On Sun, Apr 12, 2009 at 1:49 PM, jdd <jdd(a)dodin.org> wrote: As someone who had supported SuSE since 1999, I'm very disappointed with the current 11.1, and still use 11.0 on most of my stuff while I

I can't speak for Larry, but I can relate my experiences on several systems with Nvidia graphics

Nvidia is not the end, it's the means. The end is having a reliable SuSE desktop.

Cristian Rodríguez wrote: Hi Christian, Please understand, I'm not picking on SuSE here. I'm not giving up, but just waiting for 11.2. But look at it from my perspective: 11.0 works, 11.1 doesn't. I don't care if it's Novell or Nvidia: it doesn't work. So I'm going to take the path of least resistance and retreat to what's working. This will also giv KDE 4 some time to mature.

I found that when I bumped my nVidia binary drivers up to 180.x things improved in a BIG way with KDE4.. that and making sure I was using 4.2.2. (latest as of this email). GL screen savers are fine... full screen, not stuck up in the top left corner.... KDE desktop performance is fine... ie, I don't notice any slowdowns or performance issues at all anymore. I pull my nVivia drivers from here: http://www.nvnews.net/vbulletin/showthread.php?t=122606 and am currently using 180.44. I don't know what's available in the openSUSE repos, but if you have a nVidia card, and you're not on at least 180, you will have major issues with KDE4.

On Sun, Apr 12, 2009 at 02:20:00PM -0400, Larry Stotler wrote: Why? Ciao, Marcus

Ditto - I spent some time on the opensuse wiki and the KDE pages looking for info about moving PIM data from KDE3 to KDE4 - nada.

I'm working on extending kde4-migrate so it automatically pulls all the PIM data to KDE 4 for 11.2.

Problem is that majority of supporters do a lot to help those with KDE4, and spread a message about OBS.

Well majority, seems to stick with forums, Usenet, IRC even Facebook/Myspace/YouTube openSUSE communities seems to be comparable or bigger then ML. What I do read here, on couple of KDE ML, in forums.opensuse.org, around the web (google opensuse) is not near as tragic as you see. Users en mass go for extra repos, in any distro, that provide 4.2.2.

I gave you in another post link to small periferal feature that makes desktop look really cool, with neon glove below windows instead of dark shade. Is desktop slower, no. I have forgotten to upload settings that make old PCI nvidia FX5200 run with desktop effects as fast as without, but I'll do it right now.

On Sun April 12 2009 2:20:00 pm Larry Stotler wrote: > On Sun, Apr 12, 2009 at 1:49 PM, jdd <jdd(a)dodin.org> wrote: >

<snip> That is their favorite tactic, not addressing the underlying issues.

Sorry the previous administration and govermental mismanagemt at all levels has invaded your life. ... however, don't run from openSuSE, it is still worth saving, instead, change the audience you write to be more effective. What you say is basically right, the audience is wrong. Management needs to be 'educated' about how their profits can and will suffer if they continue to make poor decisions at the highest levels.

And for many, it still isn't ready and at the moment, a developer group has defacto control of an entire distribution by the way they are doing 'business'. Novell needs to regain control.

That is a valid reason to not upgrade, but for those that MUST upgrade to address some of the very issues you mentioned previously but will lose the current, functional version of their desktop environment. KDE3 is an essential part of the functionality of openSuSE for many reasons and whether or not you find KDE4 an acceptable replacement, the fact remains that for some, they are being forced to live with existing bugs in KDE or other packages in the openSuSE distro, OR to give up their current environment in order to address other issues such as drivers, kernel, etc. which are fixed in later distro releases. It is simply pre-mature to elimiate a functional option in deference to something that is not yet ready for prime time for many people.

I don't fault KDE devs for wanting to develop a replacement for KDE3, but I do fault openSuSE management for their head-long rush to the same cliff the rest of the lemmings are jumping off of.

... Remembering that openSuSE is a development tool for their SLE* product line, and bugs and bad decisions made there will increase their costs in the end product that they sell because it will require more time and money to fix the bugs that openSuSE 'WONTFIX' and will cost money by having to provide support to their paying customers fixing bugs that could/should have been fixed. SLE* lags openSuSE by one or several openSuSE releases, which if those base distros contain 'WONTFIX' bugs will remain to cost them in debugging and support issues later.

Dan Goodman a écrit : it doesn't need so long text to say that, and Kde is not Linux.

try puppylinux, for example, and with openSUSE Gnome or Xfce are good

and yes, in opensource world, devs do what they want... like it or leave it

jdd

Dan Goodman a écrit : did you notice in the example I give, there is openSUSE Gnome? that is mostly SLE, no? so no, Kde is not mandatory (even if I like and use mostly kde 3.5)

and if we want something done, it's better to be constructive (and short)

I notices there is a PIM problem. hope somebody noted.

On Tuesday, 2009-04-14 at 18:36 -0400, Dan Goodman wrote: it not? No, it is not. There is no recommended desktop. Having one would offend the rest of the people using another one.

short) If your mail is too long, we'll simply skip it entirely or partially. No time to read it all, sorry.

