Some considerations about 12.1

  1. **SMOLT: **
    IMO the 12.1 has a severe bug. The smolt database does not start automatized with the upgrade from 11.4. (At least for me it stayed off). All compatible install information will be lost. Particularly, as a user experiencing a smooth and satisfying update will not have any second thought on launching SMOLT, the most compatible hardware combinations of long term users will be lost. Suggestion for 12.2: SMOLT should launch automatized after an upgrade. 1. The install should be preceded by clear indications of hardware requirements
    . E.g. on my Athlon XP 3600 with 3 GB ram and SATA-150 HDD, Nvidia Nforce 410, 11.4 will run flawlessly while 12.1 will not be able to reproduce DVDs and Videos given the enormous resource requirements of the nepomuk framework. This should be taken into account when giving minimal hardware requirements for the standard desktop KDE. 1. Maybe the system should warn about this during install when detecting low performance hardware
    (e.g. only one core) and offer to de-aktivate from the install nepomuk and strighi. This would be probably not bad as users will then avoid to have a hanging system and try to find the reason. This is IMO particularly important for Linux as a lot of users take advantage of the OS to give new life to aging hardware (if they wouldn’t care about money and cost, they would probably by new material and often stay on what is installed, arriving only after some time to unsatisfactory performance and the wish to change). 1. Core programs of KDE, in a final version, should be backward compatible
    and **working migration programs **should be written PRIOR to the roll out, or migration procedures should be published and proven prior to the release, known to be working. The bigger part of the problems of 12.1 that lower the overall impression of reliability and viability are based on errors apparently well know before the roll out. So, if systemd was the reason for delaying the goldmaster, why wasn’t it delayed by such an important issue than a non working kdePIM. KDE is called SC that is software collection. If only the desktop should work when doing a roll out I propose to drop back to the KDE desktop environment designation. All other things would be wishful thinking. So, the reason of this point being so IMO, paramount is that Linux is getting more important for a lot of users for day by day work. So there is a lot of prior personal data available and must be migrated (if one intends Linux to be an OS and KDE to be a desktop environment. If it is instead a technical toy, then things are O.K. just as they are. 1. There have been voices saying: wait for a month or so to go by. I do not subscribe the “sit out policy” on bugs. I do this for my laptop, waiting for KDE4 PIM suite to get production quality…since 11.1. Useless to say, I am still waiting. I think it is not a problem to find bugs hidden and not visible in a final release. It is actually motivating But I think we all agree that the KDE PIM status was not acceptable for the final release. The system should be viable and users
    should find bug, but in my view at least, they should not face real **“showstopper” bugs **right from the install. 1. there has been some confusion about the DVDs on sale,
    Balsam Professional and Opensuse DVD, parallel on the market here in Europe. It would be good if from the project there would be an official approach to what are these products . If there is an official sale of the openSUSE DVD to raise funds, this should be clearly indicated as users might wish to actively contribute to the project while buying such a product.

Conclusion: this 12.1 release is overall a very good release, why do you let such few issues make it look less consistent? And why do you loose out an unique opportunity to create a better hardware database?
Well, I am sure you will tell me.

PS. I will see if there is some consensus on these points and if there is, the ones where we do most agree upon I may file them as features in openFATE (even if may say it does not help much, I think it could, if one uses it). At least I am still optimist. :wink:

Much as I like KDE as a DE, there’s always a lot of KDE stuff that I disable/decide to leave alone from the start: Nepomuk/Strigi (use Recoll), KDEpim (useless, doesn’t communicate with anything outside its own solipsistic universe), anything remotely related to KOffice or its more recent descendants. Kwallet too.

