Support NeoGAF

[flowsnake]

I did.. and when I hit a point where I had to work to get fonts going - I just said "fuck it."

Maybe there's a greater experience to be gained by using Arch, but I just don't see it. Same goes for Debian or Gentoo, although I'd imagine in terms of ease of use it would be Debian > Arch > Gentoo.

So if something seems to take all this effort, the underlying question is "what's the point?"

My personal answer is A) it isn't that much harder and B) it's more than worth it to avoid certain details of something like Ubuntu. I won't go into what those things are because they're 100% personal preference and probably won't apply to you. Not everyone is going to like every distro, that's why so many exist.

I think my perspective might change if I was using it on a different computer, because I mostly use linux on a crappy old laptop and not my primary desktop (although I did use the laptop as my main computer for a couple of years).

If you're going to try another distro, maybe another one more similarly targeting user-friendliness would be better. Debian is more of a server distro and Arch and Gentoo are for more tinkering-minded folk I would say.

 

[zoku88]

The true reason to go to Gentoo

I just can't bring myself to leave Ubuntu.. I tried out ElementaryOS but wasn't thrilled with it. I tried to install Arch a while back but there were loads of issues with it. Then there's Debian, and I'm honestly not too sure of what advantage there would be to going with Debian over Ubuntu.

Those of you on Arch - how long does it take you to get things to a point where you're happy and can do everything with your setup? Hours-wise.. or days.... I couldn't even get fonts looking the way I wanted to with Arch, and I'm just not sure of why people opt for it (and it has gained a ton of popularity in the last 2 years or so).

Idk, I'm not sure what you mean by 'fonts the way you wanted to'. Are you talking about things like sub pixel rendering?

I pretty much just installed google-webfonts from the AUR, set my default terminal font (inconsolata) and was done.

 

[Massa]

I think vertical integration serves more as a lock-in than anything else...

I think the main point of GNOME OS is to decouple applications from the system, so for example if you need the new features in LibreOffice 4.0 you don't have to worry about the new kernel or Xorg not working well with your system.

Linux distributions pioneered the idea of a centralized repository for distribution decades ago, but now that literally everything has caught up it's tough to ignore the multitude of problems the old approach has. It's about time someone fixes it.

 

[zoku88]

I think the main point of GNOME OS is to decouple applications from the system, so for example if you need the new features in LibreOffice 4.0 you don't have to worry about the new kernel or Xorg not working well with your system.

Linux distributions pioneered the idea of a centralized repository for distribution decades ago, but now that literally everything has caught up it's tough to ignore the multitude of problems the old approach has. It's about time someone fixes it.

But then it replaces the coupling with another altogether. Instead of applications being coupled with the system (although, meh, I don't think this coupling exists for the majority of applications), they're replacing it with tighter coupling with GNOME.

I think that's worse.

 

[Massa]

But then it replaces the coupling with another altogether. Instead of applications being coupled with the system (although, meh, I don't think this coupling exists for the majority of applications), they're replacing it with tighter coupling with GNOME.

It certainly does. If you're an application developer the simple act of distributing a new version to your Linux users is a nightmare. In the end people just put out the source code and hope they find their way eventually, which takes months and even years. It's incredibly frustrating to constantly get bug reports for things you fixed long ago.

I don't see a tight coupling with GNOME of anything that is currently independent of it. On the contrary, the idea is to make GNOME's use of the underlying technologies less abstract. For example it will focus on working well with systemd and Wayland, instead of putting something out there and saying "well, hopefully GNOME 3.8 will work with whatever your distribution of choice packs under the hood". The result is that they can improve the user experience significantly by making these choices, as well as the developer experience. A third party developer wants a stable and well defined system to target.

 

[zoku88]

It certainly does. If you're an application developer the simple act of distributing a new version to your Linux users is a nightmare. In the end people just put out the source code and hope they find their way eventually, which takes months and even years. It's incredibly frustrating to constantly get bug reports for things you fixed long ago.

I don't see a tight coupling with GNOME of anything that is currently independent of it. On the contrary, the idea is to make GNOME's use of the underlying technologies less abstract. For example it will focus on working well with systemd and Wayland, instead of putting something out there and saying "well, hopefully GNOME 3.8 will work with whatever your distribution of choice packs under the hood". The result is that they can improve the user experience significantly by making these choices, as well as the developer experience. A third party developer wants a stable and well defined system to target.

I'm not really sure what you mean by "system", then. I assumed you meant lower level system stuff (like systemd and Wayland, as you said).

Anyway, I don't think most third-party developers want to target GNOME. (I mean, there's nothing really stopping them from doing it now,) if for nothing else but the fact that the dependency tree is so big. (if there's a problem with your application, is it a big in your code? Or maybe is it somewhere in that big dependency chain you have?) That, and it also means that your upgrade cycle is related to the GNOME upgrade cycle.

And even if the dependency tree weren't a problem at all, unless you want your application to only run on GNOME systems, you'd have to deal with users having annoyances with run-time dependencies (like, for example, unlocking gnome-keyring. If they don't use GDM, for example.)

