Thoughts on Linux, Ubuntu, Debian, Slackware
Posted: 3. Aug 2014, 21:28
Today I was unfaithful and installed Ubuntu 14 on my HTPC.
I had OpenELEC installed on it, but I wanted to add some emulators and additional software, and OpenELEC was a little bit unflexible for that purpose.
I chose Ubuntu because my favourite emulator just got packed into a brand new Ubuntu package using the latest version. The Slackbuild version was remarkably older. Another argument was the trouble I had fallen into, trying to install XBMC on Salix RP some time ago. I tried to avoid all that struggle with Java not in %PATH for root, and was afraid to encounter similar problems with the emulator.
So I tried Ubuntu, and within 30 minutes, everything worked flawless.
Hmm...
After that experience I had the following thougts.
1) The most successful Linux distribustions seem to be **untu and Debian.
But why are they successfull?
a) Debian:
Has a rediculously bureaucratic organisation in the background and they are open source fanatics. Binary proprietary WLAN adapter firmware is not included by default, even if a user desperately needs it... Debian is popular for webservers and professional use.
Why? I came to the conclusion that you can compare the wy the distribution is made with some ISO 9001 processes in companies in real life. For decision makers this can be a good argument: A good and well-organized creation process of Debian leads to a good distribution in the end. The open source fanatism is also a plus for that purpose: The risk of being sued due to a violation of patents (because LAME is lying around somewhere on a webserver installation etc..) is minimized. Everything is as free as possible, and in the end Debian is first choice for professional use.
b) Ubuntu/**untu:
Derived from the professional distribution described above, with the goal of being as end-user friendly as possible. Due to the massive invest of a generous donor, this goal gets archived very well... Everything works as it should to the user, and there are lots of packages of consumer software for it. Big support communities online, thousands of books and manuals offline.
If you look at it that way, you'll recognize that there's not much room left for other distributions. Professional use: Debian, Ubuntu: End-User... If you add "commercially supported" distributions like Fedora, Open Suse or CentOS, the remaining need of some traditional community distributions like Slackware or Mandrake/Mageia is very limited.
Especially for Mageia I see dark prospects, because most of their goals are accomplished better by Ubuntu, which is supported by a lot of more packages.
And what about Slackware?
Slackware has some interesting features.
-simple package system
-Slackbuild concept
-configuration with config files
-?
I think it would be intersting to find out which kind of users use Slackware for what purpose. Did Slackware gain a numerable amount of new useres in the last 3 years? Are there usage scenarios where Slackware can display its special features?
For example, I wanted to learn a little bit Linux and already had some config file experience from MS DOS and Win 3.1 I like the ideas of having no graphical tools messing up my configuration. But I think this kind of audience is quite rare...
Could that be a purpose: Using Salix/Slackware to learn something about Linux, config files and KISS?
KISS is also a good cue: IMHO, if you look at the original Slackware philosophy, isn't it some kind of contradiction to install the latest KDE or Gnome desktop with all their graphical eye-candy on it? The more so as they often bring their graphical configuration tools with them...
So I think the ready-made Salix Ratpoison distribution was a good approach. Unfortunately ratpoison is a little bit limited for certain scenarios, especially if a single program comes up with many little windows. Maybe Salix/Slackware can develop a window manager/desktop aligned to the distribution's philosophy:
-KISS
-configuration files
-able to run many little windows and dialog boxes...
-no graphical config tools
And Salix/Slackware needs to define its audience, as stated above... And then focus on the needs of that users...
I had OpenELEC installed on it, but I wanted to add some emulators and additional software, and OpenELEC was a little bit unflexible for that purpose.
I chose Ubuntu because my favourite emulator just got packed into a brand new Ubuntu package using the latest version. The Slackbuild version was remarkably older. Another argument was the trouble I had fallen into, trying to install XBMC on Salix RP some time ago. I tried to avoid all that struggle with Java not in %PATH for root, and was afraid to encounter similar problems with the emulator.
So I tried Ubuntu, and within 30 minutes, everything worked flawless.
Hmm...
After that experience I had the following thougts.
1) The most successful Linux distribustions seem to be **untu and Debian.
But why are they successfull?
a) Debian:
Has a rediculously bureaucratic organisation in the background and they are open source fanatics. Binary proprietary WLAN adapter firmware is not included by default, even if a user desperately needs it... Debian is popular for webservers and professional use.
Why? I came to the conclusion that you can compare the wy the distribution is made with some ISO 9001 processes in companies in real life. For decision makers this can be a good argument: A good and well-organized creation process of Debian leads to a good distribution in the end. The open source fanatism is also a plus for that purpose: The risk of being sued due to a violation of patents (because LAME is lying around somewhere on a webserver installation etc..) is minimized. Everything is as free as possible, and in the end Debian is first choice for professional use.
b) Ubuntu/**untu:
Derived from the professional distribution described above, with the goal of being as end-user friendly as possible. Due to the massive invest of a generous donor, this goal gets archived very well... Everything works as it should to the user, and there are lots of packages of consumer software for it. Big support communities online, thousands of books and manuals offline.
If you look at it that way, you'll recognize that there's not much room left for other distributions. Professional use: Debian, Ubuntu: End-User... If you add "commercially supported" distributions like Fedora, Open Suse or CentOS, the remaining need of some traditional community distributions like Slackware or Mandrake/Mageia is very limited.
Especially for Mageia I see dark prospects, because most of their goals are accomplished better by Ubuntu, which is supported by a lot of more packages.
And what about Slackware?
Slackware has some interesting features.
-simple package system
-Slackbuild concept
-configuration with config files
-?
I think it would be intersting to find out which kind of users use Slackware for what purpose. Did Slackware gain a numerable amount of new useres in the last 3 years? Are there usage scenarios where Slackware can display its special features?
For example, I wanted to learn a little bit Linux and already had some config file experience from MS DOS and Win 3.1 I like the ideas of having no graphical tools messing up my configuration. But I think this kind of audience is quite rare...
Could that be a purpose: Using Salix/Slackware to learn something about Linux, config files and KISS?
KISS is also a good cue: IMHO, if you look at the original Slackware philosophy, isn't it some kind of contradiction to install the latest KDE or Gnome desktop with all their graphical eye-candy on it? The more so as they often bring their graphical configuration tools with them...
So I think the ready-made Salix Ratpoison distribution was a good approach. Unfortunately ratpoison is a little bit limited for certain scenarios, especially if a single program comes up with many little windows. Maybe Salix/Slackware can develop a window manager/desktop aligned to the distribution's philosophy:
-KISS
-configuration files
-able to run many little windows and dialog boxes...
-no graphical config tools
And Salix/Slackware needs to define its audience, as stated above... And then focus on the needs of that users...