30 KB!

<shortening the last reply considerably, but adding a lot of my own replies> Richard wrote: I run Linux because (unlike Brand X) it doesn't emphasize glitz over functionality. I don't object to eye-candy provided it is optional, and doesn't impose performance or maintenance penalties. But when it does, I don't mind not having something zooty to show to the uninitiated...I just want a strong, fast, stable platform. And I did briefly try to engage one of the lead KDE4 designers about the need to get the basics working, as well as to make a "classic" mode in KDE4 to ease the transition. Even M$ gets that one. Instead I was told that KDE4 was introducing a radical new paradigm that was of such great value that it needed the new desktop stuff ASAP, and then the apps could be ported over to the new platform. And that keeping KDE3 functionality was a technical drain on their resources, unless they retrofitted it once they got all the new goodies working. (Really, how many users care *what* icon set or theme they are using, once they find one that isn't visually obnoxious?) Obviously, they must think that they are going to conquer the world with KDE4 -- that it is Linux' new "killer app". I began with KDE because I basically was in agreement with what Linus Torvalds said a few years back, that Gnome tended to dumb things down, thereby limiting Linux' utility. After my sad attempts at co-existing with KDE4 (both stable and factory, to try to fix the stable version), and after probing the mind-set of the developers, I ran across another comment of Linus', that he just wants a desktop that allows him to easily do whatever he wants to do, and doesn't get in the way. And for him right now (recently) it was Gnome. (Sorry, haven't the time to dig up the details, but it wasn't an April Fool's or Onion type page, it was one of the standard sources for info on Linux -- just can't remember which one. I read a lot online.) While Gnome has become less "dumbed down", I'm sure he must have more reason than that to switch back to Gnome. I work in a predominantly SLES shop, and am encouraged to run either SLED or openSUSE on my personal work machine. I chose openSUSE to be on the front edge of where SLES/SLED are going. I originally went with o/S 10.3 about a year ago, running KDE3.5. I skipped over o/S 11.0 and went to 11.1, initially trying KDE4. When I investigated and found out what the repair/enhancement strategy was -- this new paradigm MUST be running first AND users MUST learn to use this powerful new paradigm for their own good -- I attempted to switch back to KDE3.5. All I ended up with was a hodge-podge. Finally, out of frustration, I switched to Gnome (no, not because Linus did -- I found out about his switch while setting up Gnome), and it has been much more useful. And I haven't once woke up thinking "I could have been a part of a new desktop paradigm, but I have chosen to hang back and avoid change. Woe is me!" Others in my group have not simply defected from KDE4.5 after upgrading; instead they elected to go for simplicity of administration -- so they defected to Ubuntu. If I thought 3.5 would be available for as long as I wasn't satisfied with the state of 4.x, I would wait for KDE 4 to get to its "promised land", then switch. But this thread shows the futility of that. I thought KDE3.5 was better than Gnome, by a small amount, and that Gnome was much better than KDE4 (from a user perspective). If I believe the KDE design team leads, KDE4 is the greatest, because its fundamentals are more sound, and that when it is finished, it will blow both Gnome and KDE3 away. But I must live in today... What the KDE developers fail to realize is that not all of us who use Linux, KDE, or whatever, have time to become lab rats for the developers over an extended period of time. Even if I buy their argument that I must be willing to test the new KDE in order for it to improve, I still reject their assertion that I must choose to do so on their timetable, regardless of whether or not it interferes with my job or my other Linux interests. And Linux is all about choice, so why would I choose the pain, unless I were a true believer in the vision. But any vision that doesn't provide a decent migration path, or a comprehensible list of benefits, isn't a vision I am likely to buy into. And no, working Plasmoids just don't rock my world. And I wholeheartedly agree with David Rankin's post earlier, basically describing the fallacy and danger of "letting the tail wag the dog". If the KDE developers don't want to take a year to get apps working, then go back to building the desktop to conquer the world, as someone else suggested, then why not at least stop new work on the eye-candy, and devote more effort into responding to the needs of the KDE community, in both versions for now. And no, Mr. Asiago of the KDE team, although I had been a member of the KDE community, I do not need a new desktop paradigm, may not even need one, and I do not feel that, as a Linux or openSUSE community member, I am obligated to work to advance your vision of KDE, regardless of how right that might be someday. Especially since most of the benefits you enumerate have to do with issues that are at best of marginal interest to us as users. Had I sensed that KDE leadership was going to zealously allow KDE3.5 to be fully available to all KDE users for as long as it took for them to be comfortable with KDE4, I might have muddled through. But instead, I was told that KDE4 is so different that unless users are forced to try and learn the new paradigm, they might never cross over to the KDE4 promised land, so they felt that they had to guide users into KDE4. And elsewhere they suggest that Factory versions will provide fixes faster. But in a production shop, I elect to only run Factory fixes when and if I have a specific problem that I need to resolve -- not to enable me to learn a new paradigm while it is being debugged, or as part of my standard update strategy. I suggested supporting both in parallel -- sorry, the infrastructure is different, and we don't want to take time away from making the apps work in v4. But we are hampered in doing that as well, they claim, because they need to finish evolving their new paradigm in order to have a stable base in which to get apps working again. I have added this post because I feel it adds another user perspective, one not fully spelled out elsewhere -- the need of users who work professionally with SLE[SD] to be able to run an openSUSE release that does not interfere with their work, and that does not require extensive use of Factory builds to try to overcome what could have been, and should have been, pre-release testing in many cases. In the end, I feel sorry for just about everything about KDE -- because it *was* the best interface (except for a couple of features in OS/2), and because it could have evolved (albeit more slowly) into