Weekly Edition
Return to the Front page
| |||||||||||
Fedora release cycles: longer or shorter?
The Fedora 10 release is currently planned for November 25 - somewhat later
than had been originally intended. Delays in Fedora releases are certainly
not unheard-of, even when the project isn't coping with a major compromise
of its fundamental infrastructure (the full story of which, it should be
noted, still has not been told). Fedora 10 looks like it will be
worth the wait, but the project is not waiting for the release to start
thinking about its upcoming release cycles. A couple of discussions
related to this topic provide some interesting insights into the pressures
being felt by Fedora's leadership.
A recent video review of Fedora 10 was seen by the project as being something other than entirely favorable. But the biggest complaint expressed by the project is on a different subject: credit for work which is done by Fedora developers. Quoting Fedora leader Paul Frields:
Another point that had me scratching my head was the same host
indicating that Fedora had a lot of features that were in Ubuntu
8.10. This is certainly true, but the differentiator is that many
of these features were *built* by Fedora contributors, inside and
outside Red Hat. It's important for us to keep emphasizing this
fact.
Subsequent discussion indicates that a number of Fedora developers feel that other distributions - Ubuntu in particular - are stealing Fedora's thunder by shipping Fedora-developed improvements first. This is not the first time this kind of concern has been raised; it has been asserted that Novell's behind-closed-doors XGL work was done that way to keep Ubuntu from shipping it first. Fedora does not appear to be considering pulling its development from public view - that would run counter to the project's open nature - but some other responses are being discussed. More than anything else, the Fedora project would like to ensure that the world knows about the work its developers are doing. Initiatives like the feature list for each release help to get information out ahead of the actual software release. There is also talk of more aggressive blogging, outreach to news sites, etc. The project has even posted a proposed marketing schedule which would help to ensure that all the right marketing activities are happening at the right points in the release cycle. Former Fedora leader Max Spevack had a different suggestion to offer:
If "features" and "first" are hurting because of where we are in
the calendar compared to the Ubuntu release, allowing them the
chance to release their new distro first and to receive a lot of
credit for new features when reviewers and press don't understand
where the upstream work is being done (in Fedora, for example),
then Fedora Marketing should ask the Fedora Board to think about
altering our "May Day" and "Halloween" release targets by a little
bit, so that Fedora's cycle finishes before Ubuntu's.
This proposal brings to mind a vision of distributors racing to be the first to release, leading to ever-shorter cycles and a corresponding decrease in release quality. It is hard to imagine that the first mover has such an overwhelming marketing advantage; there must be a better way. It does not look like Fedora will attempt a "first post" counterattack anytime soon. In fact, if the recently-posted Fedora 11 release schedule proposal is adopted, the exact opposite will happen. In the past, Fedora has responded to a much-delayed release by shortening the following release cycle in an attempt to get back on schedule. For Fedora 11, it would appear that this will not happen; there will be no attempt to go for a "May Day" release. The reasoning against shortening the Fedora 11 cycle comes down to this:
Fedora 11 will be extremely important to Red Hat Enterprise Linux
(otherwise known as RHEL). RHEL 6 planning has looked to use
Fedora 10 and Fedora 11 as releases to work out new technologies
and features that are desired in RHEL 6. This includes a lot of
upstream work that is being done, and targeted to land in these two
releases.