When I said 'lock-in', I somewhat meant stuff like the latter.

 

[lmpaler]

Can you all recommend a different distro that is still pretty newbie friendly? It could just very well be the AMD Mobility card of my laptop, but Ubuntu 12.10 works flawlessly and then sometimes it won't boot in which I have to press ESC and let it fix the errors in the drive.

Which makes little sense to me how the drive can be failing when I have Win 7 on this laptop as well and have had no problems whatsoever with my drive. I am loving the look and layout of Ubuntu, but the issues are bothersome. I am not giving up though! I will get Linux working and make it my primarily OS on my laptop.

 

[Massa]

I'm not really sure what you mean by "system", then. I assumed you meant lower level system stuff (like systemd and Wayland, as you said).

Anyway, I don't think most third-party developers want to target GNOME. (I mean, there's nothing really stopping them from doing it now,) if for nothing else but the fact that the dependency tree is so big. (if there's a problem with your application, is it a big in your code? Or maybe is it somewhere in that big dependency chain you have?) That, and it also means that your upgrade cycle is related to the GNOME upgrade cycle.

And even if the dependency tree weren't a problem at all, unless you want your application to only run on GNOME systems, you'd have to deal with users having annoyances with run-time dependencies (like, for example, unlocking gnome-keyring. If they don't use GDM, for example.)

When I said 'lock-in', I somewhat meant stuff like the latter.

In general dependencies are not a problem for app developers. It's always better to find a bug in gnome-keyring than having to write your own code to store keys, you'll end up with a lot more work and more bugs. Keys is a good example actually, because apps like firefox and pidgin used to (or maybe still do) store passwords in plain text in your home dir, which is awful for security. gnome-keyring, on the other hand, had a lot of work put into it to make it secure. As an app developer I just want to say 'this is a sensitive key, store it securily'. I'd rather worry about the details of my actual app than things like that.

Things can be problematic when the developers of the platform try to cater to everyone and build a lot of abstraction layers. KDE 4 was notoriously bad for this, when they wrote abstraction layers on top of abstraction layers. For example it's much easier to debug a problem if you're using gstreamer to play sounds in your app than if you're using phonon+gstreamer or phonon+xine or whatever else they supported. Then bugs may happen only under circumstances and things are much tougher to debug.

To go back to GNOME OS, one of the goals is to define clearly what's in the system and what isn't. So for example say they define gnome-platform-1.0, and it includes certain versions of GTK+, gstreamer and tracker, but not libofthemonth. Basically only the stable and tested versions of core libraries. With the tools they're writing now you'll just build an app bundle that will include non-standard libraries with it. Then any user of any system that implements the gnome-platform-1.0 interface will be able to just drag and drop that bundle and it should just work. The idea isn't to restrict users or developers, it's simply to bless a standard set of libraries and provide an easy way for developers to distribute their software, and users to actually install them.

 

[zoku88]

In general dependencies are not a problem for app developers. It's always better to find a bug in gnome-keyring than having to write your own code to store keys, you'll end up with a lot more work and more bugs. Keys is a good example actually, because apps like firefox and pidgin used to (or maybe still do) store passwords in plain text in your home dir, which is awful for security. gnome-keyring, on the other hand, had a lot of work put into it to make it secure. As an app developer I just want to say 'this is a sensitive key, store it securily'. I'd rather worry about the details of my actual app than things like that.

When I was bringing up gnome-keyring, I wasn't suggesting that people just write their own key-storing thing. I was just giving an example of a GNOME specific dependency. I don't see the reason why you whould want a keyring which is, unless users know what they're really doing, would only be normally unlocked if they used GDM. That's why I call it a 'lock-in', because it's somewhat annoying to use if you're not a GNOME/GDM user.

For the dependency thing, I was mostly talking about the weird graph for gnome. I mean, for example, tracing back the dependencies, empathy 3.6 seems to rely on gnome-desktop-3.6 , which I found odd. I'm sure there's a clear line to trace (I didn't really care to find out why.) I was trying to say that it might be hard to debug problems due to the amount of non-direct dependencies you have, depending on the problem. Then again, maybe you could just find which direct dependency is not working, and then ask upstream to deal with it, even if the bug isn't in their code.

EDIT: I'm explaining myself poorly. I'm basically trying to say that having non-obvious, unintuitive dependencies can possibly make debugging harder.

Things can be problematic when the developers of the platform try to cater to everyone and build a lot of abstraction layers. KDE 4 was notoriously bad for this, when they wrote abstraction layers on top of abstraction layers. For example it's much easier to debug a problem if you're using gstreamer to play sounds in your app than if you're using phonon+gstreamer or phonon+xine or whatever else they supported. Then bugs may happen only under circumstances and things are much tougher to debug.

This is kind of what I'm talking about, too. Why do we need phonon/gstreamer (which I assume is KDE specific?) Why do we need these DE specific backend stuff?

To go back to GNOME OS, one of the goals is to define clearly what's in the system and what isn't. So for example say they define gnome-platform-1.0, and it includes certain versions of GTK+, gstreamer and tracker, but not libofthemonth. Basically only the stable and tested versions of core libraries. With the tools they're writing now you'll just build an app bundle that will include non-standard libraries with it. Then any user of any system that implements the gnome-platform-1.0 interface will be able to just drag and drop that bundle and it should just work. The idea isn't to restrict users or developers, it's simply to bless a standard set of libraries and provide an easy way for developers to distribute their software, and users to actually install them.

But I don't think the end result will be easy for most non-GNOME users. I think the end result (if this takes off) will be in a lot of code that is not-quite-so portable.

Ubuntu specific things somewhat give me the same feelings (since it feels like people using Ubuntu as a development platform is also rising.)

 

[Vic]

The true reason to go to Gentoo

I just can't bring myself to leave Ubuntu.. I tried out ElementaryOS but wasn't thrilled with it. I tried to install Arch a while back but there were loads of issues with it. Then there's Debian, and I'm honestly not too sure of what advantage there would be to going with Debian over Ubuntu.

Those of you on Arch - how long does it take you to get things to a point where you're happy and can do everything with your setup? Hours-wise.. or days.... I couldn't even get fonts looking the way I wanted to with Arch, and I'm just not sure of why people opt for it (and it has gained a ton of popularity in the last 2 years or so).

Once you're quite familiar with Arch, you can get things the way you want it within a day.

 

[Massa]

When I was bringing up gnome-keyring, I wasn't suggesting that people just write their own key-storing thing. I was just giving an example of a GNOME specific dependency. I don't see the reason why you whould want a keyring which is, unless users know what they're really doing, would only be normally unlocked if they used GDM. That's why I call it a 'lock-in', because it's somewhat annoying to use if you're not a GNOME/GDM user.

You can use gnome-keyring with other DM's, I believe Ubuntu which uses LightDM also uses gnome-keyring (unless they already forked that too). You just need a good enough DM so that it will unlock the keyring at the same time as it unlocks your session.

But to your actual point, most GNOME-specific libraries have been killed in the past few years, they have made a concentrated effort to provide everything for application developers in GTK+, glib and a few others. The GNOME platform is significantly simpler now than it was just a couple years ago.

For the dependency thing, I was mostly talking about the weird graph for gnome. I mean, for example, tracing back the dependencies, empathy 3.6 seems to rely on gnome-desktop-3.6 , which I found odd. I'm sure there's a clear line to trace (I didn't really care to find out why.) I was trying to say that it might be hard to debug problems due to the amount of non-direct dependencies you have, depending on the problem. Then again, maybe you could just find which direct dependency is not working, and then ask upstream to deal with it, even if the bug isn't in their code.

There are a few specific cases like Empathy and Epiphany where the apps are designed specifically to work under GNOME. This is not the platform that they present for third party apps though, these are core apps that define what the GNOME core experience is like. For example, gnome-desktop (terribly misleading name) is just a very unstable library that offers neither API nor ABI stability between GNOME versions, it's used for functions that are not yet ready to be promoted to a stable platform.

This is kind of what I'm talking about, too. Why do we need phonon/gstreamer (which I assume is KDE specific?) Why do we need these DE specific backend stuff?

Well, the goal of these desktop environments is to provide a nice and integrated experience for users and developers alike. They both go about it in very different ways though.

But I don't think the end result will be easy for most non-GNOME users. I think the end result (if this takes off) will be in a lot of code that is not-quite-so portable.

Ubuntu specific things somewhat give me the same feelings (since it feels like people using Ubuntu as a development platform is also rising.)

There's a pretty big distinction there: GNOME is and will continue to be an open platform. Ubuntu makes it specially hard to integrate the technologies they use into other open source operating systems. Even Unity, which in theory is supposed to be a simple thing, can be quite a pain to integrate elsewhere. I believe Debian, which must support over 100 different environments these days, still can't integrate it.

When looking at GNOME OS you should look at their goals. If you're the kind of user that runs a rolling distro and has no problem compiling every program you use from a terminal command then you're probably not the target. If you dislike things like udev or dbus you're definitely not the target. But if you just want to be able to download Libreoffice when a new version comes out and install it with a few clicks, have it auto-update, integrate with the MIME system, provide launchers and things like that, then that kind of integration and ease of use is what the goal of GNOME OS is.

 

[zoku88]

Apparently, a new version of kerbel space program was launched with Linux support.
http://kerbaldevteam.tumblr.com/post/45540856025/0-19-out-you-betcha

I'll probably get it when I get home in an hour.


EDIT: Lol, the site is totally hammered.

 

[zoku88]

You can use gnome-keyring with other DM's, I believe Ubuntu which uses LightDM also uses gnome-keyring (unless they already forked that too). You just need a good enough DM so that it will unlock the keyring at the same time as it unlocks your session.

But to your actual point, most GNOME-specific libraries have been killed in the past few years, they have made a concentrated effort to provide everything for application developers in GTK+, glib and a few others. The GNOME platform is significantly simpler now than it was just a couple years ago.

That isnt actually default behavior. Ubuntu probably has some stuff added to a session wrapper script that they have lightdm call in order for this to happen. Or, it's related to one of their ubuntu specific patches. I actually use lightdm. It's not really a factor of a DM being good enough. I mean, should they be expected find every different keyring installed on the system (like, what if someone had KDE and GNOME keyrings.)

Perhaps if there was some standard directory for keyings desktop files to be, I guess. Then when you installed the keyring, the desktop file (which would have an exec statement which would have necessary arguments in it) would be in this directory. Then the only things DMs would have to do is do a foreach on that directory and unlock them.

https://wiki.archlinux.org/index.php/GNOME_Keyring#Use_Without_GNOME
This is what I found from the arch wiki anyway. Not something a normal user would know. Then again, maybe a normal user wouldn't install an application that depended on gnome-keyring without actually running GNOME, either.

Actually looking at the platform as defined in the link, most of those seem pretty standard. Not really what I was thinking when talking about vertical integration.

There are a few specific cases like Empathy and Epiphany where the apps are designed specifically to work under GNOME. This is not the platform that they present for third party apps though, these are core apps that define what the GNOME core experience is like. For example, gnome-desktop (terribly misleading name) is just a very unstable library that offers neither API nor ABI stability between GNOME versions, it's used for functions that are not yet ready to be promoted to a stable platform.

Did not actually know that. I'm actually using empathy 2.x right now, since 3.x is when they switched to using gnome-desktop. I had always assumed that that was, idk, something more desktopy (something that gnome-shell runs on? IDK.)

Well, the goal of these desktop environments is to provide a nice and integrated experience for users and developers alike. They both go about it in very different ways though.



There's a pretty big distinction there: GNOME is and will continue to be an open platform. Ubuntu makes it specially hard to integrate the technologies they use into other open source operating systems. Even Unity, which in theory is supposed to be a simple thing, can be quite a pain to integrate elsewhere. I believe Debian, which must support over 100 different environments these days, still can't integrate it.

Yea, I just worried about some other programs being designed specifically with GNOME in mind. It's something I've heard people complain about w.r.t. udisks2, for example. Though, since I'm not actually a user of that, I have no idea what the specific complaints are.

I guess it's not really the same as Ubuntu. Maybe something like Chrome OS, where they get support for things like netflix, but that doesn't trickle down to the rest of us? IDK.

When looking at GNOME OS you should look at their goals. If you're the kind of user that runs a rolling distro and has no problem compiling every program you use from a terminal command then you're probably not the target. If you dislike things like udev or dbus you're definitely not the target. But if you just want to be able to download Libreoffice when a new version comes out and install it with a few clicks, have it auto-update, integrate with the MIME system, provide launchers and things like that, then that kind of integration and ease of use is what the goal of GNOME OS is.

Perhaps.

One a sidenote, I really like dbus, but I don't think it goes far enough.

From what I know, you communicate with components specifically by name. Like, you send a message to NetworkManager or to Pidgin.

For the latter case, I think it should be much more general. There should be some sort of standard IM interface that IM-clients that use dbus are expected to implement. Instead of saying @pidgin write-message to[email protected] message:"blah"

It should be

@IM-clients write-message to[email protected] message:"blah" and any running IM-client could receive the message.

Something thing for like, a network manager or whatnot.

 

[Massa]

I guess it's not really the same as Ubuntu. Maybe something like Chrome OS, where they get support for things like netflix, but that doesn't trickle down to the rest of us? IDK.

Not really, there's nothing specifically different about GNOME OS than what you get on your <insert distro name> already. Not to say someone couldn't use Gnome as a base for a proprietary system like ChromeOS, Google is using Gentoo as their base right now.

Perhaps.

One a sidenote, I really like dbus, but I don't think it goes far enough.

From what I know, you communicate with components specifically by name. Like, you send a message to NetworkManager or to Pidgin.

For the latter case, I think it should be much more general. There should be some sort of standard IM interface that IM-clients that use dbus are expected to implement. Instead of saying @pidgin write-message to[email protected] message:"blah"

It should be

@IM-clients write-message to[email protected] message:"blah" and any running IM-client could receive the message.

Something thing for like, a network manager or whatnot.

dbus works pretty much like that. For example, recently the Gnome settings panels were ported to use "systemd" for setting the hostname and date (which used to be a pain since every OS and distro used a different method). Except they don't actually use systemd, they used a set of generic dbus interfaces that were implemented by systemd, but could also easily be implemented differently (I believe someone already has a systemd independent implementation working for Debian).

Other things also define generic dbus interfaces. The problem in your example is simply a matter of different IM app developer getting together to define them, so if they were motivated to do something like that it could easily be done.

 

[zoku88]

Not really, there's nothing specifically different about GNOME OS than what you get on your <insert distro name> already. Not to say someone couldn't use Gnome as a base for a proprietary system like ChromeOS, Google is using Gentoo as their base right now.

That is true. Speaking of ChromeOS, I was almost thinking of buying the new one and then completely wiping it and putting a "real" OS on it. Unfortunately, I pretty much depend on Super and I'm too lazy to remap.

dbus works pretty much like that. For example, recently the Gnome settings panels were ported to use "systemd" for setting the hostname and date (which used to be a pain since every OS and distro used a different method). Except they don't actually use systemd, they used a set of generic dbus interfaces that were implemented by systemd, but could also easily be implemented differently (I believe someone already has a systemd independent implementation working for Debian).

Other things also define generic dbus interfaces. The problem in your example is simply a matter of different IM app developer getting together to define them, so if they were motivated to do something like that it could easily be done.

Ah, I see. All of the examples I've seen were always point-to-point so I just assumed that dbus was always like that.

 

[Vanillalite]

On a side note if you want a light weight music player without the flash of a banshee, amarok, or rhythmbox DeadBeef works surprisingly well. I'd definitely recommend it.

 

[Elfforkusu]

On a side note if you want a light weight music player without the flash of a banshee, amarok, or rhythmbox DeadBeef works surprisingly well. I'd definitely recommend it.

Clementine.

It's a fork of Amarok from when Amarok was good.

 

[flowsnake]

Clementine.

It's a fork of Amarok from when Amarok was good.

Tried it. It's not what I would call "lightweight".
At least in comparison to DeaDBeeF.

 

[peakish]

Can you all recommend a different distro that is still pretty newbie friendly? It could just very well be the AMD Mobility card of my laptop, but Ubuntu 12.10 works flawlessly and then sometimes it won't boot in which I have to press ESC and let it fix the errors in the drive.

Which makes little sense to me how the drive can be failing when I have Win 7 on this laptop as well and have had no problems whatsoever with my drive. I am loving the look and layout of Ubuntu, but the issues are bothersome. I am not giving up though! I will get Linux working and make it my primarily OS on my laptop.

You could try 12.04 instead of 12.10. It'll lack some advancements for Unity (although I can't remember what they've done since) but should be more stable since it's a Long Term Support version. Linux Mint is recommended quite a bit as an alternative, but it's Ubuntu based and should run into similar problems.

I wouldn't discount hardware problems though, I'm by no means an expert but hard drives are fickle. No harm done in trying alternatives of course.

There's Fedora Core as well, looking at popular options. I don't think it's a great user experience since you'll have to mess around a bit to get mp3 and other closed codec support, but it is very nice as a good vanilla (Red Hat doesn't develop their own desktop spins like Canonical does, the standard Fedora is plain and beautiful Gnome 3) desktop environment experience. After I was fully set up (RPM Fusion covers codecs and stuff) I had a great time a couple of years back, but not everyone likes going over a speed bumb to get a great desktop experience. The new installer in 18 is stellar, too.

http://fedoraproject.org/
http://rpmfusion.org/

 

[qizah]

Someone released a Ubuntu Touch icon theme a while ago, I tried it out today, liking it a lot.

Is Unity going to get Ubuntu Touch icons in 13.04? Would be nice.

 

[ThatObviousUser]

Unity.

 

[lmpaler]

Threw up Fuduntu on my laptop and I must say, I am impressed with the speed of it. SImple and quick. I enjoy it. I came upon it on the net by reading some lists of the best distros, although there isn't a specific one apparently as it depends on the user, but Fuduntu apparently is very great on laptops and netbooks.

Now if only the mirrors would work so I can update everything and get WINE installed..

 

[Madtown_]

I never have really played SimCity in all that much detail. Last night, finally got it set up around midnight. Whew, today I think I'll be a bit tired. Thanks zoku for the commandline args to up the resolution.

 

[Vanillalite]

Has anyone tried out Krita yet?

Seems Krita is like an advanced version of MS Paint with some GIMP like features though not really meant to be a GIMP/Photoshop replacement.

 

[O.DOGG]

A couple of months ago at work my supervisor gave me a separate machine that had XP on it. I think it was infected with a virus because it was running out of memory and crashing my browser and Thunderbird pretty much every time I would start them at the same time. I didn't care much about it because I only work on weekends, so it's really not that big of a deal. Last weekend however after the fifth crash, I had enough and installed xubuntu. Runs like a charm. Linux, I've missed you so much!

 

[b0b]

i like gnome3. it's a nice interface, it's usable and it looks fine.



but the development is somehow stupid. why tying gnome3 to GDM? why tying it to systemd? Why load nearly the whole gnome by starting GDM? Why isn't it possible to disable some stuff I don't need - like "Universal Settings"? It's fine, sure - but I'm not disabled, so I don't need it. Gnome Extensions are very (VERY) stupid. Go to a website (still beta) and download something from somewhere from someone you don't know you can trust. Great idea, really... The next great idea Gnome people had are Gnome Apps. this is gona be gud...

gnome people go linux-only. they don't care about non-linux-systems. gtk and gnome go hand-in-hand and if gnome isn't used that much besides linux, gtk (3) as an framework won't get the same acceptance as KDE/QT ie. There is the GTK3 port for windows? how good is it? any GTK for OSX? I know of GTK 2.x OSX port and it's just horrible...



as for music player:

I'm using cmus. It's ncurses-interface without mouse-support, but it's damn fast. as second choice there is also exaile installed.

 

[Treechopper]

Soooooooooooo Brettison got me to distro hop.

I was on Ubuntu 12.10 but ran into plenty of bugs that were not letting me get work done so I jumped to Crunchbang. This particularly sucks since Andrex's glowing praise of Gnome shell made me try it and love it to death. Not the biggest fan of Crunchbang but it gets the job done. Problem is now I can't get the native Linux version of Spotify run. Downloads and updates fine but it won't launch. Any help?

 

[ThatObviousUser]

https://mail.gnome.org/archives/devel-announce-list/2013-March/msg00004.html

Gnome 3.8 release candidate is OUT.

Soooooooooooo Brettison got me to distro hop.

I was on Ubuntu 12.10 but ran into plenty of bugs that were not letting me get work done so I jumped to Crunchbang.

This particularly sucks since Andrex's glowing praise of Gnome shell made me try it and love it to death.

Problem is now I can't get the native Linux version of Spotify run. Downloads and updates fine but it won't launch. Any help?

As a temporary solution you can check out Spotify for Chrome.

 

[Polari]

Soooooooooooo Brettison got me to distro hop.

I was on Ubuntu 12.10 but ran into plenty of bugs that were not letting me get work done so I jumped to Crunchbang. This particularly sucks since Andrex's glowing praise of Gnome shell made me try it and love it to death. Not the biggest fan of Crunchbang but it gets the job done. Problem is now I can't get the native Linux version of Spotify run. Downloads and updates fine but it won't launch. Any help?

Try opening it in a terminal and then post the output here. Might just a missing library. Otherwise I used to use the Windows version under WINE (that was a couple of years ago though, no idea if it still works).

 

[Treechopper]

https://docs.google.com/file/d/0Bwh6whYLZYPjdE1aVVd0MlMtWHM/edit?usp=sharing

Mind you I'm still relatively new to Linux.

Also, Gnome Shell is awesome. For anybody lurking, get it.

 

[injurai]

yeah the only people that hate gnome are people that have used gnome 2 for years and or running on old hardware. But really they have option so it shouldn't be the end of the world. I'm still on cinnamon b/c im lazy and just use mint.

 

[Polari]

https://docs.google.com/file/d/0Bwh6whYLZYPjdE1aVVd0MlMtWHM/edit?usp=sharing

Mind you I'm still relatively new to Linux.

Also, Gnome Shell is awesome. For anybody lurking, get it.

Hey you might want to give this a try "apt-get install libqt4-network"

Edit: actually, Googling around it appears " libqtnetwork4-perl" is the package needed.

 

[Treechopper]

Yeah, I'm going to stop multitasking now. I could have fixed that on my own :/

*Edit*

Ok, yeah this might go a little deeper than I know how to deal with.

https://docs.google.com/file/d/0Bwh6whYLZYPjbzlRMzFONWFaNzA/edit?usp=sharing

I ran sudo apt-get install -f to fix broken packages but it didn't change anything.

 

[-KRS-]

Yeah, welcome to APT dependency hell. =/
It's probably because libqtnetwork4-perl depends on older libraries that are not compatible with your other installed applications. So if the older libraries were installed it would break other things. Therefore APT won't let you install them. I wish APT would actually say what's wrong though, instead of just saying that 'it depends on this but it wont be installed'.

I'm not sure what can be done except just to wait until crunchbang updates the libqtnetwork4-perl package. Well I guess you could try to manually install it by downloading the source code. But that's kind of a lot of work since you would probably need to also manually install everything it depends on as well. You could also try to google something like 'crunchbang libqtnetwork4-perl' to see if others have the same issue and perhaps has a solution.

Just to make sure, you did run apt-get update and apt-get upgrade before trying to install right?

 

[Treechopper]

Yeah, welcome to APT dependency hell. =/
It's probably because libqtnetwork4-perl depends on older libraries that are not compatible with your other installed applications. So if the older libraries were installed it would break other things. Therefore APT won't let you install them. I wish APT would actually say what's wrong though, instead of just saying that 'it depends on this but it wont be installed'.

I'm not sure what can be done except just to wait until crunchbang updates the libqtnetwork4-perl package. Well I guess you could try to manually install it by downloading the source code. But that's kind of a lot of work since you would probably need to also manually install everything it depends on as well. You could also try to google something like 'crunchbang libqtnetwork4-perl' to see if others have the same issue and perhaps has a solution.

Just to make sure, you did run apt-get update and apt-get upgrade before trying to install right?

Craaaaaap. At least I'm learning a lot.

I ran them a few days ago after I had downloaded spotify but not right before the screenshot.

 

[Xisiqomelir]

Canonical and Chinese standards body announce Ubuntu collaboration

- Chinese Ministry of Industry and Information Technology defines a new reference architecture for standard operating systems in China
- CSIP, Canonical and NUDT for the CCN Open Source Innovation Joint Lab to further that standard
- CCN Joint Lab will produce Ubuntu Kylin, with first release expected in conjunction with Ubuntu 13.04 in April 2013.

London and Beijing, 21 March 2013: The China Software and Integrated Chip Promotions Centre (CSIP) today announced a new reference architecture for Chinese operating systems. CSIP, which is a part of the Chinese Ministry of Industry and Information Technology (MII), has selected Canonical’s Ubuntu as the basis for that reference architecture in order to provide a flexible, open, widely-used and standardised operating system. The announcement is part of the Chinese government’s five year plan to promote open source software and accelerate the growth of the open source ecosystem within China.

Additionally, CSIP, Canonical and the National University of Defense Technology (NUDT) have formed the CCN Open Source Innovation Joint Lab in Beijing. The Joint Lab hosts engineers from each organisation who will accelerate the development of a China-focused version of Ubuntu for desktop and cloud.

"CSIP, Canonical and NUDT share the goal of widespread adoption of high quality Linux on personal devices and in the cloud," said Jane Silber, CEO of Canonical. "This collaboration will bring local investment and participation to ensure that the platform is relevant for the Chinese market, and close coordination with the global Ubuntu project ensures that it is familiar to software and hardware vendors, and useful for export products made by Chinese companies as well."

The initial work of the CCN Joint Lab is focused on the development of an enhanced version of the Ubuntu desktop with features specific to the Chinese market. The new version is called Ubuntu Kylin and the first version will be released in April 2013 in conjunction with Ubuntu’s global release schedule. Future work will extend beyond the desktop to other platforms.

Ubuntu Kylin goes beyond language localisation and includes features and applications that cater for the Chinese market. In the 13.04 release, Chinese input methods and Chinese calendars are supported, there is a new weather indicator, and users can quickly search across the most popular Chinese music services from the Dash. Future releases will include integration with Baidu maps and leading shopping service Taobao, payment processing for Chinese banks, and real-time train and flight information. The Ubuntu Kylin team is cooperating with WPS, the most popular office suite in China, and is creating photo editing and system management tools which could be incorporated into other flavours of Ubuntu worldwide.

“The release of Ubuntu Kylin brings the Chinese open source community into the global Ubuntu community,” said Mark Shuttleworth, founder of Ubuntu and Canonical. “With Ubuntu Kylin, China now has its own secure and stable desktop operating system, produced alongside Ubuntu's global community. Ubuntu combines proven technology with a mature ecosystem and strong OEM and ISV partners, and this initiative allows the Joint Lab to bring those strengths to China across the full range of platforms: desktop, server, cloud, tablet and phone.”

In before "Communism"

 

[Slavik81]

Yeah, welcome to APT dependency hell. =/
It's probably because libqtnetwork4-perl depends on older libraries that are not compatible with your other installed applications. So if the older libraries were installed it would break other things. Therefore APT won't let you install them. I wish APT would actually say what's wrong though, instead of just saying that 'it depends on this but it wont be installed'.

Why not just install versioned libraries? Didn't Windows 98 fix this problem with WinSxS?

 

[lmpaler]

After a few days of testing, Fuduntu kept crashing so I burned Ubuntu to a DVD and did a clean install and everything runs fine, which is odd. So I guess I will stick with this for now and eventually try a few different distros when I get more acquainted with Linux(I have Mint 14 on DVD as well if I am feeling froggy)

Now for the real fun:installing those old Inifity Engine games with mods and getting used to wine and such.

 

[-KRS-]

Why not just install versioned libraries? Didn't Windows 98 fix this problem with WinSxS?

Much of the point of package managers and one of the ideas in linux in general is that you don't have several versions of a library installed and taking up unnecessary space unless you absolutely have to. Everything should be compiled to use the same library versions etc. In general this shouldn't be an issue for the user. Only for the people who packages all the applications and libraries. So say they upgrade libjpeg to a new version and that somehow breaks everything that uses it, it's the job of the package maintainers to recompile and repackage everything that depends on it to use the new version, so that the end user doesn't have to care. But for some reason crunchbang seems to have fucked this specific package up. Or something is wrong with his APT install which is also possible. APT can be somewhat of a mess sometimes.

It makes sense for Windows though, because most things in Windows are not open source. So you can't just download the source code and recompile it to use a newer version of a library. So the only way to support everything is to have every version of a library installed or depend on statically compiled libraries shipped with the application.

 

[PandaL]

Can you all recommend a different distro that is still pretty newbie friendly? It could just very well be the AMD Mobility card of my laptop, but Ubuntu 12.10 works flawlessly and then sometimes it won't boot in which I have to press ESC and let it fix the errors in the drive.

