- Site menu:
- Latest Discussions
- Latest Newswire
- Archives
- Submit Story
- Story HOWTO
- Syndicate
- E-mail Alerts
Why Mark Shuttleworth Is Important to Desktop Linux
|
|
|
Comments (7)
False Premise
Wow!
We really have lost sight of why we have distro's, don't we? The goal was to provide a free operating system to users and developers. Different distro's have different defaults to suit different needs. While distro's are not binary compatible and they differ in composition, historically, they all draw from a common pool of upstream sources and they all offer a broad selection of the available FOSS software.
Somehow, along the way, the goal seems to have shifted towards getting marketshare and getting a majority of computer users on something, anything that runs a Linux kernel, regardless of the compromises made to shoehorn a Linux based thing into the consumer market.
Android is a Linux based thing, but the community accepts the huge market share gingerly as a "Linux" success, as Android is a shadow of the vision behind Free Software and Open Source.
When it comes to Ubuntu, it started out as a (GNU/)Linux distro, but increasingly it is becoming a Linux based thing. Linux maybe powering it on the lowest levels, but it is slowly moving away from what we understand as a distro. Maybe that is a compromise that needs to be made if you want to capture a market that doesn't care what powers their digital life. It is a far cry from what a healthy distro ecosystem could have been.
In essence, I believe, we are throwing the baby out with the bathwater. What does the GNU/Linux community gain if another "Android" arises in the form of Ubuntu? They say that the tide raises all boats, but Ubuntu moors more and more often in a different (foreign) harbor than the GNU/Linux boats.
We'll eventually get to a situation, where we will have a behemoth Canonical (instead of MS) providing a Linux kernel powered Ubuntu OS (with some degree of freedom) that on a grey day in the past used to be a Linux distro. Then we will have the GNU/Linux distro's, who will be exactly where they we're in the days of MS, toiling away to compete with the ubiquitous Ubuntu, while not being able to make use of the incompatible and vast Ubuntu software catalog.
The only difference and a big, beneficial one, is that the more stuf uses a Linux kernel, the more drivers for devices will be written for Linux.
Somehow, along the way, the goal seems to have shifted towards getting marketshare and getting a majority of computer users on something, anything that runs a Linux kernel, regardless of the compromises made to shoehorn a Linux based thing into the consumer market.
Android is a Linux based thing, but the community accepts the huge market share gingerly as a "Linux" success, as Android is a shadow of the vision behind Free Software and Open Source.
When it comes to Ubuntu, it started out as a (GNU/)Linux distro, but increasingly it is becoming a Linux based thing. Linux maybe powering it on the lowest levels, but it is slowly moving away from what we understand as a distro. Maybe that is a compromise that needs to be made if you want to capture a market that doesn't care what powers their digital life. It is a far cry from what a healthy distro ecosystem could have been.
In essence, I believe, we are throwing the baby out with the bathwater. What does the GNU/Linux community gain if another "Android" arises in the form of Ubuntu? They say that the tide raises all boats, but Ubuntu moors more and more often in a different (foreign) harbor than the GNU/Linux boats.
We'll eventually get to a situation, where we will have a behemoth Canonical (instead of MS) providing a Linux kernel powered Ubuntu OS (with some degree of freedom) that on a grey day in the past used to be a Linux distro. Then we will have the GNU/Linux distro's, who will be exactly where they we're in the days of MS, toiling away to compete with the ubiquitous Ubuntu, while not being able to make use of the incompatible and vast Ubuntu software catalog.
The only difference and a big, beneficial one, is that the more stuf uses a Linux kernel, the more drivers for devices will be written for Linux.
I wanted to correct the notion that distributions are not binary compatible. In general distributions are binary compatible (as long as you don't change hardware architectures of course). They may not have a version of the same package manager, and they may not have some libraries installed, but technically, those things could be overcome by installing the missing libraries and installing the contents of a package manually (or installing the missing package manager). They don't constitute binary incompatibility.
Quoting:In general distributions are binary compatible (as long as you don't change hardware architectures of course).
Yep. I have temporarily used unmodified Debian libraries in a running PCLinuxOS in order to overcome a printer problem. It worked perfectly. The PCLinuxOS devs later corrected the printer setup problem which rendered using the Debian libraries moot.
It isn't always 100%, though,... It's not so much a binary executable problem as it is a system architecture problem... The more a thing is tied into the system architecture changes of a distro, the less likely it is to work on another... I've run into this in the (fairly distant) past...
Yes, I'm aware that you can Frankenstein your setup, but the burden of keeping it all running increases with the amount of "foreign DNA" introduced. Used as is, a foreign package will probably not work out of the box.
> Yes, I'm aware that you can Frankenstein your setup, but the burden of keeping it all running increases with the amount of "foreign DNA" introduced.
Depends. Did you create a package for your system from the installation? If so, it will largely handle keeping it running for you.
Depends. Did you create a package for your system from the installation? If so, it will largely handle keeping it running for you.
Post a comment
A free account is needed to post. Write your comment here first: your text is kept in this browser while you sign up or log in, and is waiting for you when you come back.
Your message:Markdown: **bold**, *italic*, [text](https://link), - lists, > quotes, $x^2$ for math. Formatting help
Sign up and post (it is free) or Log in and post
To your point of Canonical's work being of use to the rest of the non-Ubuntu universe...the R[he]D that Canonical has poured into tools that are _trully_ distro-agnostic seems (no quantitative data) largely negligible. Sure, they do a lot of work for the kernel and other miscellaneous bs...but again, not to an extent that would make me worried of losing Canonical. Most efforts are focused into tools that never make it outside of Ubuntu...interfaces, artistic work, upstart, mir, etc. Sure...GPL, but GPL that other people cannot easily use or modify is only nice in a strictly philosophical sense[/he] pragmatically, not so.
The article raises some interesting points nevertheless...but I am a bit tired of people trying to convince me that linux (or desktop linux) owes anything to X ( here...X = Ubuntu). To me, it always seems like a blatant case of missing the point of linux. The "year of desktop linux" might make some people all warm and fuzzy inside...but gnu/linux doesn't care, and neither do I.
Note: It is late...so I might have to come back and add some clarifications later.