So a shortened Fedora 11 cycle would make it harder to get all of the changes planned for RHEL6 in. That's problematic for Red Hat, and, since Red Hat pays for much of Fedora's existence, Red Hat's problems become Fedora's problems. Beyond that, though, it seems that a number of core Red Hat engineers will be working on Fedora during the next cycle to help get RHEL6-targeted features into shape. If the next cycle is shorter, Fedora will get less attention from those developers. Fedora would like to avoid that situation and take advantage of the RHEL team's attention while it can. So the proposal is to retain the six-month cycle for Fedora 11 and release around the beginning of June. The Fedora 12 cycle, though, would be shortened to get the project back to the original schedule. The hope is that the advance notice will make it easier to plan for a short release cycle; Jesse Keating also suggests that the project "could even focus more on polish issues in F12 than large sweeping features." The more cynically-minded among us might conclude that Fedora 11 will be stuffed full of bleeding-edge new stuff that the RHEL team wants to evaluate, and Fedora 12 will be the release where all of that work is actually stabilized. But your editor would never want to be cynical. The initial response to the proposed schedule is almost entirely positive, so it seems likely that things will go that way. Some Fedora developers may feel that releasing behind Ubuntu gives the project a public relations disadvantage, but other concerns are seen as being more important. Since those "other concerns" can be seen as "take the time to focus a lot of work on pulling together new features for an upcoming stable release," this set of priorities seems hard to argue with. (Log in to post comments)
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 2:47 UTC (Thu) by zooko (subscriber, #2589) [Link]
Did anyone bring up Mark Shuttleworth's offer to synchronize Ubuntu and Fedora release schedules?
http://www.markshuttleworth.com/archives/159
It seems like that would solve the problem that is bothering Fedora folks, quite in addition to the "combined forces is greater than the sum of its parts" idea that Shuttleworth advocates on that blog post.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 2:57 UTC (Thu) by sbergman27 (subscriber, #10767) [Link]
Very true. It also might help if the devs were not quite so fixated on the topic of getting "credit". That Linux distros are advancing is somewhat more important, in the grand scheme of things, than the credit accounting that Fedora devs always seem to want to focus upon.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 3:09 UTC (Thu) by chromatic (subscriber, #26207) [Link]
I can understand the desire for credit, but surely some of the blame rests on the fact that most reviews of new distribution releases are exceedingly bad. (Perhaps I have little in common with the intended reader, in that I find installing an operating system unbearably dull.)
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 10:35 UTC (Thu) by nim-nim (subscriber, #34454) [Link]
Credit is a big motivator for free software work. People can do free or even gratis contributions but they want at least to be credited them.
To be completely cynical, increasing the length of the Fedora cycle would have the same effect as trying to race Ubuntu. Ubuntu largely relies on Fedora to do the ground work, when they release shortly before Fedora they can claim the credit without expending a lot of resources since the Fedora features are almost finished.
It remains to be seen if Ubuntu will have the resources to stabilise Fedora stuff for the Ubuntu release date, when the Fedora release date is far enough Fedora people haven't gotten to this stage yet.
Fedora release cycles: longer or shorter? Posted Nov 14, 2008 4:39 UTC (Fri) by interalia (subscriber, #26615) [Link] It remains to be seen if Ubuntu will have the resources to stabilise Fedora stuff for the Ubuntu release date, when the Fedora release date is far enough Fedora people haven't gotten to this stage yet. I don't see how this makes sense. If Fedora delays their release and feature X ends up not being ready, then the next Ubuntu release won't have it. The one however, might. Fedora lives on the bleeding edge. This means that Ubuntu is probably always going to get the credit. If a new feature was stable, then Ubuntu's short release cycle means it gets to release with the feature first. If the feature doesn't prove stable, then likely Ubuntu may not include it but Fedora will. Then it bakes a bit more, Ubuntu puts it in their next+1 release and gets credit for "getting it right". I'm not sure Fedora can win either way, as long as they decide to be a bleeding edge distro.
Fedora release cycles: longer or shorter? Posted Nov 14, 2008 7:59 UTC (Fri) by nim-nim (subscriber, #34454) [Link]
> I don't see how this makes sense. If Fedora delays their release and
> feature X ends up not being ready, then the next Ubuntu release won't > have it. The one however, might.
But the next Ubuntu one will be four months after the Fedora one, leaving ample time for reviewers to understand who did what.
Fedora release cycles: longer or shorter? Posted Nov 17, 2008 13:36 UTC (Mon) by Ze (subscriber, #54182) [Link]
It's not a matter of winning.
If Fedora ships with it after Ubuntu , the code gets tested more and debugged more. I think when Fedora 10 ships the 3G HAL support will be better.
