OpenSUSE 11.0 was released on 19 July. Version 11.1 was released on 18 December. The release dates are 6 months apart – to the day! That pings strongly into my radar and makes me wonder if time was too strong an imperative during the development of openSUSE 11.1.
Compare these intervals:
10.2 to 10.3 = 10 months
10.3 to 11.0 = 10 months
11.0 to 11.1 = 6 months
Sadly, the product quality is being questioned widely around the Internet. I wonder if this is the price of rushing a product to market.
Perhaps it’s time for a policy change: tone down the time-driven imperative and focus more on quality. Look at the intervals between releases. Version 10.3 is the best release so far. Version 11.0 was a lesser release, understandably because of the enormous challenge of integrating the KDE4 desktop which was and still is very much under development and thus still is a great challenge.
But then the crash came when on top of the ongoing burden of the new KDE4 Desktop from 11.0 , the latest 11.1 release arrived after a scant six months. And not only does 11.1 carry unresolved serious bugs through from 11.0, it arrived with a fresh crop of new bugs. The compounding of these is most unfortunate. Obviously, it’s best to resolve old bugs before rolling out a new release.
For my part, I find the networking in openSUSE 11.0 was down on the networking in 10.3. Maybe that’s understandable. But the networking in 11.1 is down again on 11.0 and the combined degradation is a disappointment.
I say it’s time to build the quality back into the distro that we liked so much and hope to continue to like so much in the future.
It may be that 11.1 was a bit rushed to try to allay concerns with KDE4 which began in 11.0. Also it seemed to me that trying to get it out by Christmas made the release date somewhat less flexible.
So far the one bug which is of most concern to me is the ownership of optical drives. Anybody trying to burn a disk hits this one. My networking setup is rather plain so I haven’t seen your bugs. The modules to open ports for particular services are newish and need more testing. It seems each service contributes a config to SuSEfirewall2, as you can see by doing a rpm -qf /etc/sysconfig/SuSEfirewall2.d/*.
I should also add that from what I have read zypper dup also contributes some bugs since this is new functionality and only possible since 11.0. Obviously it will take a while before in-situ version upgrades work smoothly. Even old hands at this feature like Debian and Ubuntu still have glitches.
While I’ve had no problems that were major, I will say 11.1 has a
slapdashed feel to it. My question though what was the stimuli?
Since the tentative release date for 11.2 is Sep10,2009 I surely hope it’ll be better than this.
I’m with you on this ‘swerdna’. I for one cannot see the need to rush on release. I have 11.1 on my laptop but will not be putting it on my Box. The Laptop is running really sweet now, but from the point of view of catering for the less experienced, I feel 11.1 will have some running away from a bad experience, to the realms of another distro.
I’d rank myself ‘Hardcore’ SUSE user. And even I found myself considering six months of apostasy - towards F10. But then I came to my senses. If all else fails just go back to 11.0 - it worked brilliantly, so long as you dump kde4.0. But as it is, I have 11.1 now and will be content with that.
Incidentally, I am also ‘Hardcore’ KDE. And recent usability issues/bugs with KDE has had me on the fence. I’m running beta 4.2 now, but I dread to think what will happen in the days and weeks ahead.
On a broader note. Issues in general that still need improvement in order to comfort the Newbies among us are: Wireless & Graphics. (I know, it’s not strictly a SUSE issue). Nevertheless, more could be done to help a user. Many are still arriving at NO UI/Desktop. I digress - this is a long standing issue and not related to your point really.
A 12 month release cycle would be fine for me, so long as there is good support via the build service to add in progress in the desktop environment and other areas.
As a GNOME user my experience is that 11.0 is the best release to date.
Apart from the bug at initial release when using reiserfs & beagle I’ve had no issues with the 11.0 release.
openSUSE 11.1 has been good for me except for some small glitches that I hope will be solved soon through the standard update channel.
I think the time driven release has allot to do with SLES 11 having it’s close relation with 11.1.
One thing I’ve said before is that I strongly believe that many issues spring from this being the first release that’s been using the build service to totally build the media.
There have been many issues and glitches when building , causing delays and incomplete media, making it very hard to keep up testing the aplha and beta releases… due to this my feedback was quite limited and I can imagine this is also true for others (testers and developers alike).
When developing and testing such a widespread product it’s important that people can count on release schedules and not have too many unexpected delays and unstable results.
I also think there is more negative feedback now as more people are getting into Linux or switching over to openSUSE. The product is getting more exposure… so again, I hope they bring out fixes soon. It’s been very quiet with updates so far (logically due to the holidays).
