Hi everyone,
I've been an intermittent user of SuSE and openSUSE for well over a decade, starting back in the 6.x days if I remember correctly. I've always admired the project for its professionalism and broad scope. I recently switched back to openSUSE thanks to the rolling Factory (now Tumbleweed) announcement, and I also have 13.2 installed on other machines. I'm really enjoying the experience, and this time, I'd like to stay.
I really want to see openSUSE succeed, but unfortunately I tend to see it dismissed by potential new users as "a nice distro, but..." followed by certain annoyances or "paper-cut" issues. This week's openSUSE 13.2 review on Distrowatch is the most recent example of such comments. I feel that openSUSE is fundamentally extremely well designed and executed, but I have also noticed the same recurring "paper-cut" issues year after year, release after release. So I'm wondering if anything could be done about them, or if the developers have any interest in fixing them.
Without going into detail, here are a few general areas of concern that I often hear and/or have personally noticed throughout many openSUSE releases:
1. Ugly font rendering 2. Automatic installation of extraneous/unrequested packages, especially after initial installation. 3. Crippled packages meant to prevent compatibility with proprietary multimedia formats. 4. Difficult to install multimedia and/or proprietary formats and drivers. 5. Breakage of YaST modules.
All of the above issues do have workarounds, and think it would be best for this thread to not get into the technical details of these aforementioned issues. But in general terms, these issues come up time after time in reviews and forum posts about openSUSE. So my question is: Would the openSUSE project be interested in working to resolve any or all of these issues? And if so, where would be the best place for me to bring up these issues and work with developers to make improvements? Some of the issues are not related to any one specific package, but are rather of a more systemic nature. In some cases legal/patent issues are probably involved. (Again, let's please avoid legal discussions in this thread.) Although I am not a coder or developer *at all*, I do have an eye for detail and polish, and I have a good idea of what typical users expect out of a Linux distro on the desktop. I'd like to help to identify and test solutions to these papercuts if any developers are willing to look into some of these long-standing issues with a fresh eye.
Any comments? Thanks in advance for your time and consideration.
The easiest that can be done here is to propose possible solutions and also involve our team at Factory that could take a closer look at the technical aspects of this problem. For fonts, I have proposed in the past switching to a different font family that could be more suited for what we use in modern screens. I am not sure that we were able to implement it. It mostly remained as an idea.
It might be possible also to have certain screen specific configurations (font only discussion) but that is something that we need to discuss with our factory team.
On Mon, Nov 17, 2014 at 12:48 PM, S. sb56637@gmail.com wrote:
Hi everyone,
I've been an intermittent user of SuSE and openSUSE for well over a decade, starting back in the 6.x days if I remember correctly. I've always admired the project for its professionalism and broad scope. I recently switched back to openSUSE thanks to the rolling Factory (now Tumbleweed) announcement, and I also have 13.2 installed on other machines. I'm really enjoying the experience, and this time, I'd like to stay.
I really want to see openSUSE succeed, but unfortunately I tend to see it dismissed by potential new users as "a nice distro, but..." followed by certain annoyances or "paper-cut" issues. This week's openSUSE 13.2 review on Distrowatch is the most recent example of such comments. I feel that openSUSE is fundamentally extremely well designed and executed, but I have also noticed the same recurring "paper-cut" issues year after year, release after release. So I'm wondering if anything could be done about them, or if the developers have any interest in fixing them.
Without going into detail, here are a few general areas of concern that I often hear and/or have personally noticed throughout many openSUSE releases:
- Ugly font rendering
- Automatic installation of extraneous/unrequested packages, especially
after initial installation. 3. Crippled packages meant to prevent compatibility with proprietary multimedia formats. 4. Difficult to install multimedia and/or proprietary formats and drivers. 5. Breakage of YaST modules.
All of the above issues do have workarounds, and think it would be best for this thread to not get into the technical details of these aforementioned issues. But in general terms, these issues come up time after time in reviews and forum posts about openSUSE. So my question is: Would the openSUSE project be interested in working to resolve any or all of these issues? And if so, where would be the best place for me to bring up these issues and work with developers to make improvements? Some of the issues are not related to any one specific package, but are rather of a more systemic nature. In some cases legal/patent issues are probably involved. (Again, let's please avoid legal discussions in this thread.) Although I am not a coder or developer *at all*, I do have an eye for detail and polish, and I have a good idea of what typical users expect out of a Linux distro on the desktop. I'd like to help to identify and test solutions to these papercuts if any developers are willing to look into some of these long-standing issues with a fresh eye.
Any comments? Thanks in advance for your time and consideration.
To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org
Thanks for the reply, Andy. So create threads about specific issues at the opensuse-factory list?
Thanks for the reply, MH. I'm very well aware of the legal issues related to proprietary formats. But I don't think this list is the right place to discuss specific legal/patent issues? Or is it?
In general, most big, international Linux distros also don't pay any software licensing fees and therefore they also don't distribute proprietary software. But they do make it considerably easier than openSUSE does to install such software when the user needs/wants it. In some cases, considerably better defaults could be configured using the available open source software that could help some users to not recur to a proprietary/licensed alternative.
Here's the Distrowatch review: http://distrowatch.com/weekly.php?issue=20141117#feature
Le 17/11/2014 21:15, S. a écrit :
In general, most big, international Linux distros also don't pay any software licensing fees and therefore they also don't distribute proprietary software. But they do make it considerably easier than openSUSE does to install such software when the user needs/wants it. In some cases, considerably better defaults could be configured using the available open source software that could help some users to not recur to a proprietary/licensed alternative.
could you describe such config?
apart Canonical (ubuntu) that build itself in Man Island?
thanks jdd
Thanks to everyone for their replies!
Roger Luedecke wrote:
the font rendering issue is being worked on actually and should be fixed in an update soon.
Great to hear this! What does the fix involve?
I'm aware that the subpixel rendering used by the Infinality project (arguable the best font rendering available for Linux) is patent-encumbered, but in my experience Arch has fonts that are *almost* as good as Infinality, and I'm pretty sure they don't use anything patent-encumbered. In my experience, the DejaVu Sans fonts with "slight" hinting and RGBA antialiasing look pretty nice, even with the default openSUSE fontconfig. So that's an example where considerable improvement could be made in the default openSUSE settings, even within the confines of purely open source solutions.
Roger Luedecke wrote:
As for installing proprietary drivers, I'm not sure how much easier we could really make it.
Well, I do think that the Ubuntu Restricted Drivers tool is pretty fantastic. The reason is that most newish Linux users aren't really aware of the existence of libre/proprietary drivers, they just want their hardware to work. Many users aren't even all that sure about what sort of hardware they have, and it gets even more complicated with dual-graphics Intel/Nvidia Prime systems. So it requires more than a little bit of manual investigation to get certain hardware pieces working. The users has to find the hardware device in the YaST hardware tool, and then has to Google around to figure out what the names are of the drivers that correspond to the hardware, then they have to find the openSUSE package name as well as the repo that contains it. Sometimes there are libre drivers that work to a certain extent, but some users require the more complete functionality of a proprietary driver. Yes, it's not *terribly* difficult, and us seasoned Linux users can do it with our eyes closed, but I think that a lot of openSUSE users aren't interested in learning about their system-- they just want it to work correctly. A tool like the Ubuntu Restricted Drivers does all the hard work for the user, it tells them what pieces of hardware have an available proprietary driver, which versions are available, and a one-click button to enable it if desired.
Roger Luedecke wrote:
I don't know about our packages being crippled to complicate using codecs etc
The most concrete example I can remember off-hand is K3B. To rip MP3s from an audio CD, it required LAME. But the openSUSE version of K3B was not compiled with the flag that made it look for the LAME libs, and therefore it couldn't use LAME even if it was installed.
Perhaps a more modern example are the gstreamer -bad and -ugly packages. They are offered in the official openSUSE repos, but they don't seem to actually contain any proprietary codec support, because even after installing them I couldn't play MP3 or MP4 files.
I'm sure there's more examples of this sort of crippling of openSUSE packages, but these are some easy examples. Of course, us seasoned openSUSE users know that it's just a matter of adding Packman, setting its priority to a lower number to increase the priority (rather counter-intuitive for beginners) and switching installed packages to the Packman versions, then adding the gstreamer -bad and -ugly -addon packages. But that process is long and convoluted compared to other distros.
I know that openSUSE can't offer proprietary codecs out of the box with the ISOs, but would it be possible to add a YaST module called "Proprietary Software Support" that offers a "Install proprietary codecs from Packman" checkbox that runs a simple script to do the whole process I described in the previous paragraph?
Roger Luedecke wrote:
We need an integrated 'command-center' for the project so that communication will be more transparent and issues can be tracked in order to avoid regressions.
I see. Meanwhile until that gets implemented, would Bugzilla reports of these individual issues I mentioned be sufficient?
jdd wrote:
so, the good way now should be to enable packman, give it some lower number as priorité, go to yast software, *repositories view* and call for software from there. A bit complicated.