Yes NetworkManager 0.7 supports 3G devices but without the appropriate HAL support for hardware that people use it's a PITA. This reminds me I need to submit the HAL information for the MF636 3G USB Modem.
The Kernel drivers work with it but it just needs the HAL support so NM 0.7 knows about it.
Honestly the whole Fedora vs Ubuntu thing strikes me as a bit of a rehash of the old Redhat Vs Debian rivalry.
I can see it from both sides though. I'm sure some folks in the Fedora camp aren't happy about how popular Ubuntu has become. I'm also sure that Canonical and Ubuntu aren't doing themselves any favours by not developing launchpad in the open (open it all ready , an open source company developing a closed source application isn't really a good image).
Mark Shuttleworth also failed to understand it's not who has the most users that decides the direction of the future , it's who contributes the code. If you want Linux to go in one direction then you need to contribute the code for that direction and keep committing it and people will join you if they like that direction.
The motto for linux isn't "Show us the money", it's "Show us the code".
Fedora release cycles: longer or shorter? Posted Dec 20, 2008 2:17 UTC (Sat) by boss (guest, #55702) [Link]
Hi,
You say, "This reminds me I need to submit the HAL information for the MF636 3G USB Modem.".
I've just got this modem and I've having trouble getting it working under Ubuntu 8.10. Any chance you can post, or send, your configuration to make it work?
Thanks,
- Robi
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 9:19 UTC (Thu) by AlexHudson (subscriber, #41828) [Link]
If someone else gets the credit for the work you do, why would you bother doing more work in the future? I think you're missing how disheartening that actually is, particularly to volunteer contributors.
Sync? not worth it. Posted Nov 13, 2008 3:12 UTC (Thu) by mmcgrath (subscriber, #44906) [Link]
That is very true, Fedora publishes its schedule and tries to stick to it. I'm sure it wouldn't be that hard for the Ubuntu guys to sync up.... But of course therein lies the problem. Who decides the release schedule? The canonical folks don't have a great reputation for giving back or playing well with the ecosystem. What if Ubuntu's schedule slips and Fedora is ready to release? What if it's by more then a month? Two?
It's a "wouldn't it be nice" thought but just doesn't align with reality. The costs are fairly high affecting what features get in, which ones get cut and when. The benefit? PR? Something else? It's muddy at best.
Sync? not worth it. Posted Nov 13, 2008 6:38 UTC (Thu) by tajyrink (subscriber, #2750) [Link]
Ubuntu schedule is directly aligned to GNOME's, and so far Ubuntu has always released on schedule (with the exception of 6.06 LTS which was quite early decided).
Sync? not worth it. Posted Nov 13, 2008 8:18 UTC (Thu) by rahulsundaram (subscriber, #21946) [Link]
GNOME schedules themselves were originally aligned to Red Hat Linux.
http://fedoraproject.org/wiki/Releases/TimeBased
We are coming full circle.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 3:09 UTC (Thu) by bojan (subscriber, #14302) [Link]
Keep 'er steady, boys! No need to panic, just keep being brave and releasing new stuff. Anyone that has a clue will know where the stuff is coming from...
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 11:23 UTC (Thu) by eru (subscriber, #2753) [Link] Anyone that has a clue will know where the stuff is coming from...Unfortunately, that means the set of people who knows is rather small :-(. I don't know how, but somehow Ubuntu has managed to publicize itself so well that outsiders could be forgiven for thinking Ubuntu invented desktop Linux. Fedora really should "advertise" its achievements a bit more.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 14:33 UTC (Thu) by skitching (subscriber, #36856) [Link] Anyone that has a clue will know where the stuff is coming from.Agreed. Yes there are a lot of basic desktop users now who don't have this "clue". But the people who matter are
Those two categories are the people that need to be impressed; the others are just unimportant - or even a burden. I don't mean that making linux usable by Aunt Tillie is a bad idea - but it isn't a big deal if Aunt Tillie chooses some other distribution either.