Long story short, I’m glad they are going back to the 9 month release schedule. It gives more breathing space.
I hope 11.1 will get back the polished openSUSE feel in the next month… Also it seems like a good thing if the (at the very least) LiveCD’s get a remake as there are mostly negative reports when installing from that media.
11.1 is the best openSUSE release I have used thus far.
I have found 3rd party packman packaged apps are available for 11.1 a lot sooner than they were for ANY previous openSUSE release. As a multimedia fan, I rely heavily on Packman packaged apps.
My view is along the lines noted, that 11.1 (with KDE-4.1.3) was intended to make up for 11.0 deficiencies in KDE4.0.4, and to a large extent I believe it has done that. The die hard KDE3 fans are still not happy with KDE4 (and I’m probably a die hard KDE3 fan), and some of these fans are very vocal in their discontent. Having stated that, despite my being a KDE3 fan, I think KDE4 has come a long ways, … I think openSUSE-11.1 is an excellent example of the big strides that have been made in KDE4 improvement, and I am very optimistic about the future of KDE4, if this current improvement trend continues.
openSUSE-11.1 was released on 18-Dec. 10 days ago. My experience with every SuSE and openSUSE release that I can remember is it takes more than 10 days after the release before a distribution version “settles down”. The beta, alpha, and RCx testing only catches a limited number of bugs, and it is only once a release is opened to the masses, are the bugs that either slipped by, or were inappropriately categorized in terms of priority, are properly addressed and fixed.
In terms of attempted fixes, I believe 11.1 had a limited scope to start with (compared with other openSUSE updates/versions) and hence I do not believe it was rushed inordinately so. If any SuSE release was rushed, IMHO it was 10.1 much much more, than openSUSE-11.1, where 10.1 had the package management disaster, and had a community not ready to support the dropping of proprietary driver support.
No, … I see 11.1 as a good release. The best yet. Just give it a couple more months, give it time like every other SuSE release required, … and then judge it.
To add to the positive note… I agree 11.1 has much potential to become the best yet! Give it a little time and it will be.
But like the girl with the golden goose said, Veruca being the name I believe, ’ I want it NOW! ’
It would have been good though to have more time to wipe out some more bugs for 11.1. 11.2 might be the best of the 11.x series in that sense… for many reasons already mentioned here.
OK, maybe my time would be better spent carefully cataloguing the facts around why Samba browsing in KDE doesn’t work so well in KDE4, and making a clear and helpful bug report about it – rather than venting here tooooo much.
And then wait for the polish to come as oldcpu reminds us is normal for all new releases.
And take heart from the target of nine months rather than six for the next round.
That’s why i prefer to wait for SLED11 to update my linux system.
I am sure these bugs will be resolved when launching the enterprise version. I can’t imagine a company executive complaining on a forum for bugs as they want and buy stability.
For me, the best version is SLED v10 (the original one). Is there a way to update the kernel for this SLED version in order to detect my 3g-cellphone-usb-modem?
That’s why i’m waiting for SLED11… if upgrading the default kernel with the newest one solves my issue, i’d go ahead (and later i will ask how to upgrade the kernel for SLED without breaking my current stable system)
The damage done by releasing a still-buggy product is hard to reverse in today’s competitive open source environment. We mature SUSE users would do well to keep newbies focused on stable versions like 10.3. Once a new user experiences problems during his first installation he/she is unlikely to think kindly of the product. We lose them for good.
I agree with swerdna.I have thought that the Open Suse releases was rushed. I would rather see them come out with a more polished release once a year.
And what oldcpu mentioned proves to me that they are rushing things.Granted a release is always going to have a few bugs. But not a couple months worth.
I will also admit I don’t know what goes on behind the scenes.But it seems like KDE is doing there thing.The kernel developers is doing theirs etc.But it would seem that they don’t communicate between each other.
And that is Linux as a whole.
I also feel that Open Suse is probably one of the best distro’s out there. But it also takes the most time to tweak and set up. As someone mentioned.
I feel the 2 biggest problems with Suse is wireless and Graphics problems. And is keeping a lot of people from using it.
Go to the forum topics “wireless/hardware” and look at the pages of problems. That would tell me there is a problem and it needs to be improved. But has it?
Also right now I am a man without a desktop.>:( I have tried running KDE4. I’m just not ready for it. Or maybe it’s not ready for me.
It’s not to giving in letting you set it up the way you want to. And it is slower than KDE3.
But a little of the problem with KDE4 is a lot of people hate change.And have problems adjusting to it.
When I go back to KDE3 it seems old and dated. But fast and familiar.
So I thought I would give Gnome a try.But I am just a KDE person I guess.
The difference between fedora 10 and openSUSE is like night and day when it comes to setup time. IMHO Fedora 10 requires significantly more setup time. openSUSE is far from the “most” . I can point to other (less popular) distributions as well.
I have always found fedora a SUSE on a par as far as setup and usability is concerned. I was sorry to see Fedora still have, in my opinion, a poor package manager. You cant see the wood for the trees. Makes you realize just How Good Yast and zypper are.
I had F10 set up in a VM in less than 2hrs, full multi-media functionality.
I have been using nautilus for this purpose in 11.0. The problem is not due to nautilus, but to the firewall settings (highports), and it has been there since ever in SUSE.
Just set your firewall properly and it will work.
I agree with oldcpu when he says that we just need to wait for some fixes, exactly as it happened for other releases, and as it happens with all distributions out there. I don’t understand all the bad complaints around openSUSE 11.1, and I didn’t see many of the complainers try to help during the development
11.1 is definitely a poor release. Maybe the worst since I started using SUSE at 9.2. 10.1 had disastrous package management, but once you disabled that crap and installed alternative package management, it was a very solid release. On 11.1 everytime you cough you trigger some bug.
The problem is not the time driven release in itself, nor is it the length of the release cycle. For me the problems seem to be over-ambitious and unrealistic feature planning, lack of respect for the freezes (note that 11.1 even had an unusually long feature freeze, ~2 months, extra betas and whatnot), and perhaps sometimes wrong bug fixing priorities - and a little bit of bad luck, but that’s what you get when running too many risks. All of these problems are probably more or less directly attributable to 11.1 being the base for SLE11.
Novell and openSUSE definitely need to figure out a way in which SLE can be developed without sacrificing openSUSE in the process. Because disappointing users and contributors in a major way with every fourth release (the x.1s) - won’t be made up by the other three decent releases. To grow, you need to consistently do at least decent releases (good or great releases would be even better of course, but let’s be reasonable).
I’ve installed 11.1 on my laptop - but my workstation, which is the production machine for me, will remain on 11.0 for the foreseeable future.
@cb400f: I agree on the bad planning, lack of real testing, lack of respect of freezes. It has been a long time problem around openSUSE, and it makes both developers’ and contributors’ life harder, lowering contributors motivation (and number).
I think one of the problems is that openSUSE is the “de-facto” laboratory for SLE, but only in the negative sense: openSUSE is used to test new features for SLE, it suffers for hurried development and a lot of attempts (sometimes done without a clear idea in mind: read yast modules changing at every release), but does not benefit of SLE testing and stabilization, because this is done independently, and in parallel to openSUSE development, duplicating the efforts for the developers. It sounds like another lack of planning in terms of resource usage at Novell.
That is my view as well. Most complainers are paying less than $100 for their openSUSE. Many are paying nothing.
They have ample time and opportunity to test the development releases.
Instead they simply download after the release, pay little to nothing, and degenerate the efforts of others in their complaints. And exaggerate too.
For example, the big shock with 10.1 was not only the package management hiccups (which had an easy fix with smart) but it was the dropping of proprietary drivers, which is having impact even until today.
That release had far great impact on users than 11.1, and yet complainers of 11.1 still blithely ignore that important philosophical change.
Anyway, I’m leaving this thread. … I’ve got better things to do than listen to complaints that are not looking for solutions.
It’s the old Project Management Cost/Quality/Time thing, except that in this case, as far as most individuals are concerned, the cost dimension is reduced to zero. So, most users perception of the product is strictly 2-Dimensional i.e. how long must I wait and what do I get? If the product falls short of expectations, it must be because the product was rushed to market, mustn’t it? Or is it because expectations were raised to high beforehand? If you raise expectations by over-selling exciting new features you have to deliver on them, or risk a raft of disgruntled users. When the experience doesn’t match the hype this is the kind of fall-out you get. Some insights here,
Personally, I agree with oldcpu. I’ve been with SuSE since 9.1 and I have experienced a “bedding-in” period with every single release. There’s always something, graphics, sound, package management, wireless etc. This time around it looks as if it’s samba. But I anticipate that any current problems will be resolved within the first few update cycles. I like the look and feel of 11.1/KDE4 and I’m sure it will serve me well in future. But speaking as someone from outside of the IT sector software development seems to be a good example of the “cliff-hanger” theory of development i.e.
“Every problem solved introduces a new unsolved problem”