something above and beyond that, introducing a new visual and working paradigm -- but instead, KDE4 languishes in the end-user community because it is being force-fed, and because KDE3.5 has now become something that has to be "reverse-engineered" into new releases of openSUSE, if indeed even that can even be done effectively. I personally will not return to KDE, based on my experiences thus far, until such time as I see that they have (re-)built a satisfied user base with a version they intend to continue to fully support. And I will remain with it only if the new paradigm doesn't continue to get in my way after I have learned the basics, and only if there is sufficient perceived benefit to make it worth investing the time. Had they (the KDE team) taken a less aggressive and imperative approach to the introduction of KDE4 (and especially its claimed radically different new desktop paradigm, plus a reduced set of functionality in areas other than the desktop itself) -- had they made the transition more gradual, voluntary, and on the users' timetables for change -- I would probably gladly have worked through bug issues with the transition AND made a sincere effort to comprehend the new paradigm as well. But instead, I find myself in total agreement with much of what has been said here by other frustrated users. I have added this "rant" in part because it is the best forum for being heard about my experiences, and also because I am a member of what Novell should consider an important sub-community of the openSUSE community -- those who use it as a way to be at the leading edge of SUSE because they actively support Novell's commercial efforts on a daily basis as part of their work, but who don't work for Novell. For me, and for most of those I know in a similar situation, we cannot invest a lot of time in trying to help stabilize what is unstable, unless it is a part of our assigned work duties. And we cannot routinely run the Factory versions of things, as we need to be able to depend on our machines working today at least as well as they did yesterday. After investigation, "no KDE4" became an easy choice. And after trying to roll back, and then finding out that I was going to be caught on a weird backwards-compatibility bleeding edge, trying to run KDE3.5 out of Factory repositories, etc., I felt like I had no choice but to jettison it. Gnome just happened to be the least painful transition -- although if there had been no Gnome, I would have gone to xfce or something like it. It was not a pro-Gnome decision. And it was not based on an initial dislike of KDE. The KDE crew can go paradigming to their heart's content. But they definitely risk facing a reduced user base, at best only temporarily. And their "brave new world" may very well fail to attract users and/or recapture those who have defected, either because others don't like the new paradigm even after they learn it, or because a large group of users develop a backlash against being forced to upgrade. After all, isn't that one of the biggest complaints of users in M$-land, that they are forced to upgrade when they don't want to? Everywhere else in my experience, Linux distros have tended to allow the older version to co-exist with the new, until there has been enough time for not only the early-adopters, but those who want to see acceptance before jumping onboard, to move to the new on their own pace. KDE is breaking this model, and since it is the desktop of choice for SuSE Linux, from a new user perspective, it looks like Novell and SuSE are breaking this "no forced upgrade" implicit promise to the user community. (Other than reasonable and gradual new versions that do not force a major shift until stable new functionality has been introduced, of course.) Ubuntu would be easier for me to install and operate, but openSUSE represents a better overall experience, because of the community and the add-ons available, as well as because of its relationship to my work. But this only works for me because Ubuntu could only make it so much easier, as long as openSUSE didn't become too much of a hassle. But the KDE thing has the potential to turn the openSUSE overall experience into more of a hassle, from a lot of perspectives. And if KDE3 is gone from 11.2, and someone tries out 11.2 with KDE4 for the first time, then tries Ubuntu, the gap will be much wider. Even those who are technically proficient enough to handle it will be much more inclined to defect, if the experience begins to go downhill. (Had I not been working in a SUSE shop, I might very well have just started with a new distro, likely Ubuntu, instead of trying to retrofit Gnome to my 11.1 upgrade.) If it were up to me, and I speak only for myself, in spite of the work that I do, -- if it were up to me, I would bring out 11.2 with KDE3.5 as the default, require KDE group to provide a clean and tested upgrade path to KDE4, plus an easy path to roll-back to 3.5 (or Gnome) if not satisfied. Or even provide Gnome, plus both KDE's, and do as some other distros do -- "The choice of a window manager is too personal for us to make a recommendation", followed by several choices and a way to find out about other choices. And if the KDE team can't make the transition in both directions painless for 11.2, then put 4.X in as an option, not the default. Unfortunately, switching "on the fly" in SUSE Linux is not anywhere near being a "box check" or even zypper change to go cleanly between KDE3 and KDE4, much less between other desktops as well. If the KDE team ever gets to where it can win the support of the user community for the version it elects to support, I will consider changing on the next OS upgrade. But if it is the default choice, simply for historical reasons, that will only make me more suspicious of the choice, given the history...and I suspect I will not be alone. Nice sentiment, but even though "management" may need to be educated, I have enough other "problems of the world" to deal with. For me, I won't abandon openSUSE yet, but I clearly have abandoned KDE (as opposed to not trying or not initially liking it). And I have done so because of the management of the KDE team. And I have corresponded directly with them, shortly after my first 4.1 experiences, before arriving at this conclusion. And I tried adding repositories on their recommendation, tried to understand what could possibly be so great that I would endure a year or more of reduced functionality to obtain it, etc... I only abandoned it after trying to do "the right things" about getting the KDE team to see how this looked from the users' end, and was told that it was for the users' good, but that they wouldn't be smart enough to migrate to what was better for them, so they had to given only the "better" choice in order to get them to understand how great their lives will be under the new paradigm. I made a lot of the same points as in this thread with the team lead about how this implicitly assumes that the users are either too dumb to switch to another alternative, or are contradictorily, smart enough and altruistic enough to put up with the pain. Sorry, I don't want to wander in the desert, even if it isn't for forty years... And if they do, I may return to a stable and useful version of KDE. But if they do not, some will defect SUSE, others like myself (and Linus) will remain with our distros but abandon KDE. I agree that KDE (3) is a significant part of the SUSE experience, but if the developers cannot find a way to continue to maintain 3 while building 4, then perhaps either a fork would be in order (a new "old paradigm" team, with the current team free to innovate), or the KDE team needs to reevaluate its priorities from the community perspective, as opposed to their perspective of the user community. (However, since the current team feels that their paradigm is so radically different that it must be pushed out to the users to get them to use it, I doubt that this will happen.) Yes, openSUSE management and Novell management should continue to push for the new. But not at the expense of utility and functionality for the end user. Another example of the "make them take it and they will be better off" mentality, is Novell's (and Miguel de Ycaza's) desire to see the MONO project succeed, and somewhere along the line, the decision is made to make MONO-based Beagle the default search engine, and to have it on by default, even though it can be a negative on a small system, especially if the user doesn't need it. Instead of this approach, perhaps both the openSUSE community and Novell need to consider letting these "leading edge" projects exist, but not as defaults, and provide painless transition between using them and disabling them. As I said before, the tail (KDE design team) is wagging the dog (Novell, openSUSE community, and prospective new users). But if the tail turns out to be a problem on a dog breed, the tail is cropped off. You may not like that idea in the dog world, but it is a good paradigm for what will happen to this particular "dog wagger". Let us hope that the tail doesn't cause the breed of dog to fall out of favor in this case. Experienced users can make the split, even if it isn't the official position. But new users are more likely just to end up with a different "dog". Sorry for the length of this, but I know that if I left out a lot of this, the next replies would be: Did you try to communicate with the KDE team? Isn't this just one example, and isn't it just a matter of a bit of schedule slippage that will be long forgotten in a year? At least they are doing something, why do you think it could be done better? Etc. etc., etc.... So I have tried to address all these points, because I have first hand knowledge that it is more than a schedule slippage, it is an attempt to forcefully wean users off of 3.5, with a misplaced confidence that they will all end up happy on v4. "I have seen the future, and I think I'll draw another hand..." For me, that hand is a workable Gnome, since 4 is not a great KDE now, and 3 is being made more and more inaccessible as time goes by. I agree that KDE 3.5 definitely enhanced the Suse experience, but Brett Favre was a great quarterback in American football, but at the present time, and in his current condition, he might make a good coach, but I doubt anyone would willingly choose him as a starting QB based on his current ability. For some, he might also be rejected because of his recent past. And KDE looks to me to be a lot like Brett Favre's career...once the greatest, still better than many, etc., but not the one you'd choose today, because he may be smarter today, but he is not a better starting QB today. And that is about where the KDE team is at: they may have a smarter vision of the future, but they are not able to execute it in a way that won't cost them a lot of their backing. And even then, there is no guarantee that they will be able to deliver, or that whatever they deliver will look as great to the rest of the world as it does to them. I did not write this to try to start a bonfire, but rather because I want to see a better solution to the desktop issue in SUSE. And KDE, unfortunately, stands out in Linux-land as the single thing that most resembles the unpleasantries I, and others, have had to put up with in M$-land. Not because of how it used to work, but because of how it is today, and because I am asked to buy into, and even support, a better future taken on faith. And because I am told that I (as a Linux user) owe it to that project to help it become even greater...sorry, Linux may be a movement, but only for as long as it continues to be a more useful tool. There are a lot of things that might make the world a better place that I do not participate in, because I find that I have something else more important to me, to become involved in. And I am first unconvinced that KDE 4 represents a better paradigm, and secondly I am unconvinced that v.3 could not be supported while v.4 is being developed. (No, it isn't developed yet, it is a prototype.) Yes, it would take longer that way, but you may not have much of an active KDE community left, if they spend enough time in the status quo. Indeed, unless the KDE team has a very large community of users who buy into their vision and needs, keeping v.3 alive may be the only way for them to prevent a mass defection from KDE. And I'm not seeing or hearing that most of us are sold on KDE4, even as it will be, much less as it is. So does anyone want to make any bets about where KDE will be in two years? Very sad...but the world moves on, sometimes to the beat of a visionary different drummer, but more often, to the same or similar daily beats, while the visionary takes to their soapbox to castigate the world for failing to adopt their better way. Tell us again, please, Mr. Asiago and the KDE team, why anyone who is not a "true believer" would put up with this. Alas, many of us will not, and if we are proven wrong, all we will miss is a couple of years of painful debugging of a project that is not ours. (Oh, yeah, that and you will look down on us for not helping to develop your vision for the next couple of years...guess I'll have to live with that.) Consider this my eulogy for the KDE that could have been, but will not be -- a KDE that leads us gently into a better world, with minor discomforts perhaps, but without the radical, leave it all behind you and join us building Utopia, mentality. Still, I missed what KDE looked like it was, and was going to be, a while back. But then again, perhaps it is a victim of too much binding to the OS, instead of being a pluggable layer on top of it. But regardless, it has long since stopped "being fun". -- Dan Goodman Senior Systems Administrator <openSUSE 11.1 w/Gnome on a Lenovo T61> (Still trying to get all the vestiges of KDE out of my way.)

On Mon, Apr 13, 2009 at 17:53, Dan Goodman &lt;Dan.Goodman(a)coat.com&gt; wrote: [SNIP] His name is Aaron Seigo (aseigo). Just thought I'd help you out there. ne... A happy user of oS11 with KDE 4.2.2 and who does not update from OBS.

The oS devs have made their position clear,

On Mon April 13 2009 3:48:47 pm ne... wrote:

First, the KDE devs are making changes to the OBS that modify earlier versions of KDE with KDE4 code. Second, There is no one responsible for fixing changes made by 'virtaully anybody' that break existing code/libraries/packages. Third, many, even most of the users are not programmers and lack essential troubleshooting skills and Fourth, even for the Cummunity members that are or can program, virtually only the Devs have any real understanding of the code in the OBS that will be 'maintained' by Community and because of poor or no documentation in the code or elsewhere, User patches submitted to the OBS run the real risk of inducing subtle logic errors that often have far-reaching side effects. Lastly, no one really expects the Devs to 'maintain' KDE3.x, just to quit destoying it for those that need or want it, warts and all.

If the OBS copy of 3.5.x were left untainted by the development of v4.x, that might be somewhat viable, but what is in the OBS as 3.5 is 'tainted' with replacement code that has been modified for 4.x compatibility at the expense of compatibility with 3.x except in name only in some cases.   OBS is NOT a viable solution at this time.

I do, often, but not everyone has that ability and don't have the skills to be 'proactive' as you suggest.   They just need 'leave well enough alone' mentalities to prevail somewhat better.    No one is asking the Devs to stop developing, just not to do it at the expense of those that don't want to play that game.

As already indicated, the last 'untainted' version of KDE3.5.x is no longer in the OBS with key features/utilities like Konqueror for instance, having been replaced by the KDE4.x replacement of the same name.   There is nothing wrong with PORTING KDE4.x code to be compatible with 3.x, just keep it separate in case the Port is undesireable for some.

You have restated the problem very well.    It is only the solution is not good because it is already comprimised and lacks adequate responsibility and quality control  considerations. However, we are not far apart in our understanding of the problem.

On Sun, Apr 12, 2009 at 1:49 PM, jdd &lt;jdd(a)dodin.org&gt; wrote: Unfortunately, this has been tried before, and the KDE3 users were basically harrassed and ignored and told to just suck it up and deal

In &lt;200904140017.11309.ajohansson(a)suse.de&gt;de>, Anders Johansson wrote: Debian will provide security and RC-level bug fixes for KDE 3.5.9/3.5.10 that was shipped with Lenny (on the 14th of February this year) until (at least) Squeeze is released.

On Monday 13 April 2009 17:25:27 -0500 Boyd Stephen Smith Jr. wrote:

> Debian will provide security and RC-level bug fixes for KDE 3.5.9/3.5.10 > that was shipped with Lenny (on the 14th of February this year) until (at > least) Squeeze is released.

I was not aware of that.

I have no idea what "RC-level" means, but again, it

will be very interesting to see how they cope.

Not at all. It's just that the KDE3 users refused to answer simple questions like "what is it that you use in kde3 that isn't there in kde4". They insisted on giving non-answers like "everything", or "the devels should know already".

Oh, so I'm a 5 year old now?

I ask for KPersonalizer to be ported so I don't have turn waste hours trying to figure out how to turn off that glitz and I'm ignored.

I ask that we be able to remove unneeded crap like avahi and pulseaudio and I'm looked at like I'm crazy.

Why do we need all this crap as dependencies? What's so much better than alsa and is pulseaudio a stable replacement for alsa?

And why do I NEED a search tool? What's that new thing that's integrated into KDE4 - neopunk or something. Last time I checked, I removed all that garabage. Now I HAVE to have it?

Why don't you tell me what is in KDE4 that I NEED to have? Show me one feature that makes it compelling for ME to use it. Show me how it's rock solid and stable and doesn't need a new video card. Tell me I don't have to constantly update it to make it usable. Don't include any glitz - that's not something I care about. The ball's in your court now. Tell ME how much I NEED KDE4. I'm willing to listen.

[...] The issue is that I would need to use the build service to get 4.2.x, but I've recently seen that we are being told that the build service isn't supported, so if something breaks, I'm SOL. But, you're telling me that the current 4.2.x is fixed, so I guess I need the build service to get that. So, is the build service considered part of the system or not? [...]

On Tue, 14 Apr 2009 13:51:29 Larry Stotler wrote: Can we please get one thing straight? You *don't* need to use the build service to get KDE 4.2.2 on 11.0, 11.1 or even 10.3. If you enable the correct repositories in Yast and you can install it from there (and I suspect have many fewer problems with packaging, updates etc).

You can remove pulseaudio, except one small library, but avahi can not be removed, a lot is linked to it. It can be dissabled, though. If you go back some time in the list, there was a thread about the system being almost "destroyed" by attempting to remove those. Even yast went away, because of deps.

[...] All the same, I did not send my previous email without testing what I said. Believe me, in 11.1 I was able to remove avahi with absolutely no other dependencies removed (other than nss-mdns, as I mentioned in the other email)

Just to confirm. # zypper rm avahi Loading repository data... Reading installed packages... Resolving package dependencies... The following packages are going to be REMOVED: avahi nss-mdns After the operation, 498.0 K will be freed. Continue? [YES/no]:n Carlos?

You can't eat your cake and have it too! Banshee (or one of its extensions) obviously makes use of zeroconf (the daap plugin: iTunes sharing) and you want to remove its provider: avahi

Sure, multicast dns - or zeroconf in general - isn't something that makes things work. It's something that makes things work without configuration. The whole idea of zeroconf is that "normal people" can just plug in their machines, and have more or less everything work, including networking, automatic discovery of file shares etc.

For example, if I block or disable avahi, my PS3 system doesn't find my file share where I store movies for it

But I agree that even banshee/mono shouldn't have a hard dependency on it

Sure, multicast dns - or zeroconf in general - isn't something that makes things work. It's something that makes things work without configuration. The whole idea of zeroconf is that "normal people" can just plug in their machines, and have more or less everything work, including networking, automatic discovery of file shares etc.

For example, if I block or disable avahi, my PS3 system doesn't find my file share where I store movies for it

But I agree that even banshee/mono shouldn't have a hard dependency on it

Perhaps not, but Linux and openSUSE are not supposed to be just for the technically proficient. I think they need to be in a modern Linux distro, its just that they have to be pretty bug free (only narrow HW issues, etc.) to be ready for prime time.

In &lt;49E8EC7F.1060104(a)coat.com&gt;om>, Dan Goodman wrote: I think that, for users that don't want them they should also be removable without breaking software that doesn't strictly require it. Because of that, an RPM dependency is not appropriate. A less strict relationship like APT's "Recommends" or "Suggests" is more appropriate.

All the same, I did not send my previous email without testing what I said. Believe me, in 11.1 I was able to remove avahi with absolutely no other dependencies removed (other than nss-mdns, as I mentioned in the other email)

It did. You even quoted it above. There was a complaint about illogical dependencies on avahi, and it was fixed in 11.1

Surely you don't expect a complete remastering of 11.0

On Wed, Apr 15, 2009 at 2:13 PM, Anders Johansson &lt;ajohansson(a)suse.de&gt; wrote: Must have missed that. Since 11.1 is not an option for me, I guess I'm stuck with it. Honestly Anders, I've compained about stupid dependencies in SuSE for as long as I can remember. My question is who decides the dependencies? You guys who create the packages for the distro, or the author of the program? Nothing but bug fixes.

Carlos E. R. wrote: Larry, I don't remember whether you were around for the discussion last year concerning "stupid" dependencies, so I'll repeat my argument here, for your and others' benefit. Not being a linux programmer (but having written hundreds of other programs of several kinds), I can't speak for them; however: It's standard practice when developing a large program to use resources from other programs and their dedicated libraries. Two good reasons for this are that it reduces (often tremendously) the labor involved in writing a large program, and that it makes development (often much) less risky. So avahi has a library that contains several functions that would help me make my own program easier and better. I'm going to use it. That means that avahi is now a dependency of my program. Avahi (I don't know what avahi is, really, and for this argument, I don't care) is a large program with a single large binary library that I can't easily get into, even if I wanted to. So (for example) the 20K of code that I wrote, with the 100K of avahi code I linked to, just became a 4M program, much of the code likely unused. But I feel it's worth it. I might never have finished it if I'd had to do all that stuff myself. Now I have the benefit of a successfully running program without too much work, and you have the benefit of a solid program you can use for your purposes much sooner than you would have otherwise (if ever). And it's much more robust because I used proven techniques and code from avahi.

On Thu, 2009-04-16 at 08:22 -0400, Ken Schneider - openSUSE wrote: !!!!!!!!!!!! As a programmer, what control could I possibly have on which version of a library is installed at runtime on some other system? Not all libraries identify their version to the app, so I could not refuse to run if the point release does not match some magic list of good ones. At best, the version of the library is controlled by the name of the .so file. The program linker (not the code programmer - maybe the configure/Makefile author) controls that. But that still requires this magic list of good versions that can be consulted. Or are you meaning something else?

On Thu, 2009-04-16 at 09:40 -0400, Greg Freemyer wrote: All these things are packaging issues, not programmer issues. A least I am defining the programmer as the guy who writes the code that will be compiled. Not the guy who makes the packaging. Of course, that is programming as well. But I think it is better to call them packagers, or package designers, or anything other than programmers simply because the word programmer has prior meaning.

On Sunday 12 April 2009, jdd wrote: A Fully working Quanta for one not Quanta for KDE 3.5.x that is half dead because things are not there cus it needs KDE 3.5.x that is the one that directly affects me .

... Or in a southern drool (What We Have Here Is A Failure To Communicate) ...

Pete.

I just shake my head at the derisory decision making taking place and the newfound *urgent* need to dump kde3. We were initially told that KDE3 would be part of 11.2, now this? (Note the *loud absence* within Novell of anyone willing to take "credit" for making this important decision)

You are correct.  The original discussion caused KDE3 to *be* included in 11.1 as a *last* issue rather than 10.3 being the last issue.  ll.2 was scheduled to be all KDE4, and other window managers.

KDE3 should not be dropped until the KDE4 devs can show we are NOT sacrificing functionality. I can't wait for KDE4 to be finished, it should be great. But to consider KDE4 a replacement for KDE3 at present is intellectually dishonest.

Quanta is Case-In-Point, a MAJOR app and arguably the best at what it does. I was using it not more than 30 seconds before checking email and finding this very depressing thread. You can't even upload the files you are working on to your server in the KDE4 version with fish. In KDE3 it is as simple as hitting F8.

Presently, for quanta and for other major areas of functionality, KDE4 is a bust, it's not there yet, and there is NO way Novell, openSuSE, or God for that matter, can expect KDE4 to mature sufficiently between now and the 11.2 release to be a workable replacement.

I just shake my head at the derisory decision making taking place and the newfound *urgent* need to dump kde3. We were initially told that KDE3 would be part of 11.2, now this? (Note the *loud absence* within Novell of anyone willing to take "credit" for making this important decision)

I mean just keep telling people kde4 will be ready by 11.2 enough times, and you can calculate 90% or so will be stupid enough to believe it. How many time did it take hearing "Sadam has weapons of mass destruction, wmds, and we know where they are", before you bought in?

Hopefully, Novell is privy to some hidden information (intelligence if you will) that guarantees the KDE4 glamor toy we now have will be ready by 11.2 -- I doubt it. This dice-roll decision making within Novell concerning functionality for what was once the hands down, most functional distribution with the greatest attention to detail... is disturbing to say the least -- or downright alarming to be honest and candid about it.

The management question still lingers. Would novell ever put out a release of its openSuSE distribution that cannot do what its last prior release did with the addendum of "uhh, we know its broken and doesn't quite have the functionality the last version did, but we take such pride in saying we are the best, we just decided to release it anyway...without a working KDE3.5 this time..." ???WTF??? That is the type of corporate thinking and corporate decision making that brought us credit default swaps and we see how good that turned out.

Hopefully 12.0 will be a solid release with a working version of Quanta in a mature KDE4. Sad, I used to alway look forward to the new capabilities in the upcoming release. Now, I get to look forward to loss of capabilities in the upcoming release?? My, "the times, they are a changing." (catchy phrase, probably make a good song)

OK, now just chime in with how we are "just in the same type of transition we were in with the KDE2-KDE3 transition" and "not to worry", but please be kind enough to provide citation to the references you use to show that lack of comparative functionality was still being debated *three* releases down the KDE3 road. (still just shaking my head in utter disbelief, with a dry chuckle of solemnity...) Beware the ives of Mandrake...

Cheer up, " (and therefore Quanta) is still in Factory.