Fedora release cycles: longer or shorter? Posted Nov 14, 2008 6:43 UTC (Fri) by chema (subscriber, #32636) [Link]
I think the problem on this is that a huge amount of "people who matter" usually choose what they are familiar with.
See Wikipedia example, for instance. I think the problem here is not Aunt Tillie but young Bob...
In my experience, I have seen some, let say, "enterprise" decisions about which distribution to use, only based in people's past experiences. And it is reasonable also, they are not looking for the best solution, they look for a solution that fit their needs.
--
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 6:11 UTC (Thu) by drag (subscriber, #31333) [Link]
One of those things that Ubuntu has so far failed to do is do a very nice integrated Server-Desktop combo.
That is a product similar to Microsoft Small Business Server. One of the chief advantages that Microsoft with Active Directory is that everything is pre-configured out of the box to work well with it. That is if you have Windows Vista, Microsoft Office, Active Directory, and Exchange server and, I suppose (even though I have no experience with it) Sharepoint.
Everything is designed to work together so you can get a fairly full-featured enterprise-ish desktop/server/application system up very quickly with little hassle and relatively little training.
It's very 'easy'.
Getting a similar thing in Linux is deadly difficult for most people. LDAP server, GUI management tools, calendering, etc etc.. There are components that work and are reliable, but just don't integrate easily and there are big gaps in functionality. It's hard and it's spotty.
------------------------
Ubuntu is having a hard time gaining traction in business, and that's were the money is (since Ubuntu is trying to follow a support-oriented business model similar to Redhat). But due to the lack of application certifications and whatnot Ubuntu is not gaining traction well.
Well so far Linux is good in the server room, but in the business it has a huge perception problem about being very difficult and expensive to deploy Linux desktop solutions for the desktop. Lots of management issues and such.
So what Ubuntu should be doing, if they want to be taken seriously, is focusing on the desktop, server, and domain integration. Sure they have a nice desktop and home users use it, and Ubuntu makes a decent server, but that's not were the money is and most people who are going to be interested in Linux are going to be using and supporting software/computers professionally. They use the software they use at work and that's were most of their energy is focus is going to be targetted at.
----------------------------------
Well with Fedora you have things like Fedora directory server and management tools.
Samba4 is coming along. Openchange is very nice also... (not to be confused with OpenExchange) their stuff really does a good job of integrating with Microsoft Exchange. As in 'it works'. Calendering, MAPI, and all that. When trying to use Linux in a place that uses Exchange it's always a struggle and Exchange's web interface sucks big donkey. But OpenChange is working on a evolution plugin that will solve all this.
Then there are things like FreeIPA. As well as Redhat's purchase of Qumranet as well as Redhat's web-based virtual machine management system.
-----------------------
If Fedora can release a functional equivelent to SBS.. were you have a server with Email, Domain controller, virtual machine management, SQL server, etc etc... and then you have a client Fedora version that tightly integrates with it. (As well as Windows client integration).. And the devil is in the details.. like you can allow user access to a SQL server on the server so they can play around with OpenOffice.org's DB integration.
Like a Fedora Re-spin or whatever it's called.
I think that sort of product would be met with massive interest.
If the a average admin, who is used to managing Windows systems, can get a old box out of the closet, install "Fedora Server" on it, and then dual boot "Fedora Client" and get a domain up and running with email, VM, and all that good stuff with minimal knowledge of Linux then that's a huge advertisement for Redhat and Fedora.
It doesn't need to be perfect and make no illusions about the cutting-edge nature of this sort of thing, but if you give people something that people can really sink their teeth into then users will like it very much and it's not going to be something that Ubuntu will be able to duplicate any time soon. (since most of the software I described isn't packaged by Debian yet..)
Fedora release cycles: longer or shorter? Posted Nov 14, 2008 7:41 UTC (Fri) by jengelh (subscriber, #33263) [Link]
>One of those things that Ubuntu has so far failed to do is do a very nice integrated Server-Desktop combo.
Maybe because that's just not their target. Live with it. Sounding great does not always imply being great.