My thoughts exactly. I think that if openSUSE can offer a wiki page to do this whole process manually, it should be possible to add a YaST module and script to do the same thing, still pulling from Packman and thus avoiding hosting proprietary software on openSUSE's repos.
Lars Müller wrote:
Has one of you thought about filing bug reports for the individual issues and create one meta bug which references all these reports?
I can do that, I just wanted to make sure it wouldn't be too vague to report some of these issues that don't relate to a specific package.
Regarding my #2 issue "Automatic installation of extraneous/unrequested packages, especially after initial installation":
I think that there should be a prompt or something when using YaST Software for the first time, asking users if they want the resolver to #1) install recommends for already installed packages, and/or #2 install recommends for future installed packages. Additionally, I think that there should be a YaST Software option to enable/disable installation of recommends for future installed packages-- currently the option is only available for previously installed packages' recommends. Of course it is possible to adjust the /etc/zypp/zypp.conf option to not install recommends, which is obeyed by YaST, but I think a simple GUI checkbox would be better.
Also, I think that adding a YaST module with checkboxes for installing common stuff like Flash and openJDK would be better than using the update mechanism with packages like "pullin-fluendo-mp3" and "pullin-flash-player" to arbitrarily add this proprietary software, whether the user wants it or not.
Sorry for the long email. I'd be very interested to read your further thoughts. Thanks for your time!
On Tue, 2014-11-18 at 18:43 -0500, S. wrote:
Thanks to everyone for their replies!
Roger Luedecke wrote:
the font rendering issue is being worked on actually and should be fixed in an update soon.
Great to hear this! What does the fix involve?
I'm aware that the subpixel rendering used by the Infinality project (arguable the best font rendering available for Linux) is patent-encumbered, but in my experience Arch has fonts that are *almost* as good as Infinality, and I'm pretty sure they don't use anything patent-encumbered. In my experience, the DejaVu Sans fonts with "slight" hinting and RGBA antialiasing look pretty nice, even with the default openSUSE fontconfig. So that's an example where considerable improvement could be made in the default openSUSE settings, even within the confines of purely open source solutions.
As I understand it, they are actually merging the Infinality stuff into the libraries. Either they are not including encumbered bits, or they aren't as encumbered as we assumed.
Roger Luedecke wrote:
As for installing proprietary drivers, I'm not sure how much easier we could really make it.
Well, I do think that the Ubuntu Restricted Drivers tool is pretty fantastic. The reason is that most newish Linux users aren't really aware of the existence of libre/proprietary drivers, they just want their hardware to work. Many users aren't even all that sure about what sort of hardware they have, and it gets even more complicated with dual-graphics Intel/Nvidia Prime systems. So it requires more than a little bit of manual investigation to get certain hardware pieces working. The users has to find the hardware device in the YaST hardware tool, and then has to Google around to figure out what the names are of the drivers that correspond to the hardware, then they have to find the openSUSE package name as well as the repo that contains it. Sometimes there are libre drivers that work to a certain extent, but some users require the more complete functionality of a proprietary driver. Yes, it's not *terribly* difficult, and us seasoned Linux users can do it with our eyes closed, but I think that a lot of openSUSE users aren't interested in learning about their system-- they just want it to work correctly. A tool like the Ubuntu Restricted Drivers does all the hard work for the user, it tells them what pieces of hardware have an available proprietary driver, which versions are available, and a one-click button to enable it if desired.
I'm aware of this tool, but I'm not sure that we can port it. Might not be a bad idea, though in a way I would almost rather have it be less obvious. I get exhausted very quickly by Ubuntu users and their tendency towards petulant demands and insults.
Roger Luedecke wrote:
I don't know about our packages being crippled to complicate using codecs etc
The most concrete example I can remember off-hand is K3B. To rip MP3s from an audio CD, it required LAME. But the openSUSE version of K3B was not compiled with the flag that made it look for the LAME libs, and therefore it couldn't use LAME even if it was installed.
Perhaps a more modern example are the gstreamer -bad and -ugly packages. They are offered in the official openSUSE repos, but they don't seem to actually contain any proprietary codec support, because even after installing them I couldn't play MP3 or MP4 files.
I'm sure there's more examples of this sort of crippling of openSUSE packages, but these are some easy examples. Of course, us seasoned openSUSE users know that it's just a matter of adding Packman, setting its priority to a lower number to increase the priority (rather counter-intuitive for beginners) and switching installed packages to the Packman versions, then adding the gstreamer -bad and -ugly -addon packages. But that process is long and convoluted compared to other distros.
I know that openSUSE can't offer proprietary codecs out of the box with the ISOs, but would it be possible to add a YaST module called "Proprietary Software Support" that offers a "Install proprietary codecs from Packman" checkbox that runs a simple script to do the whole process I described in the previous paragraph?
Ah, now that would explain some things. I wonder if that is why Totem segfaults when I try to play an .avi. That seems like a lousy policy to disable their ability to use proprietary codecs if they are available. I think this would be a case for bug reports. I'd bet it is kept that way since it was done once and simply hasn't been changed yet.
Roger Luedecke wrote:
We need an integrated 'command-center' for the project so that communication will be more transparent and issues can be tracked in order to avoid regressions.
I see. Meanwhile until that gets implemented, would Bugzilla reports of these individual issues I mentioned be sufficient?
It'd be the most effective mechanism we currently have.
jdd wrote:
so, the good way now should be to enable packman, give it some lower number as priorité, go to yast software, *repositories view* and call for software from there. A bit complicated.
My thoughts exactly. I think that if openSUSE can offer a wiki page to do this whole process manually, it should be possible to add a YaST module and script to do the same thing, still pulling from Packman and thus avoiding hosting proprietary software on openSUSE's repos.
Our wiki is an unsearchable disaster zone. Frankly I don't really understand what structure it is supposed to have. I believe there are articles on how to enable all the proprietary multimedia support. But articles don't matter much if new users don't know they exist. I'd frankly want each desktop to have a 'greeter' that emphasizes such caveats that new users would have.
Lars Müller wrote:
Has one of you thought about filing bug reports for the individual issues and create one meta bug which references all these reports?
I can do that, I just wanted to make sure it wouldn't be too vague to report some of these issues that don't relate to a specific package.
In response to Lars; yes, and I do file reports. After a certain point I do get fatigued with filing reports, and that tends to be by the time I start fussing with fun things like multimedia. This is especially the case since I tend to forget how to get to relevant logs and things like that once an issue has been resolved since I'm not exactly the most technical user.
Regarding my #2 issue "Automatic installation of extraneous/unrequested packages, especially after initial installation":
I think that there should be a prompt or something when using YaST Software for the first time, asking users if they want the resolver to #1) install recommends for already installed packages, and/or #2 install recommends for future installed packages. Additionally, I think that there should be a YaST Software option to enable/disable installation of recommends for future installed packages-- currently the option is only available for previously installed packages' recommends. Of course it is possible to adjust the /etc/zypp/zypp.conf option to not install recommends, which is obeyed by YaST, but I think a simple GUI checkbox would be better.
I think we simply need saner 'recommends'. I once installed some small thing from KDE, and the next PK update pulled in the ENTIRE KDE ENVIRONMENT. I was very much not happy about that. Most recently I installed the broadcom-wl driver and it though xboards was a dependency. Xboards is a chess engine...
Also, I think that adding a YaST module with checkboxes for installing common stuff like Flash and openJDK would be better than using the update mechanism with packages like "pullin-fluendo-mp3" and "pullin-flash-player" to arbitrarily add this proprietary software, whether the user wants it or not.
I think maybe doing that in the installer is feasible, but as a dedicated YaST module seems to me to be outside of the scope of YaST.
Sorry for the long email. I'd be very interested to read your further thoughts. Thanks for your time!
Roger Luedecke wrote:
I'd bet it is kept that way since it was done once and simply hasn't been changed yet.
I think you hit the nail on the head. Well expressed. That's basically the point of this thread, to see if openSUSE devs could maybe take a look at some of these issues with fresh eyes. I was basically trying to gauge interest from the current group of devs in actually changing/improving the standard openSUSE defaults, and if there is interest I'd be glad to file detailed bug reports to get the ball rolling.
Stanislav Baiduzhyi wrote:
If I understand correctly, RGBA antialiasing *is* patented technology we are taking about. Hinting patents already expired, so grayscale+hinting can be used freely everywhere.
I had filed a prior bug report about the font issue: https://bugzilla.suse.com/show_bug.cgi?id=904238 They gave this link about the possible patent issues: http://david.freetype.org/cleartype-patents.html I know absolutely nothing about the underlying font rendering technology, and even less about patent law, but my understanding was that RGBA antialiasing is not the same as sub-pixel rendering, and that only the sub-pixel rendering is patent-encumbered. I very well could be wrong, though. I do know that on any distro, including openSUSE, for years it has been possible to enable RGBA antialiasing in the Gnome or KDE font settings, and it does improve font rendering considerably. But on openSUSE that configuration still doesn't look as good as it does out of the box on Arch or Ubuntu.