On Thursday 16 April 2009 09:45:10 am peter nikolic wrote: Pete, you, and everyone else (including me), want smooth transition. What do you think how that is going to happen? Wait until all applications are ported. That will happen when Mr. Godot comes. Ranting against eye candy? Please, read this: http://en.opensuse.org/KDE/KDE4 " Performance Issues KDE 4 is the first Free desktop environment to make use of advanced features such as compositing, alpha blending and scalable graphics at the core of the desktop, as opposed to only using a compositing window manager ..." It is not eye candy, it is just the way to use computer efficiently, not as i386 and xyVGA graphic, with higher clock rate. If you are about efficiency and better use of gear that you paid for then you will greet that, instead of repeating eye candy, eye candy until repeat key breaks. You say that you need Quanta for web development. This days one can't go by without using eye candy of some sort, static, or dynamic, so why would web developer put endles rants against it? One should appreciate to have platform that can display his work without delays. Let me know if I'm wrong? -- Regards, Rajko

... I don't think that people after so much involvement in technology think on the eye candies first.

> KDE 4 is the first Free desktop environment to make use of advanced > features such as compositing, alpha blending and scalable graphics at the > core of the desktop, as opposed to only using a compositing window > manager ..." > > It is not eye candy, it is just the way to use computer efficiently, not > as i386 and xyVGA graphic, with higher clock rate.

On Thu, Apr 16, 2009 at 11:43 AM, Rajko M. &lt;rmatov101(a)charter.net&gt; wrote: The compositing manager from my reading is openGL based. Since it's built in to KDE4, I can see why my systems were so sluggish since I have no interest in wasting money on a higher end card.

Actually, it's like Vista's Aero manager, that basically requires a good 3d card to make the desktop work. That's why a lot of people turn off the "Aero Experience" so they can actually use their computer or they replace it with XP.

To use KDE4 effectively, I would be forced to upgrade my hardware, when it works just fine under KDE3. Correct me if I'm wrong, but it seems that the KDE devs are abandoning older hardware with their shiny new toy.

On Fri, Apr 17, 2009 at 6:03 PM, Boyd Stephen Smith Jr. &lt;bss(a)iguanasuicide.net&gt; wrote: So how do you turn it off?

KDE4 has shown zero benefit over KDE3 so far. That may change, and hopefully will.

With the result that you have three options: 1) go with beta software, experience as it gets better (it will) and accept to work the way the devs want you to. 2) fight to continue using KDE 3 on distributions that mix the two and slowly see KDE 3 die away. 3) As you anyway will have to change the way you work, look at other possiblities As far as I am concerned, 1) is not satisfying, 2) is depressing

In &lt;9bb996600904171438i6459ddf9l66a6a8e66f9ba5ad(a)mail.gmail.com&gt;om>, Larry Stotler wrote: KDE 4 does not require compositing. If you turn it off, it still looks slicker than KDE 3, but it won't ask other systems to do things they are bad at. So, turn off KDE 4's desktop effects. They are not a requirement. You are wrong. KDE 4 should preform *better* on old hardware if the eye candy is turned off and it should *still* look better than KDE 3. If that's not your experience, and you are willing to work with the developers, please join them on IRC. [Ideally, KDE 4 would be smart enough to choose the appropriate level of "eye candy" based on your hardware and then let you tweak it, but I don't think that's ready yet.]

On Sat, 18 Apr 2009 18:07:57 peter nikolic wrote: 2 things: 1. The application developers are not necessarily the same people as the desktop developers (although for some apps they are...) 2. You have to have the API's and other software frameworks in place and stable in order to build the apps. It is a bit like building an office block - you have to put up the framework before installing the wall panelling, windows etc. (no pun intended). The fact that the software interfaces and infrastructure facilitate "eye candy" is really just a side-effect of the chosen design and build tools.

A Fully working Quanta for one not Quanta for KDE 3.5.x that is half dead because things are not there cus it needs KDE 3.5.x that is the one that directly affects me .

All the more reason to hold off dropping KDE3.x from 11.2

Amarok 2 is an absolute failure in plugins so far, last time I checked there wasn't even an eq.

On Tuesday 14 April 2009 04:21:56 am G T Smith wrote: That was one of questions that I can't find answer. If you are unhappy, there is quite a few desktops around.

I use ICEwm for old machine that can't handle KDE3 stripped down to bare minimum. It is lesser effort to learn how to customize it, then to write endless mails around asking someone to change KDE to fit very old hardware (1998). Not that many in number of people, but for sure in number of posts, which combined with ureasonably high effort needed for that, compared to research of alternatives, wakes suspicion that there might be other problems they don't want to publish.

Having paying customers, where selling point was "it runs on older hardware" comes in mind. Now when KDE3 will be left in a dust, and some hardware upgrades are advisable, it will be problem to explain why.

The future of KDE vs. GNOME is probably on KDE side. In few months they will have nice maintenable desktop environment, while GNOME at that time will start thinking how to do what KDE did now.

On Wednesday 15 April 2009 21:36:57 Klistvud wrote: Fixed in KDE trunk - each desktop can have its own settings now, including wallpaper and applets. Will