I think that is getting a problem. Let us assume that KDE4.8 will bring performance advances. In itself the idea of Akonadi, Nepomuk/Strigi is very appealing, in the sense that it does detach the data from the interface. Writing or installing Thunderbird or another future mailer for KDE4 could be a snap. If it would work. Now, I cannot expect as a user to have “a perfect system”, especially not from the project KDE. The project is the project, but it is the distribution that should care for the harmonious and well debugged roll out. Sometimes this may (and this should!) imply a regression on a lower version if at the time of beta, GM and release it is evident (as it was with KDE PIM) that the quality is more beta if not alfa. Saying: I do not use it anymore does not help, especially if one wants to continue the desktop KDE in the future. This because the whole underlying logic of the desktop is the integration of software, rightly a SC (software collection). So while the KDE PIM of 3.5 was to be considered a vendor lock in (as to an extend is thunderbird or worse evolution), the KDE4 mentality and setup should give users an easy choice over the proper GUI. Once the akonadi framework works. Now that makes several editions that we assist to sever usability regressions of KDE4. There are two issues that can be discussed on this point. a) the KDE approach is too complex, will never be usable and therefore is abortive. b) the KDE approach is feasible, current manpower invested in it is too low, because resources are attributed uniquely to the tablet project. The latter point might be the case because the recent decision of the QML adoption. And there we can see (wiki):

QML (Qt Meta Language or Qt Modeling Language[2]) is a JavaScript-based, declarative language for designing user interface–centric applications. It is part of Qt Quick, the UI creation kit developed by Nokia within the Qt framework. QML is mainly used for mobile applications where **touch input, fluid animations (60 FPS) **and user experience are crucial.
(emphasis added).
When I read this I am even more convinced about the latter aspect. If this will then be of any advantage to the notebook user, I doubt it. Still, technical evolution runs, and I prefer to have KDE also on the table then only on the PC. Iff…that it is reliable and fast. It is the last point that I am currently missing.
So if it should be the case that manpower is scarce, it would be, IMO an asset of a distribution to ship around it, even if this means sometimes to delay and not to use “the latest and greatest”.

On 2011-11-23 14:36, stakanov wrote:
> Conclusion: this 12.1 release is overall a very good release, why do
> you let such few issues make it look less consistent? And why do you
> loose out an unique opportunity to create a better hardware database?
> Well, I am sure you will tell me.

Who is that “you”? Us? The “us” here are not the people that made 12.1.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

Hope is last to die. One supposes (although apparently going on a website to have user feedback is impossible for developers because causing flatfeet, wrinkles and orchitis) that maybe one of the creators of the distribution is having a look here. Let me be optimistic. But in the essence you are right. Notwithstanding I strongly belief it is suicidal to do a product and never have a look on the opinion of the users.
BTW, do you know whether SMOLT is still “alive” as idea? Because I really think this is actually a bug, similar to the publication bias (for who of us that did never hear this, science suffers from a bias because the negative results are less likely to be published as are the positive ones). Here it is a bit the contrary. As SMOLT does not seem to come up when all goes well by update, the machine is not counted and compatibility data of hardware is not risen, especially for the most fitting machines.
Am I wrong?
P.S. when we use the distribution, when we communicate wishes, suggestions, when we show bugs and file a bug report, isn’t this also us that make the distribution? I see this at the end as a big advantage of Linux, although too many are “just free riding” and do not participate in the effort.

On 2011-12-13 23:36, stakanov wrote:
> BTW, do you know whether SMOLT is still “alive” as idea?

I don’t know.

Anyway, I always thought that the openSUSE implementation of SMOLT was
faulty, because the profile has to be renewed periodically or it is removed
from the database. That needs a cronjob; there is one, but it is set to run
the first of the month at 1:20. If the machine is never running that day
and minute, the profile is lost.

> Because I
> really think this is actually a bug,

Then report it in Bugzilla. I was thinking of doing it myself when I noticed.

> P.S. when we use the distribution, when we communicate wishes,
> suggestions, when we show bugs and file a bug report, isn’t this also us
> that make the distribution? I see this at the end as a big advantage of
> Linux, although too many are “just free riding” and do not participate
> in the effort.

Yes, it is a community effort, and there are many way of participation.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

It would be nice if the stupid thing would recognize my nvidia card and a blank black screen were not the order of the day.

I still have 11.2 on my laptop because newer versions do not recognize the intel video chip.

Having been caught before, doing upgrades and having to fall back to a previous version, I am testing the installation on a spare hard drive.

I know that most of the fault comes from X11 and Suse is not the culprit but perhaps the install should warn that the hardware is problematic for the version.

This problem is pandemic to Linux in general, not Suse. It is sad to say that if I want to use my computer it looks like I will have to stick to M$ Windoze. I have things to do other than trying to strong arm Linux into cooperating. Suse 11.2 was the last version that installed and worked first time. Since then things have gone down hill and the hill seems to be getting steeper.