I noted in the above bug report that using *just* the Infinality fontconfig (not the libraries that actually implement the patent-encumbered process) with the standard openSUSE font rendering libraries does considerably improve things on openSUSE. As mentioned earlier in this thread, maybe something similar to this is in the process of being implemented now for openSUSE.
Can we expect an update for openSUSE 13.2 with the mentioned font rendering improvements? Or will it only get pushed out to Tumbleweed and the next fixed released after 13.2?
On Wed, 2014-11-19 at 07:41 -0500, S. wrote:
Stanislav Baiduzhyi wrote:
If I understand correctly, RGBA antialiasing *is* patented technology we are taking about. Hinting patents already expired, so grayscale+hinting can be used freely everywhere.
I had filed a prior bug report about the font issue: https://bugzilla.suse.com/show_bug.cgi?id=904238 They gave this link about the possible patent issues: http://david.freetype.org/cleartype-patents.html I know absolutely nothing about the underlying font rendering technology, and even less about patent law, but my understanding was that RGBA antialiasing is not the same as sub-pixel rendering, and that only the sub-pixel rendering is patent-encumbered. I very well could be wrong, though. I do know that on any distro, including openSUSE, for years it has been possible to enable RGBA antialiasing in the Gnome or KDE font settings, and it does improve font rendering considerably. But on openSUSE that configuration still doesn't look as good as it does out of the box on Arch or Ubuntu.
I noted in the above bug report that using *just* the Infinality fontconfig (not the libraries that actually implement the patent-encumbered process) with the standard openSUSE font rendering libraries does considerably improve things on openSUSE. As mentioned earlier in this thread, maybe something similar to this is in the process of being implemented now for openSUSE.
Can we expect an update for openSUSE 13.2 with the mentioned font rendering improvements? Or will it only get pushed out to Tumbleweed and the next fixed released after 13.2?
As I mentioned earlier, the folks in Factory are working on patching stuff with the Infinality tech.
Regarding openSUSE paper-cut issues, another 13.2 review appeared in Spanish on openSUSE Planet: http://www.lasombradelhelicoptero.com/2014/11/opensuse-132-cosas-que-mejorar...
The reviewer starts by saying that he likes openSUSE and doesn't want to bash it, in fact he had declared 13.1 to be the best Linux distro he had reviewed to date. But he had quite a few issues with 13.2.
Regarding installation of extraneous packages, notice what happened when he installed Spanish language packages: http://3.bp.blogspot.com/-YxAPbyJJ7Wo/VGtnX4b23wI/AAAAAAAAFTw/x0XQpKsE-Xc/s1... In his screenshot, I can see the additional dependencies, including Sane, Skanlite, Synaptiks, tcpdump, telnet, unrar, and a whole plethora of mostly unrelated packages. He says that step pulled in a sizable 500MB of packages, when all he wanted was some small languages packs. The glut of packages that are pulled in after the initial installation continues to create a bad impression on almost every review I've read of openSUSE to date. I think that the recommended packages for already installed packages option should be disabled by default, and those "pullin-this-or-that" updater packages should also be eliminated.
The reviewer also ran into the codecs difficulty. He mentions: "Something that I know by experience which a novice user has no way of knowing is that due to license issues openSUSE does not incorporate hardly any multimedia codecs that are usually employed. However, Ubuntu doesn't include them either, but this is not an obstacle to a better multimedia experience thanks to a the simple checkbox during installation. This is, in my opinion, the main defect of openSUSE, easily remedied if one knows a bit about the distro or by reading articles to the tune of 'What to do after installing...' But I try to look at it from the perspective of a less experienced user, who after trying to open a video sees a Kaffeine message that warns about needing to search for 'codecs'. This takes us to YaST, where we have to choose an option to add community repositories, which then takes us to a list of sites whose names, sometimes in English, do not make at all clear which of them to enable or which packages to install. And the worse of it: after adding Packman, Main OSS and Non-OSS, the result is the same: No codecs are found to play the file. Where is the problem? Well, the addition of these repositories will activate a series of packages that will be installed as recommends in YaST, something that an inexperienced user is completely unfamiliar with."
I think he stated the point very well from the viewpoint of a normal user. The comparison with Ubuntu is perhaps a bit wearisome for the openSUSE devs, but it's definitely the elephant in the room here. Any thoughts?
On Thursday 20 November 2014 08.09:59 S. wrote:
I think he stated the point very well from the viewpoint of a normal user. The comparison with Ubuntu is perhaps a bit wearisome for the openSUSE devs, but it's definitely the elephant in the room here. Any thoughts?
What did we really want user that has sold their brain ? Sorry to be harsh. But then why under windows it is so complicated to have software it doesn't mimic ubuntu center things, nor do an apple, nor ...
Come on, it's time to relearn to read. If you start a new adventure getting some lecture is mandatory.
This doesn't excuse us, to have a non working wiki for newbies. But who apart from nearly new comer can wrote it?
Why those kind of people just show after a release, shout and disappear. Why they're still not already actively (like doing stuff, not just blabla) acting.
With the time spent to wrote and read this thread, 2/3 of easy fixes would already have been done.
Guess what, same remarks will occur next release, cause those goose have already jump on a better green grass. If you don't give time/effort whatever is needed now to factory, the next release will be the same. No magic, no fairy ...
I will tell you a secret, there's no other green place than Geeko's land. Aren't we not green for them? Who care if they don't did a bullshit minute of work for our cause.
Can we now concentrate on making factory a bit better tomorrow than yesterday.
Le 20/11/2014 20:54, Bruno Friedmann a écrit :
With the time spent to wrote and read this thread, 2/3 of easy fixes would already have been done.
well... we had already the discussion about such software (I even started one some day), but the net result is: no, we can't do better.
the only thing I could think of to make things better would to give every new repos a lower priority number, to have sofware pulled from them.
Can we now concentrate on making factory a bit better tomorrow than yesterday.
we probably can't on the subject. remember it was even spoken of creating an openSUSE foudation on Man Island on this purpose, but this was not retained
jdd
On Thu, 2014-11-20 at 20:54 +0100, Bruno Friedmann wrote:
On Thursday 20 November 2014 08.09:59 S. wrote:
I think he stated the point very well from the viewpoint of a normal user. The comparison with Ubuntu is perhaps a bit wearisome for the openSUSE devs, but it's definitely the elephant in the room here. Any thoughts?
What did we really want user that has sold their brain ? Sorry to be harsh. But then why under windows it is so complicated to have software it doesn't mimic ubuntu center things, nor do an apple, nor ...
Come on, it's time to relearn to read. If you start a new adventure getting some lecture is mandatory.
This doesn't excuse us, to have a non working wiki for newbies. But who apart from nearly new comer can wrote it?
Why those kind of people just show after a release, shout and disappear. Why they're still not already actively (like doing stuff, not just blabla) acting.
With the time spent to wrote and read this thread, 2/3 of easy fixes would already have been done.
Guess what, same remarks will occur next release, cause those goose have already jump on a better green grass. If you don't give time/effort whatever is needed now to factory, the next release will be the same. No magic, no fairy ...
I will tell you a secret, there's no other green place than Geeko's land. Aren't we not green for them? Who care if they don't did a bullshit minute of work for our cause.
Can we now concentrate on making factory a bit better tomorrow than yesterday.
Bruno Friedmann Ioda-Net Sàrl www.ioda-net.ch
openSUSE Member & Board, fsfe fellowship GPG KEY : D5C9B751C4653227 irc: tigerfoot
I certainly agree with your lament of these potential new users and clickbait review bloggers. Simply shouting and leaving helps nobody.
However, having bug reports go neglected for years doesn't inspire newbs to report. Frankly, it's part of what makes me stop writing reports. I don't know enough to be especially technical without guidance. And since I don't provide particularly technical reports, I get ignored and won't ever have guidance to learn from and make better reports. Worse though is when a fixed problem returns in a next release. My best guess is that these regressions are due to poor project management infrastructure.
Also, it isn't useful to not have any true mechanism for reporting problems. Say some package has insane dependencies or recommends, but doesn't have an explicit entry in the Bugzilla? Or, what if we take the time to send a maintainer an e-mail and never get any response and certainly no resolution?
As you say, this isn't a distro for the entirely brainless whom can't read. On the other hand, our wiki is an abomination and thus we have less documentation than we ought to. To further compound that problem, there is a lack of... presentation, not quite sure how to explain. But let's see an example. On the page where we have our downloads, it has links to comprehensive installation instructions... except on how to make a bootable USB which has risen to enormous popularity. In fact, the only way to find our article on creating a USB is through an external search engine. Mind you, this is nothing that just ANYBODY can fix. I can fix wiki articles, but I have no power over our top domain.
Further on it being an unfriendly for newbs distro... why then are we creating such large default installations? Why would say the base GNOME environment need for example Enterprise SIP telephony client? Why is that part of the pattern, or a recommend? It's unnecessary for mildly experienced users, and useless to newb consumer users. Also, why are we even marketing to the newbs at all? Frankly I find them frustrating since they come to us fleeing Ubuntu for some reason, and then promptly complain that we aren't Ubuntu.
Additionally there are relatively simple things we could do regarding messaging, and making sure new users are informed that we haven't done and nobody seems to care to do at all. Why isn't the KDE SUSEgreeter more useful, or more up-to-date? Why doesn't GNOME have something similar? Hell, we could even use the same one from KDE if we are feeling particularly lazy. Why isn't that greeter blasting users in the face with how we are a volunteer project, and encouraging bug reporting?
Hey,
On 20.11.2014 21:57, Roger Luedecke wrote:
Why
I can tell you why. It'll not be a nice thing to hear for most of us...
Why do your bug reports go unanswered? Because we don't have enough people to support the amount of packages we ship. Too much load onto too few shoulders. Package maintainers _have_ to pick their fights and they pick the ones they can win (easily). And no one feels responsible to help maintainers make the most of bugs.
Why is the wiki a mess? Because we don't have enough people maintaining it. Everyone just adds his 2¢s into it, no one makes sure the whole thing stays usable. People do the absolute necessary, next to no one goes the extra mile. No one feels responsible for taking care of 'functionality' wiki.
Why is our documentation often that bad? Like the Live_USB_stick article. Because everyone just adds whatever he just found out to it and never looks back. No one makes sure that this beast stays recent, readable, comprehensible. People just dump their ideas and run off. No one feels responsible.
Why do those paper-cut issues slip through? Because every one who does something for this distribution is neck deep in putting the open leg fractures in splints. The last 1% to 2% to make something perfect are the ones which are often hardest to do. Just because they are not obvious when your head is buried in the code. You have to have your mind free for this, which is very hard if the next open leg fracture just waits around the corner....
Why is our "marketing" often not to the point? If you simplify there are 2 big things we do each year: Releasing a new version of the distribution and running a conference. We even struggle to get the most basic announcements for those together. Simply because there are not enough people feeling responsible for it.
Why don't we add tons of 'easy' features to our distribution? Why is our infrastructure outdated and largely unmaintained?
Why? The answer is always the same: It's too much to do for not enough people.
This has been a spiral for years now. We have too much on our plate. That's why we struggle to do even do the basics for everything. That's why nothing is very polished. That's why next to no new people join and a lot of people leave.
Sad but true.
Henne
On Fri, 21 Nov 2014 12:36:57 +0100 Henne Vogelsang hvogel@opensuse.org wrote:
Hello,
This has been a spiral for years now. We have too much on our plate. That's why we struggle to do even do the basics for everything. That's why nothing is very polished. That's why next to no new people join and a lot of people leave.
I'm Linux user since '99, although after looking at Suse's release numbers it could be that I did start with 4.4 sometime in the beginning of 1997.
In any case, I left in 2003 to rolling distros - first spent >5yrs with Gentoo, then same amount with Arch and until now with Debian (Sid).
Now I'm returning back to Suse - put 13.2 on my notebook, soon will put it on the desktop and then, hopefully on my Linode VPS. (I'm still considering Tumbleweed, but in any case, it's Suse).
After becoming more familiar (again) with the distro, I'll try to help as much as possible!!
Sincerely, Gour
On Friday 21 November 2014 13.08:54 Gour wrote:
On Fri, 21 Nov 2014 12:36:57 +0100 Henne Vogelsang hvogel@opensuse.org wrote:
Hello,
This has been a spiral for years now. We have too much on our plate. That's why we struggle to do even do the basics for everything. That's why nothing is very polished. That's why next to no new people join and a lot of people leave.
I'm Linux user since '99, although after looking at Suse's release numbers it could be that I did start with 4.4 sometime in the beginning of 1997.
In any case, I left in 2003 to rolling distros - first spent >5yrs with Gentoo, then same amount with Arch and until now with Debian (Sid).
Now I'm returning back to Suse - put 13.2 on my notebook, soon will put it on the desktop and then, hopefully on my Linode VPS. (I'm still considering Tumbleweed, but in any case, it's Suse).
Nope something happened during your vacation on other distribution
openSUSE has appear, a (still) young community project. It's based on effort of (still) too few contributor's shoulder which are not payed by SUSE. Beside that, SUSE still offer to the community numerous very valuable resources and contribution.
After becoming more familiar (again) with the distro, I'll try to help as much as possible!!
Try also to check the project as a whole too ;-)
Sincerely, Gour
On November 21, 2014 7:52:23 AM EST, Bruno Friedmann bruno@ioda-net.ch wrote:
On Friday 21 November 2014 13.08:54 Gour wrote:
On Fri, 21 Nov 2014 12:36:57 +0100 Henne Vogelsang hvogel@opensuse.org wrote:
Hello,
This has been a spiral for years now. We have too much on our
plate.
That's why we struggle to do even do the basics for everything.
That's
why nothing is very polished. That's why next to no new people join and a lot of people leave.
I'm Linux user since '99, although after looking at Suse's release numbers it could be that I did start with 4.4 sometime in the beginning of 1997.
In any case, I left in 2003 to rolling distros - first spent >5yrs
with
Gentoo, then same amount with Arch and until now with Debian (Sid).
Now I'm returning back to Suse - put 13.2 on my notebook, soon will
put
it on the desktop and then, hopefully on my Linode VPS. (I'm still considering Tumbleweed, but in any case, it's Suse).
Nope something happened during your vacation on other distribution
openSUSE has appear, a (still) young community project. It's based on effort of (still) too few contributor's shoulder which are not payed by SUSE. Beside that, SUSE still offer to the community numerous very valuable resources and contribution.
After becoming more familiar (again) with the distro, I'll try to
help
as much as possible!!
Try also to check the project as a whole too ;-)
For me as a contributor, the greatest part of openSUSE is 2-fold:
- The do-apoly nature of package inclusion such that packages I care about get in the distro simply because I am willing to package/maintain them.
- OBS - that is some great infrastructure and I suspect the glue that holds opensuse together. That it got a major set of improvements in the last year, just makes it better.
Greg
Le 21/11/2014 12:36, Henne Vogelsang a écrit :
Hey,
On 20.11.2014 21:57, Roger Luedecke wrote:
Why
I can tell you why. It'll not be a nice thing to hear for most of us...
I don't agree entirely. searching to have more order send many people out. more below
Why do your bug reports go unanswered? Because we don't have enough people to support the amount of packages we ship.
bugzilla needs enhancement. Not to say it's easy. I think some more people could be allowed to give help, at least sorting bugs and helping bug writers to better make the report. Some kind of (internal?) doc is needed, for example to know to who assign a bug
Why is the wiki a mess? Because we don't have enough people maintaining it.
it's there than I most desagree. A wiki *have to be a mess*. Every body have to be able to write there without "not in subject", "have to be reviewed" etc. I used to contribute a lot and was completely disgusted by the last revamp some years ago. Google is good enough to find the data in the mess. This don't mean people can't make order in the mess, but only on a voluntary basis.
SDB and HOTOS are a good place to order things if possible
Why is our documentation often that bad? Like the Live_USB_stick article. Because everyone just adds whatever he just found out to it and never looks back. No one makes sure that this beast stays recent, readable, comprehensible. People just dump their ideas and run off. No one feels responsible.
as every where in the web. And our documentation is far from bad. I use it very often and very often with success.
sure, it would be nice to have a writer's team to write a manual for each distro version, but this is only a dream.
Why do those paper-cut issues slip through? Because every one who does something for this distribution is neck deep in putting the open leg fractures in splints.
this is not that important. openSUSE is still one of the first choice distribution.
I see this (packman) more as a permanent chapter to add to the release notes. the release notes themselve are very difficult to write because it's hard to get the info from the devs (not a rant, we are all short of time, just a fact)
Why is our "marketing" often not to the point? If you simplify there are 2 big things we do (...)
sure, but also the marketting team have a complex situation: gather informations and write papers at least. The result is not that bad. Only one people full time (not even sure, is still Jos in charge?)
Why don't we add tons of 'easy' features to our distribution? Why is our infrastructure outdated and largely unmaintained?
Why? The answer is always the same: It's too much to do for not enough people.
This has been a spiral for years now. We have too much on our plate. That's why we struggle to do even do the basics for everything. That's why nothing is very polished. That's why next to no new people join and a lot of people leave.
I was thinking french people where the only one to bash themselves, but looks like there are not alone. :-)
Henne, you are one of the most valuable openSUSE member and you make a splendid job. openSUSE is, may be, the most polished distribution in the world! and I was last week next Debian, Fedora and ubuntu fans, and I can say my booth was far the smarter
our booth:
http://dodin.info/piwigo/picture.php?/106543-dsc02518_25029/category/5675
Debian (left) and Fedora (right)
http://dodin.info/piwigo/picture.php?/106544-dsc02519_4735/category/5675
Debian goodies where for sale when ours are free and the second day there where nobody else than us...
what you describe is the life. We need more flexibility, not more stress :-)
and let me thanks also everybody in the SUSE team. It's a thing I write in the papers, we have a sponsor extremely involved, including in free time, and it's a great strenght
jdd
Hey,
On 21.11.2014 13:17, jdd wrote:
Le 21/11/2014 12:36, Henne Vogelsang a écrit :
On 20.11.2014 21:57, Roger Luedecke wrote:
Why
I can tell you why. It'll not be a nice thing to hear for most of us...
I don't agree entirely. searching to have more order send many people out. more below
Why do your bug reports go unanswered? Because we don't have enough people to support the amount of packages we ship.
bugzilla needs enhancement. Not to say it's easy. I think some more people could be allowed to give help, at least sorting bugs and helping bug writers to better make the report. Some kind of (internal?) doc is needed, for example to know to who assign a bug
And why do you think bugzilla enhancements don't get code, screening teams don't get formed or help documents don't get written?
Why is the wiki a mess? Because we don't have enough people maintaining it.
Google is good enough to find the data in the mess.
That's beside the point. If there is no one taking care of making your wiki (the tool) and it's content consumable then it doesn't matter if it gets found or not. Because once you've found it you can't do anything with it.
Why do those paper-cut issues slip through? Because every one who does something for this distribution is neck deep in putting the open leg fractures in splints.
this is not that important. openSUSE is still one of the first choice distribution.
Because we're still pulling this off. Is this sustainable in the future? I have doubts.
Why is our "marketing" often not to the point? If you simplify there are 2 big things we do (...)
sure, but also the marketing team have a complex situation: gather informations and write papers at least.
If it would be complex and cumbersome it would probably get done automatically.
The result is not that bad. Only one people full time (not even sure, is still Jos in charge?)
No one is in charge. SUSE hired someone to help with this (hey Douglas) but until he is up to speed it will be some time. And is that something that sounds healthy? That we need a sponsor to hire someone to do this?
Why don't we add tons of 'easy' features to our distribution? Why is our infrastructure outdated and largely unmaintained?
Why? The answer is always the same: It's too much to do for not enough people.
This has been a spiral for years now. We have too much on our plate. That's why we struggle to do even do the basics for everything. That's why nothing is very polished. That's why next to no new people join and a lot of people leave.
I was thinking french people where the only one to bash themselves, but looks like there are not alone. :-)
Henne, you are one of the most valuable openSUSE member and you make a splendid job.
Thank you Jean. Appreciated :-)
Still I'm more than worried about the future of this project and distribution. To quote one of my favourite (French) Movies:
There is a man falling from a 50-story sky scraper. To calm himself he is saying to himself: So far so good, so far so good, so far so good. But he misses the point, the important part is not the fall, it's the impact.
https://www.youtube.com/watch?v=yk77VrkxL88
Henne
On Friday 21 November 2014 12.36:57 Henne Vogelsang wrote:
Hey,
On 20.11.2014 21:57, Roger Luedecke wrote:
Why
I can tell you why. It'll not be a nice thing to hear for most of us...
Why do your bug reports go unanswered? Because we don't have enough people to support the amount of packages we ship. Too much load onto too few shoulders. Package maintainers _have_ to pick their fights and they pick the ones they can win (easily). And no one feels responsible to help maintainers make the most of bugs.
Why is the wiki a mess? Because we don't have enough people maintaining it. Everyone just adds his 2¢s into it, no one makes sure the whole thing stays usable. People do the absolute necessary, next to no one goes the extra mile. No one feels responsible for taking care of 'functionality' wiki.
Why is our documentation often that bad? Like the Live_USB_stick article. Because everyone just adds whatever he just found out to it and never looks back. No one makes sure that this beast stays recent, readable, comprehensible. People just dump their ideas and run off. No one feels responsible.
Why do those paper-cut issues slip through? Because every one who does something for this distribution is neck deep in putting the open leg fractures in splints. The last 1% to 2% to make something perfect are the ones which are often hardest to do. Just because they are not obvious when your head is buried in the code. You have to have your mind free for this, which is very hard if the next open leg fracture just waits around the corner....
Why is our "marketing" often not to the point? If you simplify there are 2 big things we do each year: Releasing a new version of the distribution and running a conference. We even struggle to get the most basic announcements for those together. Simply because there are not enough people feeling responsible for it.
Why don't we add tons of 'easy' features to our distribution? Why is our infrastructure outdated and largely unmaintained?
Why? The answer is always the same: It's too much to do for not enough people.
This has been a spiral for years now. We have too much on our plate. That's why we struggle to do even do the basics for everything. That's why nothing is very polished. That's why next to no new people join and a lot of people leave.
Sad but true.
Henne
Thanks.
Hi everyone,
I'm the starter of this thread, and I definitely don't think that all is doom and gloom for openSUSE. I think I mentioned that I have also switched back to openSUSE thanks to the combination of the new Tumbleweed availability and the 13.2 fixed release. I think that the OBS and YaST are two defining features of openSUSE that make it uniquely valuable for my needs, and Packman is also superb, albeit poorly integrated by openSUSE.
As far as bug handling, I have actually found openSUSE to fairly efficient. I reported a *LOT* of bugs against Factory rolling (now Tumbleweed) and the 13.1 RC, ranging from a few fairly serious, others petty, others "nice to have" requests. Overall, I was fairly impressed with the response I got. In fact, I would say that my experience with bug reporting was significantly better than Arch and Ubuntu, and infinitely better than the offshoot projects that depend completely on a parent distro and don't really fix their own core OS bugs (Mint, Manjaro, etc).
With regards to documentation, frankly I feel that *all* distros except Arch have pretty awful documentation. Ubuntu is huge and commercial, but their documentation is also horrible-- incomplete and terribly out of date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the underlying system more unified across distros, I just use the Arch wiki for almost everything, and it's mostly applicable to openSUSE too.
I don't mean to give the wrong impression with this thread. As I mentioned, I wanted to gauge interest in fixing smallish "paper-cut" issues, not core, fundamental, ideological, or infrastructure-related changes. I don't think openSUSE should be a clone of Ubuntu, but there is considerable room for improvement in terms of ease-of-use and refinement.
On Friday 21 November 2014 09.19:20 S. wrote:
Hi everyone,
I'm the starter of this thread, and I definitely don't think that all is doom and gloom for openSUSE. I think I mentioned that I have also switched back to openSUSE thanks to the combination of the new Tumbleweed availability and the 13.2 fixed release. I think that the OBS and YaST are two defining features of openSUSE that make it uniquely valuable for my needs, and Packman is also superb, albeit poorly integrated by openSUSE.
Again, this is due to all the Patent/crap troll.
openSUSE/SUSE is not allowed to make it easy (change the law, and it will be really easy for us to improve the situation) This means, next time don't vote for those who defend those way of thinking.
It's not allowed to pre-install packman or other same kind of repositories on end-user computers. The community repo you are seeing in Yast is a awfully complicated ping-pong (really). But even if we can still improve a bit the situation, with concrete proposal, I don't believe the way you desire it is really possible.
As far as bug handling, I have actually found openSUSE to fairly efficient. I reported a *LOT* of bugs against Factory rolling (now Tumbleweed) and the 13.1 RC, ranging from a few fairly serious, others petty, others "nice to have" requests. Overall, I was fairly impressed with the response I got. In fact, I would say that my experience with bug reporting was significantly better than Arch and Ubuntu, and infinitely better than the offshoot projects that depend completely on a parent distro and don't really fix their own core OS bugs (Mint, Manjaro, etc).
Good to know, and yeah the team behind triage etc on bugzilla deserve a more thankful.
With regards to documentation, frankly I feel that *all* distros except Arch have pretty awful documentation. Ubuntu is huge and commercial, but their documentation is also horrible-- incomplete and terribly out of date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the underlying system more unified across distros, I just use the Arch wiki for almost everything, and it's mostly applicable to openSUSE too.
Did you check the http://activedoc.opensuse.org/ website. Did you see that website is totally collaborative once you register and login ? Aka, to improve the doc, you can propose, comment etc ???
I don't mean to give the wrong impression with this thread. As I mentioned, I wanted to gauge interest in fixing smallish "paper-cut" issues, not core, fundamental, ideological, or infrastructure-related changes. I don't think openSUSE should be a clone of Ubuntu, but there is considerable room for improvement in terms of ease-of-use and refinement.
For sure, but as already stated, we need people interest in, that have also the strength to commitment. Sure we're legion, but we need legion^2 :-)
Thanks anyway for your report, if we're still fighting, just prove we're still alive :-)
On Mon, 2014-11-24 at 13:20 +0100, Bruno Friedmann wrote:
With regards to documentation, frankly I feel that *all* distros
except
Arch have pretty awful documentation. Ubuntu is huge and commercial,
but
their documentation is also horrible-- incomplete and terribly out
of
date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the
underlying
system more unified across distros, I just use the Arch wiki for
almost
everything, and it's mostly applicable to openSUSE too.
Did you check the http://activedoc.opensuse.org/ website. Did you see that website is totally collaborative once you register and login ? Aka, to improve the doc, you can propose, comment etc ???
THIS IS EXACTLY one of the things I've been bitching about. Yet another undocumented mystery sub-domain. Want to know why nobody sticks around? It's because you have to lurk for a year or have personal connections before you can even navigate!
I don't mean to give the wrong impression with this thread. As I mentioned, I wanted to gauge interest in fixing smallish
"paper-cut"
issues, not core, fundamental, ideological, or
infrastructure-related
changes. I don't think openSUSE should be a clone of Ubuntu, but
there
is considerable room for improvement in terms of ease-of-use and
refinement.
For sure, but as already stated, we need people interest in, that have also the strength to commitment. Sure we're legion, but we need legion^2 :-)
Thanks anyway for your report, if we're still fighting, just prove we're still alive :-)
I think we can fix all this, but when I've proposed it in the past I got resistance to the point of obstruction. Lots of people saying how it couldn't be done, only a few saying lets give it a shot. I busted my ass trying to do it (Richard Brown even provided a VPS that I couldn't remember how to use at the time) and even managed to recruit developers help from Novell! Yet we couldn't do it, because of the obstructionary nay-saying.
On Monday 24 November 2014 09.15:47 Roger Luedecke wrote:
On Mon, 2014-11-24 at 13:20 +0100, Bruno Friedmann wrote:
With regards to documentation, frankly I feel that *all* distros
except
Arch have pretty awful documentation. Ubuntu is huge and commercial,
but
their documentation is also horrible-- incomplete and terribly out
of
date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the
underlying
system more unified across distros, I just use the Arch wiki for
almost
everything, and it's mostly applicable to openSUSE too.
Did you check the http://activedoc.opensuse.org/ website. Did you see that website is totally collaborative once you register and login ? Aka, to improve the doc, you can propose, comment etc ???
THIS IS EXACTLY one of the things I've been bitching about. Yet another undocumented mystery sub-domain. Want to know why nobody sticks around? It's because you have to lurk for a year or have personal connections before you can even navigate!
Stop bitching, it is not unknown really. it was rtfm.o.o then doc.o.o and as the tool change finally got activedoc
but was always available
First link on web search engine with "opensuse documentation" keywords Available from home page of wiki -> Portal:Documentation http://en.opensuse.org/Portal:Documentation Also well known url since years doc.opensuse.org
You lost this time, with a bad argument. Doc has always been visible and accessible.
I don't mean to give the wrong impression with this thread. As I mentioned, I wanted to gauge interest in fixing smallish
"paper-cut"
issues, not core, fundamental, ideological, or
infrastructure-related
changes. I don't think openSUSE should be a clone of Ubuntu, but
there
is considerable room for improvement in terms of ease-of-use and
refinement.
For sure, but as already stated, we need people interest in, that have also the strength to commitment. Sure we're legion, but we need legion^2 :-)
Thanks anyway for your report, if we're still fighting, just prove we're still alive :-)
I think we can fix all this, but when I've proposed it in the past I got resistance to the point of obstruction. Lots of people saying how it couldn't be done, only a few saying lets give it a shot. I busted my ass trying to do it (Richard Brown even provided a VPS that I couldn't remember how to use at the time) and even managed to recruit developers help from Novell! Yet we couldn't do it, because of the obstructionary nay-saying.
At Tue, 25 Nov 2014 09:16:59 +0100, Bruno Friedmann wrote:
On Monday 24 November 2014 09.15:47 Roger Luedecke wrote:
On Mon, 2014-11-24 at 13:20 +0100, Bruno Friedmann wrote:
With regards to documentation, frankly I feel that *all* distros
except
Arch have pretty awful documentation. Ubuntu is huge and commercial,
but
their documentation is also horrible-- incomplete and terribly out
of
date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the
underlying
system more unified across distros, I just use the Arch wiki for
almost
everything, and it's mostly applicable to openSUSE too.
Did you check the http://activedoc.opensuse.org/ website. Did you see that website is totally collaborative once you register and login ? Aka, to improve the doc, you can propose, comment etc ???
THIS IS EXACTLY one of the things I've been bitching about. Yet another undocumented mystery sub-domain. Want to know why nobody sticks around? It's because you have to lurk for a year or have personal connections before you can even navigate!
Stop bitching, it is not unknown really. it was rtfm.o.o then doc.o.o and as the tool change finally got activedoc
Even after few years in community and a lot of personal connections I haven't heard about this domain so far. I don't think he's bitching ;)
Best regards,
S_W
On Tue, Nov 25, 2014 at 4:56 AM, Tomas Cech tcech@suse.cz wrote:
At Tue, 25 Nov 2014 09:16:59 +0100, Bruno Friedmann wrote:
On Monday 24 November 2014 09.15:47 Roger Luedecke wrote:
On Mon, 2014-11-24 at 13:20 +0100, Bruno Friedmann wrote:
With regards to documentation, frankly I feel that *all* distros
except
Arch have pretty awful documentation. Ubuntu is huge and commercial,
but
their documentation is also horrible-- incomplete and terribly out
of
date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the
underlying
system more unified across distros, I just use the Arch wiki for
almost
everything, and it's mostly applicable to openSUSE too.
Did you check the http://activedoc.opensuse.org/ website. Did you see that website is totally collaborative once you register and login ? Aka, to improve the doc, you can propose, comment etc ???
THIS IS EXACTLY one of the things I've been bitching about. Yet another undocumented mystery sub-domain. Want to know why nobody sticks around? It's because you have to lurk for a year or have personal connections before you can even navigate!
Stop bitching, it is not unknown really. it was rtfm.o.o then doc.o.o and as the tool change finally got activedoc
Even after few years in community and a lot of personal connections I haven't heard about this domain so far. I don't think he's bitching ;)
Best regards,
S_W
It is documented in the wiki:
https://en.opensuse.org/openSUSE:Documentation_Contribute
If you look at doc.opensuse.org you will see a note:
== Since openSUSE 13.1 all openSUSE related documentation moved to our community editing server ActiveDoc.opensuse.org and the links below are only retained here for reference. ==
Unfortunately that note is not very well highlighted so users could miss it.
Also, I note that on wiki's header template there is a pull-down for "support - documentation" that still points at the old server.
Fixing that is above my pay grade.
Greg
Le 25/11/2014 17:19, Greg Freemyer a écrit :
Since openSUSE 13.1 all openSUSE related documentation moved to our community editing server ActiveDoc.opensuse.org and the links below are only retained here for reference.
does this mean the wiki is obsolete? I don't really understand this sentence :-(
thanks
jdd
On Tue, Nov 25, 2014 at 11:23 AM, jdd lesrevesdeness@gmail.com wrote:
Le 25/11/2014 17:19, Greg Freemyer a écrit :
Since openSUSE 13.1 all openSUSE related documentation moved to our community editing server ActiveDoc.opensuse.org and the links below are only retained here for reference.
does this mean the wiki is obsolete? I don't really understand this sentence :-(
thanks
jdd
For years openSUSE has had 2 primary documentation sources:
The first is the wiki. The second is the manuals. The manuals I assume are what used to come in the box sets.
The manuals are listed at https://en.opensuse.org/openSUSE:Documentation_List_of_Manuals
In days past the manuals were maintained exclusively by SUSE/Novell employees I think.
In order to make the "manuals" community supportable they have moved them to https://activedoc.opensuse.org
For instance the security guide is at: https://activedoc.opensuse.org/book/opensuse-security-guide
If you want to contribute to the manuals: https://en.opensuse.org/openSUSE:Documentation_Contribute
Greg
very well explained, you should copy/paste this on the documentation wiki page :-)
thanks jdd
For years openSUSE has had 2 primary documentation sources:
The first is the wiki. The second is the manuals. The manuals I assume are what used to come in the box sets.
The manuals are listed at https://en.opensuse.org/openSUSE:Documentation_List_of_Manuals
In days past the manuals were maintained exclusively by SUSE/Novell employees I think.
In order to make the "manuals" community supportable they have moved them to https://activedoc.opensuse.org
For instance the security guide is at: https://activedoc.opensuse.org/book/opensuse-security-guide
If you want to contribute to the manuals: https://en.opensuse.org/openSUSE:Documentation_Contribute
Greg
Most of the info is already there:
https://en.opensuse.org/Documentation
I'd make some edits, but I'm not sure they would help.
The bigger issue is getting people to read that page. (today i read it for the first time.)
Greg
On Tue, Nov 25, 2014 at 1:43 PM, jdd lesrevesdeness@gmail.com wrote:
very well explained, you should copy/paste this on the documentation wiki page :-)
thanks jdd
For years openSUSE has had 2 primary documentation sources:
The first is the wiki. The second is the manuals. The manuals I assume are what used to come in the box sets.
The manuals are listed at https://en.opensuse.org/openSUSE:Documentation_List_of_Manuals
In days past the manuals were maintained exclusively by SUSE/Novell employees I think.
In order to make the "manuals" community supportable they have moved them to https://activedoc.opensuse.org
For instance the security guide is at: https://activedoc.opensuse.org/book/opensuse-security-guide
If you want to contribute to the manuals: https://en.opensuse.org/openSUSE:Documentation_Contribute
Greg
-- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org
On 11/25/2014 10:53 AM, Greg Freemyer wrote:
Most of the info is already there:
https://en.opensuse.org/Documentation
I'd make some edits, but I'm not sure they would help.
