Ubuntu may move to monthly releases like Firefox

[t3g]

http://www.pcworld.com/businesscenter/article/239804/will_ubuntu_linux_switch_to_a_monthly_release_cycle.html

There is a story floating around that Canonical may move Ubuntu releases to a monthly release much like Google Chrome and Firefox. For the many Linux distributions based off of Ubuntu, I don't know how this can be a good thing since teams like Linux Mint spend a month or so to tweak the OS and tailor their own intentions to it. If Ubuntu does indeed move to a monthly release, how does this affect the Ubuntu branch of Linux Mint? Essentially there would be no need for Mint based off of Ubuntu because Linux Mint Debian does about the same thing with the monthly update packs. Granted, there is a greater software selection and support for Ubuntu in comparison to Debian, but the Mint team cannot be the only ones affected by this.

People want stability in an operating system and usually do not care if they are using a 10 year old OS like Windows XP as long as it works and programs run on it. For Linux being already a niche market and Ubuntu being the popularity leader, this is a huge risk for Ubuntu and distributions based off of it. Its already a pain to be told to wipe my OS every 6 months with the current release cycle and it is even worse with Linux Mint due to its update manager not allowing full upgrades. I honestly want to be full time LMDE, but that project is not mature yet and the rolling release isn't as polished.

Does this mean that the Linux Mint releases should be based off of Ubuntu LTS and constantly tweaked with point releases? If not, is the Mint team ready for drastic changes to the Ubuntu codebase every month and having to release a Mint version because of it?

[craig10x]

Reading the article is a bit confusing...do they mean you would have to re-install your system each month or just do a monthly update "pack" that has been thoroughly checked over by ubuntu, making it essentially into a rolling release where you wouldn't have to re-install your system every 6 months anymore?
And that they would also release them as monthly isos for someone installing for the first time so they would be immediately up-to-date....

If they mean rolling with monthly update packs, that sounds pretty good...only thing is i wonder is how ubuntu based distros like mint would be able to work with it...

[xenopeek]

The article is not saying this will be done. It is based on a blog post from a Ubuntu team member, who shared his thoughts on this and is writing he will put forward this proposal. Not a done deal.

Ubuntu is already taking hits with ignoring part of its established community, due to drop in quality assurance on its recent releases and forcing Unity. I'm not confident it will initially bring more stability to Ubuntu if they move to a monthly release cycle, especially as Unity still has a long way to go to become stable (perhaps 11.10 is already there, but might take another release). Of course, the agile way is to do frequent releases of your project, at a set interval of 4 to 6 weeks for example, and deliver those features that are fully tested early, with features still needing work being pushed back to a next month. That would be an improvement over the forced delivery of a clearly unfinished (not release ready) product that was the 11.04 release. It may work, but if Ubuntu switches to this get ready for another year of instability and regressions...

It is a good move to cut off people basing their distro on yours But LMDE is already doing a monthly release cycle. But there are some critical steps missing from LMDE yet, like one-click disk encryption from the installer that Ubuntu has. Who still uses a laptop without disk encryption? So there is still a gap of usability between Ubuntu based Linux Mint and LMDE. If Ubuntu forces a departure to another distro as base for Linux Mint, that might be the time these gaps can be closed with renewed focus away from Ubuntu.

[ThistleWeb]

Yeah it is a blog post by an Ubuntu member suggesting it, it's not Canonical's decision to adopt it. Canonical are shaping Ubuntu towards enterprises, because those are the folks who pay for support contracts and make Canonical profitable.

No enterprise with a halfway competent IT department is going to adopt a rolling distro of any sort, so if Ubuntu changes to be be rolling, even with a monthly quality assurance fallback, they will kill any chance they have in the enterprise market in one swoop. Mozilla killed Firefox as a browser fit for enterprise deployments with the insane quick release idea.

There's a reason RHEL is a strong enterprise player, it can be tested to within an inch of it's life to make sure it's all fine. Updates only apply security fixes, they don't upgrade applications to the next version, potentially breaking something that worked the day before. Enterprise deployments need their PCs to be available at full capacity for the workers to do their jobs and make the company money, service their customers during working hours. The last thing they need is to arrive for work only to find some crucial software they need doesn't work now because of some update. All of that is meticulously planned out, including upgrade cycles.

Most home users either don't need paid support because they can support themselves, or can get easy fixes posted by Linux and Ubuntu community folks. Home users are of limited interest to Canonical because they are limited in how they can be monetized. App sales, UbuntuOne subscriptions, and search engine data are about it.

I can see the appeal of having newer versions of software faster, but to me that just means Ubuntu shifting towards more of the Fedora end of the spectrum. The problem is that it's not the market Ubuntu target. Ubuntu are aimed squarely at human beings; regular Joe's. Regular Joe's don't care how something works, they just want to use it. It's a means to an end. They wanna chat on Skype, the PC allows them to do that, they click the icon, sign in and chat. For these people, they don't know what the latest is, or any reason to have it, or the unstable risks that come with it. They care that it works, that it's stable.

