achieve. It should be seriously revisited.

Just doing a quick search of my e-mail I found some in jan '08, oct '08, and feb '09 (geez, three threads in 13 months, I wonder why I'm not thrilled to have this discussion yet again). If I remember correctly, the discussion from October was also escalated to FESCo & the Board:

https://www.redhat.com/archives/fedora-devel-list/2008-January/msg02191....
https://www.redhat.com/archives/fedora-devel-list/2008-October/msg00921....
https://www.redhat.com/archives/fedora-devel-list/2009-February/msg00652...

Saying there was never a discussion seems rather disingenuous to me.

This feature would be very useful for university CS departments like ours which run labs of Linux workstations for an academic year (Sept-Sept). At present we have the choice this Sept of deploying F11 and changing mid-year to F12, or deploying F12 beta and dealing with what breaks. Others might either deploy F11 for the whole year and hope they don't get compromised, or simply avoid Fedora altogether.

Hehe, yes, it is that time of the year... exactly the time when I am brutally forced to upgrade 3 laptops and 2 desktop machines
So, thanks for making the effort. Too bad I am not technically able to help in any way. Still I hope it will be accepted, it will save me so much strain to reinstall all 3 laptops at home:( I hate this procedure

We at the Swiss Federal Institute of Technology Zurich (ETHZ) follow your progress closely. We definitely want Fedora (instead of the currently used RHEL) on all workstations for students and all laptops. The main issue we face there is the 13-month-only life cycle. We probably could deal with this since it's not so problematic to reinstall boxes once a year - but we'd be happy if we wouldn't have the pressure to actually have to do it, leaving us the option to maybe only update every 1.5 years or so.

I would love this. I'm tired of people at work suggesting I use one of the Ubuntu LTS releases so we don't have to upgrade every 6 months when our installed boxes go EOL.