Which makes little sense to me how the drive can be failing when I have Win 7 on this laptop as well and have had no problems whatsoever with my drive. I am loving the look and layout of Ubuntu, but the issues are bothersome. I am not giving up though! I will get Linux working and make it my primarily OS on my laptop.

Xubuntu 13.04. Its gonna be released on April 25th. I assure you that you won't be disappointed. But you should be ready to upgrade it every 9months since canonical reduced the support for normal releases to 9months from 18 months.

If you're not ready to upgrade every 6 or 9 months, then Xubuntu 12.04 is the way to go.

Xubuntu 12.04 reviews:

• http://www.dedoimedo.com/computers/xubuntu-pangolin.html
• http://www.techdrivein.com/2012/07/xubuntu-1204-lts-review-perfect.html

And I strongly recommend you to stay away from unity.


PS : writing this post from Xubuntu 13.04 beta and its stable & faster than 12.10

 

[O.DOGG]

Xubuntu 13.04. Its gonna be released on 25th. I assure you that you won't be disappointed. But you should be ready to upgrade it every 9months since canonical reduced the support for normal releases to 9months from 18 months.

If you're not ready to upgrade every 6 or 9 months, then Xubuntu 12.04 is the way to go.

Xubuntu 12.04 reviews:

• http://www.dedoimedo.com/computers/xubuntu-pangolin.html
• http://www.techdrivein.com/2012/07/xubuntu-1204-lts-review-perfect.html

And I strongly recommend you to stay away from unity.


PS : writing this post from Xubuntu 13.04 beta and its stable & faster than 12.10

Awesome! Looks like I'll be upgrading next weekend. Can it be autoupdated from within the OS (xubuntu 12.10) or do I have to reinstall it from a boot drive?

 

[injurai]

can someone explain the concept of packages to me? I really am a linux noob. I know hardly anything about the OS, I just use it for software developement and learning.

 

[Massa]

can someone explain the concept of packages to me? I really am a linux noob. I know hardly anything about the OS, I just use it for software developement and learning.

Basically a package is a binary version of a program, ready to install and use.

Linux distributions are made up of hundreds or thousands of different pieces, that can range from simple programs like a calculator to core pieces of the system like the kernel. Each different module does a different job, and when combined together you end up with a fully functioning system.

The authors of each one of these programs release them in the source code form, which is what I linked above for the kernel and a calculator. So someone takes that source code, builds a package containing binaries and that's what you install and run on your system.

 

[nan0]

can someone explain the concept of packages to me? I really am a linux noob. I know hardly anything about the OS, I just use it for software developement and learning.

What Massa said plus: Packages can contain source code as well, usually indicated by the suffixes -src or -dev. Also, packages are not neccessarily executable programs, they can contain libraries (equivalent to Windows *.dll files) as well, indicated in the package name by "libwhatever".

 

[ComputerMKII]

can someone explain the concept of packages to me? I really am a linux noob. I know hardly anything about the OS, I just use it for software developement and learning.

All programs and their plug-ins are packages. Some packages can't work if some other packages aren't installed first. There are also meta-packages which sole purpose is to install a bunch of other packages at once.

 

[PandaL]

Awesome! Looks like I'll be upgrading next weekend. Can it be autoupdated from within the OS (xubuntu 12.10) or do I have to reinstall it from a boot drive?

You can directly upgrade from 12.10. The actual release date is "April 25".

Same steps as explained in this link.

http://www.omgubuntu.co.uk/2012/10/how-to-upgrade-ubuntu-12-04-to-ubuntu-12-10

 

[Vanillalite]

So Geary despite being an e-mail client all though young already in use by a decent amount of people looking for a evolution/thunderbird alternative... is now trying to really be made?

IDK... it's being crowd source funded now or at least trying...

IndieGoGo: Support Geary Page!

PS: The video while simple did address the one thing most of my friends and I all have. Why we even need a desktop client when we aren't at work. Everyone just uses Gmail in their browser or an app on their phone/tablet.

 

[peakish]

So Geary despite being an e-mail client all though young already in use by a decent amount of people looking for a evolution/thunderbird alternative... is now trying to really be made?

IDK... it's being crowd source funded now or at least trying...

IndieGoGo: Support Geary Page!

Getting funds to focus on development can only be a good thing, even if it's reached it's current state without this. Even though Geary is far too uneven for me to use it atm (prior recommendations aside I've switched to Thunderbird, which I'm not super happy with), it's definitely one of the OSS projects I'm most interested in. I'll probably drop some money to it later next month.

 

[ComputerMKII]

What's the point in using a dedicated e-mail client when you could just log on Gmail? Seriously.

 

[peakish]

What's the point in using a dedicated e-mail client when you could just log on Gmail? Seriously.

Easier switching between different accounts, native notifications, offline archives. Not having to keep a tab dedicated to a service that shouldn't need it.