State of Linux and the Desktop
- Login o registrati per inviare commenti
Warning: Long post--READ AT YOUR OWN RISK! ;)
I have been using Linux since 2005. I have gone through, literally, at least 3/4 of the linux distributions that currently exist today(as can be found on distrowatch).
Everywhere I go today, people are having the same complaints: devices not working, network problems, gui bugs, etc.--so many issues that at this point should be ironed out by the Linux community. I recently tried Fedora 16...it would not detect my Atheros wireless card. Even the three BSDs can detect(and use) my wireless card at install. I then looked at using OpenSUSE(one of the more professional grade distros for me in the past) and was reading the comments here: http://news.opensuse.org/2011/11/16/opensuse-12-1-all-green/.
I have used Windows 7 and have found it an absolute delight! I love opensource for the ideals and world-changing force it brings(and yes, F/OSS is slowling changing many aspects of life--not just the IT field). ~2-3 years ago, Mark Shuttleworth had declared that Linux had surpassed Windows and was approaching the Mac. Lately, however, it seems that all the distros and desktops have taken several steps back in usability and functionality. Gnome 3 is counter-intuitive, KDE(which I have loved) is a true resource-hog, XFCE just doesn't have that polished feel to it, LXDE(massive potential there)seems like it just needs more people tweaking it out and Enlightenment just never really seemed to catch on for the most part(hopefully Bodhi will help change that).
Now all of these desktops are part of the problem. You have a massive duplication of effort going on because everyone is working on these different desktops to do the same basic tasks. That is wasted effort. I respected Canonical back when Gnome was the only officially supported desktop because they were working on being a mainstream OS for the everyday user and wanted to put there focus on making ONE excellent version of the OS. They just dropped Kubuntu, but have moved to Unity which on my computer is inexplicably slow.
I bring this up here because as the most polished of the "entirely-free" linux distros, AND being officially supported and endorsed by the FSF--it would be a tremendous statement to pick a desktop as the "official" desktop for this distro and tweak/bugfix to make this as polished and user-friendly as it can possibly be. I love what this community has done with gnome 2. Trisquel is by far my favorite Gnome 2 based distro to date. Instead of having a Gnome version and an LXDE version, however, why not go for the lowest common denominator(LXDE) and build upward? The time that would be spent duplicating effort on two separate desktops can be put into other areas(features, bug-fixing, language support, etc.).
It has been made clear that the hope is to generate enough donations to support developers allowing them to work on this project full-time. This is just me, but I personally would love to see more fine-tuning of the desktop and applications itself. For instance, I play a lot of games and it irks me to no end when I try to play a Japanese game and I have to go to the command line to open Wine or Gedit in Japanese just to edit a file or play the game. In 2012, these applications should be able to determine the language automatically and "just work"--provided the language support is installed. Even using the unzip program on Japanese zipped files produces unusable characters that only cause errors. I have download 7-Zip for windows, install it using Wine via the Japanese locale, open 7-Zip in the Japanese locale, THEN unzip the package. As you can see, that is just "way" too many steps for something that should be effortless.
In closing, I love linux. I love what you all have done with Trisquel. I would love to see this community "lead-by-example" and do something that all of the other linux distros are failing miserably at doing: producing a solid, stable, user-friendly distro where everything "just works". Ubuntu was on the right path...but they made a HUGE left-turn. Let's show everyone that the linux desktop can be truly great! Thank you.
IMHO, we need more centralized efforts. Right now we have about 500 distros and most of them are exactly the same with a different feel, this makes new users confuse. Give a link to DistroWatch for a casual proprietary software user and you will see what I'm talking about. I was a distro hopper for a year and a half before settling to Trisquel. I have used at least 50 of DistroWatch's top 100. The problem is, different package manager needs the same porting again. So you have apt/yum/pacman/conary/pisi wars and all of them need different packagers for the same package. No repository right now is as big as Debian's. Another issue is porting to different architectures. Each distro (including a simple remaster) would need more people and more redundant effort to be ported to another architecture (PPC, SPARC, etc). In this sense I prefer to support Debian because it is "universal". It is the only distribution easy to use which has support for more than 6 different processor architectures and different kernels (kfreebsd and hurd). Debian fits on everything. Can be a server, rolling desktop, stable desktop, for Free Software enthusiasts like us (just disable the non-free repository and refuse to load the proprietary modules), serves for old computers, source based with build-dep, etc.