The bigger issue is getting people to read that page. (today i read it for the first time.)
Greg
On Tue, Nov 25, 2014 at 1:43 PM, jdd lesrevesdeness@gmail.com wrote:
very well explained, you should copy/paste this on the documentation wiki page :-)
thanks jdd
For years openSUSE has had 2 primary documentation sources:
The first is the wiki. The second is the manuals. The manuals I assume are what used to come in the box sets.
The manuals are listed at https://en.opensuse.org/openSUSE:Documentation_List_of_Manuals
In days past the manuals were maintained exclusively by SUSE/Novell employees I think.
In order to make the "manuals" community supportable they have moved them to https://activedoc.opensuse.org
For instance the security guide is at: https://activedoc.opensuse.org/book/opensuse-security-guide
If you want to contribute to the manuals: https://en.opensuse.org/openSUSE:Documentation_Contribute
Greg
-- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org
I don't think anybody trusts the wiki. It's so often out of date, and such a mess to navigate that it really doesn't lend one to trust it. Case in point, the Bumblebee article specifically makes mention of "if you don't trust this guide, here is an alternative."
Le 26/11/2014 18:51, Roger Luedecke a écrit :
I don't think anybody trusts the wiki. It's so often out of date, and such a mess to navigate that it really doesn't lend one to trust it.
do you trust internet :-)? it's even worst
google find very often solutions in the opensuse forum
jdd
On Tue, Nov 25, 2014 at 7:19 PM, Greg Freemyer greg.freemyer@gmail.com wrote:
On Tue, Nov 25, 2014 at 4:56 AM, Tomas Cech tcech@suse.cz wrote:
At Tue, 25 Nov 2014 09:16:59 +0100, Bruno Friedmann wrote:
On Monday 24 November 2014 09.15:47 Roger Luedecke wrote:
On Mon, 2014-11-24 at 13:20 +0100, Bruno Friedmann wrote:
With regards to documentation, frankly I feel that *all* distros
except
Arch have pretty awful documentation. Ubuntu is huge and commercial,
but
their documentation is also horrible-- incomplete and terribly out
of
date. Debian's is also very lacking and assumes great technical knowledge. Only Arch is a shining star of excellent documentation. Fortunately with systemd and other projects that make the
underlying
system more unified across distros, I just use the Arch wiki for
almost
everything, and it's mostly applicable to openSUSE too.
Did you check the http://activedoc.opensuse.org/ website. Did you see that website is totally collaborative once you register and login ? Aka, to improve the doc, you can propose, comment etc ???
THIS IS EXACTLY one of the things I've been bitching about. Yet another undocumented mystery sub-domain. Want to know why nobody sticks around? It's because you have to lurk for a year or have personal connections before you can even navigate!
Stop bitching, it is not unknown really. it was rtfm.o.o then doc.o.o and as the tool change finally got activedoc
Even after few years in community and a lot of personal connections I haven't heard about this domain so far. I don't think he's bitching ;)
Best regards,
S_W
It is documented in the wiki:
https://en.opensuse.org/openSUSE:Documentation_Contribute
If you look at doc.opensuse.org you will see a note:
== Since openSUSE 13.1 all openSUSE related documentation moved to our community editing server ActiveDoc.opensuse.org and the links below are only retained here for reference. ==
Unfortunately that note is not very well highlighted so users could miss it.
Why not move ActiveDoc.opensuse.org to doc.opensuse.org?
Mustafa
Also, I note that on wiki's header template there is a pull-down for "support - documentation" that still points at the old server.
Fixing that is above my pay grade.
Greg
To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org
Greg Freemyer wrote:
https://en.opensuse.org/Documentation
I'd make some edits, but I'm not sure they would help.
The bigger issue is getting people to read that page.
Yep, that's the truth. That's why I'm not a huge fan of things like "Steps to take after installing openSUSE" type of pages with 1-click installers to add support for this or that. The problem is that many users are just not accustomed to looking for solutions online, they expect a button or prompt integrated into the operating system that offers the functionality in question.
I still would like to suggest a YaST module for this purpose. There actually used to be an non-free add-on CD for old versions of openSUSE, with a YaST module to add it to the software sources. Maybe a module similar to that could be brought back to life. It could offer checkboxes for "Install proprietary multimedia support" and "Install proprietary Nvidia driver" and "Install proprietary Radeon driver". These checkboxes could run a script that would add the appropriate 3rd-party non-openSUSE repo (Packman, Nvida, etc.), set the priority, and install the packages.
On 11/25/2014 12:29 PM, S. wrote:
Greg Freemyer wrote:
https://en.opensuse.org/Documentation
I'd make some edits, but I'm not sure they would help.
The bigger issue is getting people to read that page.
Yep, that's the truth. That's why I'm not a huge fan of things like "Steps to take after installing openSUSE" type of pages with 1-click installers to add support for this or that. The problem is that many users are just not accustomed to looking for solutions online, they expect a button or prompt integrated into the operating system that offers the functionality in question.
I still would like to suggest a YaST module for this purpose. There actually used to be an non-free add-on CD for old versions of openSUSE, with a YaST module to add it to the software sources. Maybe a module similar to that could be brought back to life. It could offer checkboxes for "Install proprietary multimedia support" and "Install proprietary Nvidia driver" and "Install proprietary Radeon driver". These checkboxes could run a script that would add the appropriate 3rd-party non-openSUSE repo (Packman, Nvida, etc.), set the priority, and install the packages.
I believe we still have that module. It is from the SUSE Enterprise legacy, where an OEM might have special drivers or something. A YaST module still wouldn't be a solution. I'd say a tiny 'Hello World' type pop-up program that says, "we are not Ubuntu so don't expect Ubuntu. Do not make demands, you get what you pay for and openSUSE is free. Google is your friend, and reading won't make you turn to Satan."
Le 26/11/2014 18:57, Roger Luedecke a écrit :
I still would like to suggest a YaST module for this purpose. There actually used to be an non-free add-on CD for old versions of openSUSE, with a YaST module to add it to the software sources. Maybe a module similar to that could be brought back to life
that's a good idea. Could it be possible to create a "multimedia" dvd (or cd) let alone with the packman repo?
this cd could ba available on a community web site
jdd
On 11/26/2014 10:08 AM, jdd wrote:
Le 26/11/2014 18:57, Roger Luedecke a écrit :
I still would like to suggest a YaST module for this purpose. There actually used to be an non-free add-on CD for old versions of openSUSE, with a YaST module to add it to the software sources. Maybe a module similar to that could be brought back to life
that's a good idea. Could it be possible to create a "multimedia" dvd (or cd) let alone with the packman repo?
this cd could ba available on a community web site
jdd
I've been contemplating just that. The ability to create a driver disk from Packman and the prop. drivers. I imagine there are probably licensing issues pertaining to that, though I think it'd be nice.
Am Montag 17 November 2014, 14:48:35 schrieb S.:
( 'paper-cut' issues)
- Ugly font rendering
Are you willing to pay to license the patented technologies that make font redering prettier?
- Automatic installation of extraneous/unrequested packages, especially
after initial installation.
...dependencies. Any particular examples?
- Crippled packages meant to prevent compatibility with proprietary
multimedia formats.
see 1). Many currently widespread audio/video codecs are patented.
- Difficult to install multimedia and/or proprietary formats and drivers.
see 1) and 3).
- Breakage of YaST modules.
...examples?
So basically your list of five paper cuts comes down to one central problem that can't be solved, and that is commonly known as "software patents", and two technical problems that might be solved if there were actual examples for what the problem is.
by the way, where does one find that "latest distrowatch review", google does not know it.
cheers MH
On Tue, Nov 18, 2014 at 12:05 AM, Mathias Homann Mathias.Homann@opensuse.org wrote:
by the way, where does one find that "latest distrowatch review", google does not know it.
cheers MH
Hello Mathias,
I believe he was referring to the following review => http://distrowatch.com/weekly.php?issue=20141117#feature
Besides, we could maybe take his critic as positive. It doesn't really harm the project.
Ish Sookun
- Geek by birth ... Linux by choice. +-+-+-+-+-+-+-+-+-+-+ |H|A|C|K|L|O|G|.|i|n| +-+-+-+-+-+-+-+-+-+-+
Le 17/11/2014 21:21, Ish Sookun a écrit :
I believe he was referring to the following review => http://distrowatch.com/weekly.php?issue=20141117#feature
Besides, we could maybe take his critic as positive. It doesn't really harm the project.
very interesting. I didn't run in most of the problem signaled, but this review should be verifyed and if bug remain, fixed
jdd
On Tue, 2014-11-18 at 09:18 +0100, jdd wrote:
Le 17/11/2014 21:21, Ish Sookun a écrit :
I believe he was referring to the following review => http://distrowatch.com/weekly.php?issue=20141117#feature
Besides, we could maybe take his critic as positive. It doesn't really harm the project.
very interesting. I didn't run in most of the problem signaled, but this review should be verifyed and if bug remain, fixed
jdd
In my experience the packaging issues and YaST issues are spot-on.