>Ubuntu is having a hard time gaining traction in business,
I do not think so. It is already in the business — much to my dismay, because it seems like a totally immature choice. (As you pointed out, it is lacking the certs.)
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 6:45 UTC (Thu) by tajyrink (subscriber, #2750) [Link]
I think it should be noted that the best advertised feature in Ubuntu 8.10 desktop edition, easy 3G support, was developed by a Finnish summer-coder (http://www.coss.fi/web/coss/developers/summercode). I don't know if it has been integrated into Fedora yet (libmbca and the related operator database).
Of course, Network Manager 0.7 itself is largely a Red Hat project, but Ubuntu didn't even include it before and Fedora has had two release cycles (Fedora 8 and 9) of time to take credit for NM0.7 sweetness as such.
Same goes for Fedora's kernel-based mode setting, GEM support in Fedora 10 - Fedora-unique features.
But I agree people should not focus _that_ much on the credit issues, but FLOSS advancement in general. And others are free to copy anything community-wise from Ubuntu, because that's what has made it so popular, not the technical aspects.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 10:44 UTC (Thu) by nim-nim (subscriber, #34454) [Link]
The difference, is that Fedora makes the effort to merge all its stuff upstream, where it can easily be copied by others, whereas Ubuntu generally does not feel like upstreaming the config tweaks which are most of its contributions is useful.
In other words, Fedora's biggest problem is that it is a good community citizen, while Ubuntu benefits from not being one.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 12:54 UTC (Thu) by tajyrink (subscriber, #2750) [Link]
I think this has been already discussed and people agree to disagree. From my point of view Canonical is mostly doing what it can with the current amount of employees, and they are both aiming to improve still and are also hiring more people even they may never match even half of the amount Red Hat has. But I also agree that Fedora is a _very_ good citizen, partially because of the resources it has and partially because their vision of freedom. I hope Fedora will gather more user community and helping hands everywhere.
Fedora release cycles: longer or shorter? Posted Nov 14, 2008 23:04 UTC (Fri) by jspaleta (subscriber, #50639) [Link]
Dan Williams mentored the student project.
http://www.kaijanmaki.net/blog/2008/03/24/networkmanager-...
Part of the project plan included a Ubuntu package specifically as a deliverable.
Let me note something very clearly. A Red Hat employee, upstream NM project developer, and Fedora contributor agree to mentor the project. A project with a plan which included Ubuntu deliverables. Isn't that interesting?
Now cynics would suggest this was a missed opportunity for Fedora. It's not and let me tell you why. The important thing here was that this work was driven upstream as a primary goal of the project...not a secondary one. That's is part of Fedora's principled approach to building a distribution..to move as much innovation into upstream projects as possible because that's how we all benefit. The student, the Ubuntu community member, got the priorities right. If only Canonical employees could understand the importance of setting those priorities in the right order.
I'll leave you with this hypothetical. If a Canonical employee had been the mentor, instead of a Red Hat employee who is deeply involved in NM development, would libmbca be an upstream GNOME module right now? Or would it be stashed in the corner of launchpad along with a set of patches against NM?
-jef
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 7:12 UTC (Thu) by qg6te2 (subscriber, #52587) [Link] The more cynically-minded among us might conclude that Fedora 11 will be stuffed full of bleeding-edge new stuff that the RHEL team wants to evaluate...This has always been the case. On one hand Fedora can be obtained for little or no cost, with the users then paying for it in other ways (things breaking between upgrades, packages shipped before they're ready for consumption, updates which cause regressions, etc). On the other hand, Red Hat pays for a large chunk of open-source development (thereby making life better for everyone), with the money coming from support contracts for RHEL, which itself is a polished version of Fedora. Ubuntu fills the niche between the quality of RHEL and Fedora, though at the expense of coming across as a leech (in a number of cases this portrait is deserved). At the same time a typical Ubuntu release is much more polished than a typical Fedora release -- a big part of the reason for Ubuntu being so popular. If Fedora doesn't like to have their thunder stolen in terms of features, or indeed becoming irrelevant, the answer is simple: don't release half-baked goods.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 8:17 UTC (Thu) by rahulsundaram (subscriber, #21946) [Link]
You got to realize this. If Fedora hadn't included what you refer to as "half baken" goods, then those goods will reach other distributions in a much much worse condition.