About the desktop environments, I am currently watching the Elementary Shell. They are using Vala which is more high level than C and has the same performance. Way easier to mantain than C (the language of all the other DE's, KDE uses C++ but is not the same thing). Their code is also cleaner and just "does the job".
On 02/11/2012 06:12 PM, name at domain wrote:
> for Free Software enthusiasts like us (just disable the non-free
> repository and refuse to load the proprietary modules), serves for old
> computers, source based with build-dep, etc.
Trisquel also does all of that, no? Without having to disable anything.
:-) I've been using Debian, Ubuntu, and Trisquel in the past 5 years,
and switch back and forth a lot. But I think that since Trisquel
'refuses' nonfree software, then it's a distribution that deserves our
efforts. The problem with accepting distros that include nonfree
software is that it prevents,
1) economically rewarding hardware manufacturers that release free
firmware/drivers
2) the development of free alternative drivers and applications when
there's currently none
The more you tolerate nonfree software, the more it creeps into your
life. I've seen folks who used to use a GNU/Linux distros with nonfree
software, then more nonfree software, then slowly switched to Mac OS X
(because they can use the terminal as well and benefit from many of the
same programs...). Since I've seen this happen, I'm convinced accepting
some nonfree software will sooner or later lead to the crumbling of the
(fragile) free software ecosystem as we know it now. I'm also convinced
that the reason we are struggling with so much nonfree firmware and
drivers these days is because of the complacency of a wide range of the
GNU/Linux users with them.
If you need a functionality/program that's only available as nonfree
software, code it yourself or find someone to do it for you. If you
don't know how to code, then learn! :-) We can't just expect or wait for
freedom, we have to build it! Making exceptions is making freedom the
exception.
Yes, I agree. Trisquel's main purpose is to create a free Ubuntu rather than merely an "easier Ubuntu" and no other distro has done that on an Ubuntu base. But you see Mint, Pinguy, AriOS, BigLinux, moonOS, Super OS, Zorin n' stuff are nothing but Ubuntu preinstalled with PPA's and proprietary software. They do the same thing with a different interface. Mint at least has some software developed in-house, the others are just plain Ubuntu with tweaks. I don't think there is such need for something like that. If you don't like Unity or Gnome Shell, tweak it yourself or use Kubuntu, Lubuntu, etc.
Another disappointing story is Sabayon. I have read the leader's biography and he said Sabayon came out of a bet he made with a friend that said nothing good could come out of Gentoo. What the heck?
And yet another reason is, some distribution can't keep up with the pace of the parent's distribution and are way several releases back (BLAG,Venenux) as an example). There are also distributions marked as "active" in DistroWatch with last release being in 2007 (TurboLinux as an example). So I always tend to go back to the roots and use an already established, long lived but most importantly, a regularly released distribution.
I'm all for improving the LXDE version of Trisquel. I have LXDE on an older
computer I setup for my nephew and I believe the DE has a lot of promise.
I would recommend users check out our
[https://trisquel.info/en/wiki/introduction-trisquel-gnu-linux Introduction
to Trisquel GNU/Linux] page of documentation. It explains that we are not
simply a "linux (sic) desktop" but in fact a combination of the
[https://trisquel.info/en/wiki/introduction-trisquel-gnu-linux linux kernel
and the GNU system]. I would also like to note that we are a free software
project not an
[https://www.gnu.org/philosophy/open-source-misses-the-point.html open source
project].
I agree with both of you--that is why I am bringing it up here. All these distros that, at the end of the day, are nothing more than cosmetic differences(functionally, they all do the same things)...are the biggest problem with the adoption/growth of linux.
Imagine if the developers half of the 318 active distros(according to distrowatch) all refocused on community run distro and solely worked on features and bug-fixing alone! That distro would fly up the ranks in no time because it's growth would be stupendous!
I also agree that accepting free software is also vitally important. I was going to go with OpenBSD, but I didn't have the time to learn how to start from scratch what with school and all. So I tried Trisquel and love it!
Looking at HaikuOS and AROS--it's confirmation that it is possible to create a lightweight, fully functional desktop that could take on the bloated ones that exist now. I love how tight this distro feels. If more devs could be pulled in, and attention focused on one desktop, imagine what this could become!
I respect Ubuntu, Fedora, OpenSUSE and Mint for what they have done, but it sometimes feels like they aren't as efficient as they could be with their resources. Maybe it's just because I don't see behind the scenes.