On Tue, Nov 18, 2014 at 10:36:58AM -0800, Roger Luedecke wrote:
On Tue, 2014-11-18 at 09:18 +0100, jdd wrote:
Le 17/11/2014 21:21, Ish Sookun a écrit :
I believe he was referring to the following review => http://distrowatch.com/weekly.php?issue=20141117#feature
Besides, we could maybe take his critic as positive. It doesn't really harm the project.
very interesting. I didn't run in most of the problem signaled, but this review should be verifyed and if bug remain, fixed
In my experience the packaging issues and YaST issues are spot-on.
Has one of you thought about filing bug reports for the individual issues and create one meta bug which references all these reports?
Else I'm sure nothing will change. Cause we all only fix issues we are getting aware of. And the tool to track individual issues is bugzilla. One technical issue == one bug report.
Sending a general complain and reference to a review report doesn't drive anyone at the development side.
Cheers,
Lars
Le 18/11/2014 20:01, Lars Müller a écrit :
Sending a general complain and reference to a review report doesn't drive anyone at the development side.
oh oh, no "internet survey" on the devs part :-))
jdd
On Mon, 2014-11-17 at 14:48 -0500, S. wrote:
Hi everyone,
I've been an intermittent user of SuSE and openSUSE for well over a decade, starting back in the 6.x days if I remember correctly. I've always admired the project for its professionalism and broad scope. I recently switched back to openSUSE thanks to the rolling Factory (now Tumbleweed) announcement, and I also have 13.2 installed on other machines. I'm really enjoying the experience, and this time, I'd like to stay.
I really want to see openSUSE succeed, but unfortunately I tend to see it dismissed by potential new users as "a nice distro, but..." followed by certain annoyances or "paper-cut" issues. This week's openSUSE 13.2 review on Distrowatch is the most recent example of such comments. I feel that openSUSE is fundamentally extremely well designed and executed, but I have also noticed the same recurring "paper-cut" issues year after year, release after release. So I'm wondering if anything could be done about them, or if the developers have any interest in fixing them.
Without going into detail, here are a few general areas of concern that I often hear and/or have personally noticed throughout many openSUSE releases:
- Ugly font rendering
- Automatic installation of extraneous/unrequested packages, especially
after initial installation. 3. Crippled packages meant to prevent compatibility with proprietary multimedia formats. 4. Difficult to install multimedia and/or proprietary formats and drivers. 5. Breakage of YaST modules.
All of the above issues do have workarounds, and think it would be best for this thread to not get into the technical details of these aforementioned issues. But in general terms, these issues come up time after time in reviews and forum posts about openSUSE. So my question is: Would the openSUSE project be interested in working to resolve any or all of these issues? And if so, where would be the best place for me to bring up these issues and work with developers to make improvements? Some of the issues are not related to any one specific package, but are rather of a more systemic nature. In some cases legal/patent issues are probably involved. (Again, let's please avoid legal discussions in this thread.) Although I am not a coder or developer *at all*, I do have an eye for detail and polish, and I have a good idea of what typical users expect out of a Linux distro on the desktop. I'd like to help to identify and test solutions to these papercuts if any developers are willing to look into some of these long-standing issues with a fresh eye.
Any comments? Thanks in advance for your time and consideration.
It should be understood of course in regards to patented or licensed technologies, there is little we can do about that. On the other hand, the font rendering issue is being worked on actually and should be fixed in an update soon.
As for installing proprietary drivers, I'm not sure how much easier we could really make it. Most reviews complaining about that are due to a fundamental misunderstanding of how it works with us, and they are expecting it to be like Ubuntu which has an in your face tool for it. Whereas we simply add a repo, and things are generally handled automatically with an update so long as 'recommends' haven't been disabled.
I don't know about our packages being crippled to complicate using codecs etc, though it's easy for me at least to see why you'd think so. Multimedia packages seem to be poorly maintained, and I often have problems with them such as the segfault with .avi on Totem even after installing the Fluendo Codec Pack.
I have witnessed the problem of broken YaST modules, in my case it being the System Services module doesn't actually seem to do anything at all when you change something. I'm inclined to suspect this has to do with it not being fully updated to work with the way systemd deals with services.
In general I understand your concern regarding papercut issues. I think it has to do with the phenomenon of 'cat-herding'. We have far too many communication channels which are poorly integrated, and often underused. My solution I was pushing for was to adopt the Kablink which is the FOSS Novell Vibe, a web-based team collaboration platform developed by Novell. My idea was not merely to add it as YASD (Yet Another Sub-Domain) but to make it front and central to replace and supplement some existing channels, while being tied in with others via scripted extensions. This was however shot down since people didn't like the interface, and felt we had inadequate web developers to make it happen. However, I recently discovered the addition of a Redmine install in a subdomain which hardly anybody knows exists, and is even less useful and redundant than Kablink.
Towards the above point, I think that would still be the final solution. We need an integrated 'command-center' for the project so that communication will be more transparent and issues can be tracked in order to avoid regressions. I have been considering approaching the community as a whole with the idea again, but this time proposing a Kickstarter or IndieGoGo campaign so we can hire some dedicated developers to make sure the task is done correctly and completely.
On Mon, 17 Nov 2014 23:05:25 -0800 Roger Luedecke roger.luedecke@gmail.com wrote:
Hello Roger,
I'm atm using Debian (sid) and evaluating (under vbox) 13.2 for my desktop/notebook and eventually VPS (Linode). Interestingly, I've started Linux with Suse in 1999 (Lenz was great support guy at the time), but later moved away due to rpm-dep-hell problems.
Now I see that Zypp seems to be great dep-solver and I'm on the path back. :-)
On the other hand, the font rendering issue is being worked on actually and should be fixed in an update soon.
I'm glad to hear it!
As for installing proprietary drivers, I'm not sure how much easier we could really make it. Most reviews complaining about that are due to a fundamental misunderstanding of how it works with us, and they are expecting it to be like Ubuntu which has an in your face tool for it. Whereas we simply add a repo, and things are generally handled automatically with an update so long as 'recommends' haven't been disabled.
I also read that 'review', but didn't get that the solution is to 'just add the repo', so it does not sound bad, imho.
In general I understand your concern regarding papercut issues. I think it has to do with the phenomenon of 'cat-herding'. We have far too many communication channels which are poorly integrated, and often underused.
The number of mailing lists, IRC channels etc. is really overwhelming, so cutting those short a bit and focus/integrate those might be good strategy.
Towards the above point, I think that would still be the final solution. We need an integrated 'command-center' for the project so that communication will be more transparent and issues can be tracked in order to avoid regressions. I have been considering approaching the community as a whole with the idea again, but this time proposing a Kickstarter or IndieGoGo campaign so we can hire some dedicated developers to make sure the task is done correctly and completely.
I'm looking forward hoping that Suse will 'pass' my evaluation and that I'll be soon part of the (old) community. ;)
All the best, Gour
Le 18/11/2014 12:09, Gour a écrit :
I also read that 'review', but didn't get that the solution is to 'just add the repo', so it does not sound bad, imho.
in fact, I guess the problem is that cut down of many multimedia apps exists for compatibility reason, and when adding Packman, this repo do not trigger a higher priority for it. so it's not sure that when one install the apps there are really dl from packman
so, the good way now should be to enable packman, give it some lower number as priorité, go to yast software, *repositories view* and call for software from there. A bit complicated
jdd
On Monday 17 November 2014 23:05:25 Roger Luedecke wrote:
It should be understood of course in regards to patented or licensed technologies, there is little we can do about that. On the other hand, the font rendering issue is being worked on actually and should be fixed in an update soon.
That's very interesting. Could you give us some spoilers? Also, patents are location-specific, are they not? Is it possible to do some mechanism to provide uncrippled version for locations where those patents are not enforced?
On Tuesday 18 November 2014 18:43:02 S. wrote:
I'm aware that the subpixel rendering used by the Infinality project (arguable the best font rendering available for Linux) is patent-encumbered, but in my experience Arch has fonts that are *almost* as good as Infinality, and I'm pretty sure they don't use anything patent-encumbered. In my experience, the DejaVu Sans fonts with "slight" hinting and RGBA antialiasing look pretty nice, even with the default openSUSE fontconfig. So that's an example where considerable improvement could be made in the default openSUSE settings, even within the confines of purely open source solutions.
If I understand correctly, RGBA antialiasing *is* patented technology we are taking about. Hinting patents already expired, so grayscale+hinting can be used freely everywhere.
-
Andy anditosan -
Bruno Friedmann -
Gour -
Greg Freemyer -
Henne Vogelsang -
Ish Sookun -
jdd -
jdd -
Lars Müller -
Mathias Homann -
Mustafa Muhammad -
Roger Luedecke -
S. -
Stanislav Baiduzhyi -
Tomas Cech