Being in the forefront of new development or being conservative is a choice. Either choices has their benefits as well as costs. Assuming that being conservative is the only valid choice is wrong.
Regardless of the amount of testing that happens in development releases, Fedora frequently does find and fix bugs after those features gets into a general release. They get fixed in Fedora and in upstream as well. Many of the Fedora users understand this and are happy to be involved. Other distributions should also be thankful someone else is trail brazing development on behalf of them,
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 14:02 UTC (Thu) by kh (subscriber, #19413) [Link]
Excuse me if this changed already and I simply did not notice, but maybe the best thing Redhat/Fedora could do from a marketing prospective is to have an executive at Redhat state they have decided to go after the desktop market.
(The last I remember hearing, Redhat's (then?) CEO said they were avoiding the desktop market altogether... why should anyone then be surprised that there is more enthusiasm for a desktop project from a company who's leader has made the desktop their primary focus?)
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 15:11 UTC (Thu) by rahulsundaram (subscriber, #21946) [Link]
I don't Red Hat or Canonical disagrees much on the profitability of consumer desktops
http://www.press.redhat.com/2008/04/16/whats-going-on-wit...
http://blogs.computerworld.com/ubuntus_shuttleworth_i_don...
If there is already common understanding in the business prospects, there is then, a common challenge on making it more viable. Red Hat continues to invest heavily in desktop technologies and I am sure there is a commercial interest there for now and in the long term as well. Executive announcements just for posturing won't help.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 19:13 UTC (Thu) by drag (subscriber, #31333) [Link]
Yes.. The desktop still is critically important for Linux.
Right now Microsoft (according to common wisdom) controls the only viable platform for business desktop. Because of this they get more hardware support, more application support, more everything then Linux.
It fundamentally cripples the ability for Linux to compete in a huge and profitable market, even if your goal is to make money from servers. Microsoft has control of the desktop platform and because of this they are in a position to control how it operates and they use this to their advantage.
Fedora release cycles: longer or shorter? Posted Nov 13, 2008 15:50 UTC (Thu) by michaeljt (subscriber, #39183) [Link]
Lengthening a couple of release cycles sufficiently would also get Fedora releases earlier than Ubuntu ones...
Fedora release cycles: longer or shorter? Posted Nov 14, 2008 7:46 UTC (Fri) by jengelh (subscriber, #33263) [Link]
>Subsequent discussion indicates that a number of Fedora developers feel that other distributions - Ubuntu in particular - are stealing Fedora's thunder by shipping Fedora-developed improvements first.
There we have it, black on white (— depending on your browser configuration). No merit.
>More than anything else, the Fedora project would like to ensure that the world knows about the work its developers are doing.
Fedora /could/ switch to a license that includes the advertising clause — "this product contains Fedora software" or similar.
Fedora release cycles: longer or shorter? Posted Nov 15, 2008 1:43 UTC (Sat) by rahulsundaram (subscriber, #21946) [Link]
Fedora is very unlikely to do any such nonsense. Moral obligations if any are different from legal ones. Trying to confuse them would lead to the path of XFree86. If all software Fedora includes demand such credits, we would obviously go nowhere. So let's not even suggest using the legal tool to do the job. I don't a silver bullet solution to what essentially appears to be a marketing problem but this certainly isn't it.
Fedora release cycles: longer or shorter? Posted Nov 15, 2008 11:18 UTC (Sat) by superstoned (subscriber, #33164) [Link]
Of course they can't... Most of the stuff is GPL. And their strong values
and beliefs (FREEDOM above all) prevent them from choosing anything else. It would alienate their community, and even if it didn't, I can hardly see Red Hat do such a thing. I'm not a huge Red Hat fan, but I know they strongly believe in true Free Software.
|
|||||||||||