Someone asked some questions about Linux. Here's what I said.
From: Richard Kettlewell <rjk+ah-from=uk.comp.os.linux@sfere.greenend.org.uk>
Newsgroups: uk.comp.os.linux
Subject: Re: Serious opinions sought for research
Date: 08 Mar 2000 20:48:37 +0000
Excellent, a chance to pontificate l-)
"Tom Steinberg" <tsteinberg@iea.org.uk> writes:
> 1) What is interesting about Linux? Why do people talk about it at
> all?
Different people find it interesting for different reasons, of course l-)
When I first encountered Linux some years ago it was interesting as a cheap and convenient way of running the software I wanted - I could just buy a commodity PC from a box-shifter, shuffle floppies for a few hours, and have a reasonably complete UNIX system.
More generally, I would imagine that part of the reason for the current high levels of interest in Linux is that it's come out of nowhere, without much in the way of quantifiable investment, and yet presents a credible challenge in some areas to proprietary UNIX systems and to NT - despite the vast amounts of effort invested in the development of those proprietary systems.
Of course in reality there has been lots of investment in the development of Linux; but much of it is hobbyists doing it for fun, or people solving their own specific needs, rather than companies adding features because they think they can derive income that way, though this does happen too.
Anyway the result is that a lot of the development costs don't appear on anyone's balance sheet; in fact if anything they're more likely to turn up on people's CVs.
> 2) Why has Linux achieved what it has?
Several factors. At one level, the low[1] cost, ability to run on any old hardware, and ready availability (in particular over the net) together mean that it's very easy to try it out without much risk if it turns out to be the wrong thing for you.
[1] zero if you download it over a leased-line connection you're paying for anyway, or borrow a CD off a friend.
At another level, the fact that you can read the source; modify the source; and contribute back to the main line of development, all mean that if you want to make a change, you can (with sufficient skill) make it yourself; or (with sufficient money) pay someone to do so; and then have the results of that work available in future versions as well as the current one.
If you want to add a feature, or fix a bug, in Solaris, or NT, or Windows 98, you have to convince Sun or Microsoft to do it, and then wait for them to do so. With Linux - you can just go and do it.
Another factor - fitting in with the UNIX world has meant that huge amounts of vital software was just there waiting to be compiled for Linux. So once the kernel was written and the tool chain adapted to support it, much of the stuff you need to have a useful operating system was available with relatively little effort.
This also means that it's easy to port applications from other UNIX systems to Linux, and that skills learned on other UNIX systems transfer easily to Linux (and indeed from Linux).
> 3) What has Linux really achieved?
Well, we've got an essentially complete, Free, portable UNIX system, that runs a wide range of code not necessarily written with Linux in mind (as well as a lot of Linux specific code). That's both an achievement and a reason for its success.
> 4) Where is Linux heading, realistically, in the short, medium and long
> terms?
Here, there and everywhere! I think we'll see it spreading further still in server systems in very many businesses in the immediate future, both for internal use and for providing services to customers; hobbyist use will grow a lot; widespread desktop use will probably happen at some point.
I think there's potential for it turning up in all but the most restricted computing environments - I wouldn't expect to see a wristwatch running Linux any time soon for example, but a palmtop or a phone wouldn't surprise me at all. (I mean as a finished product, rather than as a neat hack.) The kernel is lightweight and flexible, even if not all of the userland software is.
> 5) Is Linux sustainable as a project? Is it more or less sustainable than
> non-free projects?
There doesn't seem to be any shortage of volunteers or commercial interest at the moment... While programmers find it useful I think this will continue to be true.
> 6) Is Linux actually aiming at a level of desktop usability on par with
> Windowz? If so, when? If not, why not?
Firstly I should say that I find it much more usable on *my* desktops than I do Windows, but I'm surely not representative of the world as a whole; for example the nearest I get to word processing is plain text or SGML, I don't get on with things like MS Word at all.
Also I should say that I'm not convinced it makes sense to say `Linux' is aiming at anything, rather that particular people might have particular applications for it in mind. This really a remark about terminology than an answer however...
That said, there clearly are people aiming at the desktop; Gnome for example has the eye-candy that is apparently required for this sort of thing (and I do actually find bits of it useful too), and there are a number of applications mirroring some of those found under Windows.
(Also traditional-style applications break out in menus and icons every so often - XEmacs is the best example of this that I have to hand, but there are even versions of Vi with menus and stuff...)
``A level of desktop usability on par with Windowz'' would presumably imply that you could buy a PC pre-configured with some Linux distribution and a bunch of applications and do all the things that you can do in Windows, without having to have any more specialist knowledge than Windows users currently do. I think this probably will happen, though the first attempts will probably not live up to expectations.
> 7) Is it more than just a typical manifestation of idealism which cannot
> threaten the products of the financial incentives of the proprietary
> software world? Could Linux become the CND of the modern age, if the
> Microsoft case ever ends?
Well, I wouldn't expect to see Linux dominate any particular market the way that e.g. Windows currently rules the desktop; but equally I'd be surprised if it just faded away into obscurity. There *is* idealism behind it, but in this case it is an idealism with a razor-sharp practical edge.