It would be nice if the stupid thing would recognize my nvidia card and a blank black screen were not the order of the day.

I still have 11.2 on my laptop because newer versions do not recognize the intel video chip.

Having been caught before, doing upgrades and having to fall back to a previous version and re installing all my stuff, I am testing the installation on a spare hard drive.

I know that most of the fault comes from X11 and Suse is not the culprit but perhaps the install should warn that the (video) hardware is problematic for the version.

This problem is pandemic to Linux in general, not Suse. It is sad to say that if I want to use my computer it looks like I will have to stick to M$ Windoze. I have things to do other than trying to strong arm Linux into cooperating. Suse 11.2 was the last version that installed and worked first time on the 3 computers I use. Since then things have gone down hill and the hill seems to be getting steeper.

Your problem is likely more the absence of SAX than the overall problematic of the Nvidia drivers. It is not even X11 the culprit, much more Nvidia itself when the new drivers do not work well. They could have opted time ago to join the effort for a working and well established opensource driver (as Intel did mainly for their notebook products. I would be surprised that this mentality did actually help them to maximize profit, given the performance of ATI video card. An opensource driver would have been a “buying-decision-helper”. Instead we are living in a world of “redundancy” and lost energy. You may be aware of the fact that 12.1 ships with the brand new (although still a bit bleeding edge) SAX3. With this tool it will hopefully be straightforward to overcome the “impasse” caused by the IMO “out of the belly decision” to give up SAX2. So you may reconsider.
BTW: Windows is just another operating system, so if it works better for you…however, I recall at least for XP and NT several security updates that broke completely the system, and since closed source, even so much that recovery was not possible. Well, anyway, keep a good backup.

I still have 11.2 on my laptop because newer versions do not recognize the intel video chip.
This would surprise me (as long as this is not a “historical” notebook with a pentium 4). In this case, has to be honest and say: are you sure that on such old hardware you are able even to start Windows 7? I mean start and have a usable, not totally slowed down system. Because if you stick to XP then this is like saying: the system is too old, neither Win neither Linux are supporting it any more in newer versions.
I have an old Fujitsu Siemens Laptop still running with 11.1 evergreen. Whatever else would be stupid, it is nearly 8 years old now. It perfectly works and does it’s job. If I want really the latest and greatest I buy a new one (if possible without MS-tax on it, as I did with my Lenovo X201).

On 12/19/2011 02:26 AM, lnr729 wrote:
> I still have 11.2 on my laptop because newer versions do not recognize
> the intel video chip.

what is the number of the bug you logged against this problem for 11.3,
11.4 and 12.1??


DD
openSUSE®, the “German Engineered Automobiles” of operating systems!

I think one has to go slowly in case the OP is really as green as it seems. The join date (today) and just two messages leave these two interpretations. Either an OP that had never a problem and at the first sight of some difficulty stepped back to the original version without every posting for help in this forum is surely not experienced. The second possibility (always an option) is that the OP doesn’t exist. There have been a lot of contributions lately that I consider “on MS and NOSS software’s payroll”. Time will tell. For sure I would suggest to have always the “benefit of the doubt” and treat him as he would be a “green” one. So maybe you should then also explain to him where](https://bugzilla.novell.com/index.cgi) to file a bug, how to file a bug](http://en.opensuse.org/openSUSE:Submitting_bug_reports), etc. Maybe that is not in the interest of THAT OP (as he might well be fake) but the few real ones will find their positive outcome of this. And I will be even more happy to have you posting your justified critique. Regards.

On 12/19/2011 05:36 PM, stakanov wrote:
> The join date (today) and just two messages

maybe your go-easy-on-the-new-guy is right! and, certainly had i noticed
(which i didn’t) that join date i should have given the URL pointers you
have now given…(thank you for filling in that void)

but, had i noted the join date, my reading of “I still have 11.2 on my
laptop” tells me that even if never posting here, there is a history of
many many months of use…and having tried to move to a newer version
and failing there must be some bug filed or help request somewhere!!

of course, you may also be correct that it is just another paid FUD
spreader…we will see more and more of them as they continue to lose
market share…


DD http://tinyurl.com/DD-Caveat
openSUSE®, the “German Engineered Automobiles” of operating systems!

+1
I smell the fear already :wink: