You are not logged in.

#1 2008-09-08 00:29:47

Llama
Banned
From: St.-Petersburg, Russia
Registered: 2008-03-03
Posts: 1,379

Possible evolution vector

There's a lot of different Linux distributions. Way too much above reasonable numbers, but this is neither any news nor my point. A lot of distributions exist, because different people see a lot of things in a lot of different ways smile .

But what makes a distro after all?

(1) On one hand, there is the kernel. Basically, distro teams have no business hacking into it too deep. They are expected to scratch the surface in a legitimate way, no more: the kernel is someone's else responsibility.

(2) On the other hand, there is the entire corpus of FOSS. The same logic applies: distro people have no business hacking away there; the task is impossibly huge anyway.

So a distro is just a bit of "glue", which holds together (1) and (2). No big deal? Apparently not - for the time being smile .

My point: is it possible that in due course the "glue" part will sort of fade into insignificance? Just because it's not worth the candles to create yet another distro any more? Just because no visible difference can be made any longer?

Offline

#2 2008-09-08 01:05:49

koch
Member
From: Germany
Registered: 2008-01-26
Posts: 369

Re: Possible evolution vector

Llama wrote:

...Just because it's not worth the candles to create yet another distro any more? Just because no visible difference can be made any longer?

reminds me of a story my grandpa told me about planes when he was young (about 1910 - 1920). he told me there was a discussion that planes can't go faster than, ermm, i think he was speaking about 100 mph. also people thought that there was no need to go faster, not mentioning the dangerous parts of increasing speeed to 200 or + mph...
you know what happened...

short version without the stories of an old man who is long dead...it would be very boring without other distros.

Last edited by koch (2008-09-08 01:06:57)

Offline

#3 2008-09-08 06:36:23

Mandor
Member
Registered: 2006-06-06
Posts: 154

Re: Possible evolution vector

It's about freedom - as long as there are people willing to invest time, there will be new distros. Most of these are already not worth creating by many standards, so it's not a matter of worthiness.


If everything else fails, read the manual.

Offline

#4 2008-09-08 08:43:21

iphitus
Forum Fellow
From: Melbourne, Australia
Registered: 2004-10-09
Posts: 4,927

Re: Possible evolution vector

koch wrote:
Llama wrote:

...Just because it's not worth the candles to create yet another distro any more? Just because no visible difference can be made any longer?

reminds me of a story my grandpa told me about planes when he was young (about 1910 - 1920). he told me there was a discussion that planes can't go faster than, ermm, i think he was speaking about 100 mph. also people thought that there was no need to go faster, not mentioning the dangerous parts of increasing speeed to 200 or + mph...
you know what happened...

short version without the stories of an old man who is long dead...it would be very boring without other distros.

Heavier than air flight is impossible. tongue

(and I'm studying aerospace engineering, so I should know tongue )

Offline

#5 2008-09-13 17:18:19

Llama
Banned
From: St.-Petersburg, Russia
Registered: 2008-03-03
Posts: 1,379

Re: Possible evolution vector

It's about freedom - as long as there are people willing to invest time, there will be new distros.

People invest their time for a shadow of a reason, at least. Take FreeBSD, for instance: people invest there, too, but somehow they just don't try to invent the best mousetrap ever... That's because it would be futile with FreeBSD. I'm just wondering if such luck is ever going to happen to Linux.

Offline

#6 2008-09-13 20:13:24

freakcode
Member
From: São Paulo - Brazil
Registered: 2007-11-03
Posts: 410
Website

Re: Possible evolution vector

Llama wrote:

There's a lot of different Linux distributions. Way too much above reasonable numbers, but this is neither any news nor my point. A lot of distributions exist, because different people see a lot of things in a lot of different ways smile .

But what makes a distro after all?

(1) On one hand, there is the kernel. Basically, distro teams have no business hacking into it too deep. They are expected to scratch the surface in a legitimate way, no more: the kernel is someone's else responsibility.

(2) On the other hand, there is the entire corpus of FOSS. The same logic applies: distro people have no business hacking away there; the task is impossibly huge anyway.

So a distro is just a bit of "glue", which holds together (1) and (2). No big deal? Apparently not - for the time being smile .

My point: is it possible that in due course the "glue" part will sort of fade into insignificance? Just because it's not worth the candles to create yet another distro any more? Just because no visible difference can be made any longer?

The distributions are essentially what the name indicates: they distribute FOSS in a way acceptable by it's target audience. Because FOSS is often distributed as source code, and you need an integration work to make all this source code run on a computer, the distributions are a joint effort to achieve this goal.

As such, when the target audience changes, the way this integration work happens changes too. That's why we have so much distributions: everytime, an enterprise, a client, or a bunch of people, needs FOSS software integrated in a different way than the available ones, they create a new distro. That's valid.

There's of course a BIG parcel of that number that don't change absolutetly anything in the way a distro work, but just want it rebranded. That's a weak reason to create a distro, as it just fragments effort.

Anyway, in the end it's a Darwinism: during the course of time, the distros that have a valid reason behind them will persist, and the bad ones will just vanish.

But I don't think the distro importance will ceasse in the future, unless the upstream projects integrate themselves tightly with the aim to create one, big OS - which is quite the opposite we see today, as Linux kernel is pretty autonomous, Gnome/KDE projects aim for multiplatform, etc.

Last edited by freakcode (2008-09-13 20:16:08)

Offline

Board footer

Powered by FluxBB