>Everywhere I go today, people are having the same complaints: devices not working, network problems, gui bugs, etc
Yeah, problems everywhere, GNU/Linux, *BSD, BeOS, Haiku, Windoze, Macs... you name it, complex systems tend to develop problems.
>You have a massive duplication of effort going on because everyone is working on these different desktops to do the same basic tasks. That is wasted effort.
So, one true distribution with one true desktop environment? Perhaps running on one true hardware? Sounds like a nightmare to me.
>do something that all of the other linux distros are failing miserably at doing: producing a solid, stable, user-friendly distro where everything "just works"
Trisquel is a small fish. Help us do it! :)
I am a translator!
One true distribution with one true desktop environment? You are aware, that no one could possibly force any software project to extinct in the free software world, right? As long as people like a desktop environment, they will develop for it.
The point is for a distro team to focus efforts on just one environment instead of developing a distro with many different environments and also possibly hopping from one to the other (like Ubuntu hopped from gnome to unity). There still may be different distros that focus on different environments though, nothing to say about that.
I am a translator!
I totally agree on the aspect, that we have too much distros producing same results, just in different colors. It's fine that we have a multitude of desktop environments, but there just should be a couple of distros, maybe one for each major desktop environment. It would concentrate the efforts and result in much higher quality software.
Trisquel already has a very good take on Gnome 2, one of the best I've seen. Not sure about Gnome 3, but maybe the efforts should be concentrated on that, instead of LXDE. As an option, I would really appreciate a DE-less version of Trisquel, something like a base install of Debian, where one has to install the Xserver, video drivers and the DE per hand. This way everyone would have the possibility to play around with any desktop and at the same time there would be one extremely stable and well supported desktop for use in a production environment.
-> As an option, I would really appreciate a DE-less version of Trisquel, something like a base install of Debian, where one has to install the Xserver, video drivers and the DE per hand.
But a netinstall version of Trisquel is already available in the download page.
I am a translator!
Sorry, totally forgot about it. I had issues the last time I tried to use it, so that's probably why I didn't remember the netinstall image.
I agree that fragmentation is not good for development but I actually
think that some of this fragmentation is good.
My life experiences tell me that humans have enormous ability to carry
out very technical tasks but very limited ability to co-operate,
especially over time and outside of war time efforts. When one community
becomes poisoned the code just moves to another community.
Screw open office and Oracle, I am very happy with Libre office \0/
I am a translator!
Sorry, totally forgot about it. I had issues the last time I tried to use it,
so that's probably why I didn't remember the netinstall image.
I would recommend users check out our Introduction to Trisquel GNU/Linux page of documentation. It explains that we are not simply a "linux (sic) desktop" but in fact a combination of the linux kernel and the GNU system. I would also like to note that we are a free software project not an open source project.
Trisquel meets all of the criteria laid out in the Open Source Definition (http://opensource.org/docs/osd), so it is, by definition, an open source project.
Of course, it's also a free software project, and we prefer to refer to it as such because it lets people know that the project is about freedom.
I would disagree. On paper yes all the software included in Trisquel is "open source" by the definition. However the difference between free software and open source isn't simply the diff (see what I did there) between the OSI definition and the FSF definition. It is a philosophy as well. Free software is about freedom while open source is not. Trisquel is about freedom which means its philosophy is that of free software not open source.
So yes Trisquel would meet the criteria of being an "open source project" but it is really a free software project based on philosophy.
That ambiguity between development and philosophy is what bring us to this kind of discussion.
I guess you just want to point that the term Free Software HAS to override the concept of OpenSource.
That ambiguity between development and philosophy is what bring us to this
kind of discussion.
I guess you just want to point that the term Free Software HAS to override
the concept of OpenSource.
I would disagree. On paper yes all the software included in Trisquel is
"open source" by the definition. However the difference between free
software and open source isn't simply the diff (see what I did there) between
the OSI definition and the FSF definition. It is a philosophy as well. Free
software is about freedom while open source is not. Trisquel is about
freedom which means its philosophy is that of free software not open source.
Trisquel meets all of the criteria laid out in the Open Source Definition
(http://opensource.org/docs/osd), so it is, by definition, an open source
project.
Of course, it's also a free software project, and we prefer to refer to it as
such because it lets people know that the project is about freedom.
I'm all for improving the LXDE version of Trisquel. I have LXDE on an older computer I setup for my nephew and I believe the DE has a lot of promise.
- Login o registrati per inviare commenti