Ubuntu has been accused of becoming less stable recently, less extensively tested etc. It's not always fair but in plenty of cases it is. The switch to Unity is a huge deal, so there will be some dips in quality control in the meantime, but increasing to a monthly release will just make things worse.

I seriously can't see this proposal getting anywhere among the decision makers at Canonical. They will know doing so will kill any enterprise plans they have stone dead.

[altair4]

If you read the original source: http://netsplit.com/2011/09/08/new-ubun ... e-process/
Excerpt:

So you’re an Ubuntu developer working on features for the upcoming release, you don’t have anywhere near as much time as you’d expect to actually do the development work. What happens if you’re replacing something that works with something completely new? Can’t you just target a later release, and work continually until the feature freeze of that release?

It turns out that you can’t. There is an incredible emphasis on the Ubuntu planning process of targeting features for particular releases. This is the exact thing you’re not supposed to do with a time-based release schedule.

Unfortunately Canonical’s own performance-review and management is also based around this schedule. The Ubuntu developers so employed (the vast majority) have such fundamentals as their pay, bonuses, etc. dictated by how many of their assigned features and work items are into the release by feature freeze. It’s not the only requirement, but it’s the biggest one.

Your new feature is going to take twelve months of development time to fully develop before it’s truly a replacement for the existing feature in Ubuntu. What you don’t do is spend twelve months developing and land it when it’s a perfect replacement.

What you do do is develop it in 12-13 week bursts, which means it’s going to take you roughly four release cycles before it’s ready rather than two. And you land the quarter-complete feature in the first release, replacing the older stable feature.

Anything that improves that process is a good thing.

[ThistleWeb]

And you land the quarter-complete feature in the first release, replacing the older stable feature.

Ah yes, the bleeding edge option of replacing something that works, that's feature complete and stable with a quarter complete alternative. I'm either missing something, or isn't that the reason your average Joe doesn't use a bleeding edge distro? If Ubuntu intend to go to a rolling bleeding edge model, they will not only kill their enterprise chances, but their home userbase as well, to distros long established along those lines like Fedora.

On the other hand, Canonical doing this would see Mint being DDOS'd with new folks who don't want to be guinea pigs.

[altair4]

My quote above describes the release process that Ubuntu uses today.

The suggestion by the blogger is to change it so that things are released when they are done on a monthly basis incorporating some wild out-of-the box concepts like this one ( again same blog post ):

I just introduced a bunch of new checks to the developer process there; I just introduced code review, mandatory unit tests and then piled functional tests and verification tests on top.

[zerozero]

and i find Jorge Castro's answer to that point symptomatic of Canonical priorities

Jorge Castro says:
September 8, 2011 at 11:35 pm

“I just introduced code review, mandatory unit tests and then piled functional tests and verification tests on top.”

Looking at the amount of resources available on one hand, and the amount of work required to do just this one line of your plan, how would you proceed with this?

Don’t get me wrong, I dig the Chrome-style approach here, but you’d have to at least start with a very small subset of the distro and be brutal about it, do we really have “the pieces already there”?

basically "we don't have the resources to implement quality"

[pluraldave]

Ubuntu's issues aren't caused by the release schedule, but by the culture at Canonical to push things before they are ready. Changing the release schedule won't solve the issues if the culture isn't changed, but if the culture is changed then there is no need to change the release schedule.

[Robin]

Ubuntu's LTS releases fill the need of people who require stable desktops (now built from Debian Testing rather than Unstable), and there's always SalineOS, Mepis, and Debian Stable as well. The bleeding edge stuff that Fedora and the regular Ubuntu releases is known for is for testing I'm sure, not for desktop users like me who require stability. If they were to jump to the rolling release model and still have the 3-year LTS releases, they might be offering the best of both extremes.

[Habitual]

Who cares?

[xenopeek]

Habitual speaks the blunt, but honest truth Eh, just let it happen already and we'll tackle this thing when and if it sees light of day...

[altair4]

The original poster of this topic asked the following question:

I don't know how this can be a good thing since teams like Linux Mint spend a month or so to tweak the OS and tailor their own intentions to it. If Ubuntu does indeed move to a monthly release, how does this affect the Ubuntu branch of Linux Mint?

It's a dilemma I would suspect since that's different from what happens now. Currently what passes from Ubuntu after the initial release are bug fixes which can have it's own detrimental affect on Mint. But the proposal would now pass either incremental changes to existing function or new function altogether. I'm not sure how Mint would react to this change.

The author of the proposal suggests that the changes would be slight and gradual and that may be true for Ubuntu itself but I for one have no idea what the work content is to convert Ubuntu to Mint.

[DrHu]

Aaahhgg!!!
It's bad enough how they mess with 6 month releases, greater panic will ensue if they ever contemplated this maneuver
--perhaps it's only an April fool's joke..

[altair4]

I wish people would read the original blog entry before having what is understandably an emotionally negative response:

http://netsplit.com/2011/09/08/new-ubun ... e-process/

A lot of what he is proposing are things you probably thought Ubuntu was already doing - and you would be wrong