tlug: Linux for the masses: a civil reply, I hope.

["Stephen J. Turnbull" <turnbull@example.com>]

>>>>> "tjh" == tjhaslam  <tjhaslam@example.com> writes:

    tjh> Right (or Left.  Or whatever).  And I greatly appreciate
    tjh> Chris`s sense of a Linux dev community

Hmm.  I'm going to have to rethink my idea of the Linux community.
Even today it's not unreasonable to identify "dev community == user
community == `the' community".  (As evidence, you and JDH apologized
for coming in empty-handed.)  A few days or weeks from now, that may
no longer be the case.  Certainly I need to start preparing for that
day now.  And of course I have to think about where the commercial
entities belong in the scheme of things.  (For personal philosophical
use, and that of those who care to read/listen to me.)

    tjh> GNU has no problem with GNU for the masses.  Or for that
    tjh> matter, with what they deem GNU/Linux.  Or further still,
    tjh> `ease of use` via a GUI.  (The perhaps never to appear GNU
    tjh> desktop __TEAK__, for example).  Much of what GNU wants is to
    tjh> take the mystery and elitism out of computing and information
    tjh> technologies--as well as the expense.

"Computing and information technologies" is too broad a class.  GNU
wants to provide a common, solid, well-understood _foundation_.  UI
design, for one, is going to remain an art for a while at least, and
thus by definition (mine, anyway) mysterious and elitist.  The
superstructure of new apps and so on, definitely mysterious.

And RMS does not care one bit what anybody else wants, and especially
not non-programmer users.  His community _is_ the developers, and he
does not hesitate to inflict pain on people who do not rebuild their
tools and OS on arrival of a missive from gnu-announce.  The masses
are of course welcome to use GNU products, but he does not care if the
software is user-friendly or inexpensive (to non-developers).  That is
somebody else's problem.

Believe it or not, *I* am a _lot_ friendlier in principle to non-
developer users than GNU is, or at least than RMS is.

    tjh> a.  CS and ST: not personally asking that YOU develop the
    tjh> tools I and others want.  Sorry if it sounded that way
    tjh> (though I do think Emacs development is the great blackhole
    tjh> of programming talent:

<grin> [1]

    tjh> back to my sorid reality: if by chance you and some friends
    tjh> did, however: 10000`s of people would use these tools.

Sure.  But you _are_ asking that _somebody else_ do the work.
Necessarily so: if all 170 of us on this list quit our jobs today and
started working full-time on the list of apps just you, Jonathan
Byrne, and John De Hoog have mentioned in the last few days, it would
still be years before we were done (it'd be a year just getting us
organized).  I didn't take it personally.

    tjh> d. CS: get ready--but not quite yet--to rouse the old guard
    tjh> and guard the kernel & libraries.  Even and _most hopefully_
    tjh> if the Linux dev com does not fragment/become balkanized from
    tjh> the `outsiders` picking alleged winners and losers and
    tjh> financing accordingly, some of the new code boyz and grrls
    tjh> coming in will not entirely share your values or sense of
    tjh> community: but will at least have some skills.  And GNU GPL
    tjh> might not quite offer the protection you think.  (To revive
    tjh> an earlier half-serious jest: MS Linux, anyone?)

The Linux development community has already balkanized, although not
to the point of civil war, not by a long shot.  Not to worry, it will
work itself out.

I would think MS Linux would be a good thing, from the populist view.
What is the scenario you are thinking about?  Here's mine:

Linux is dead!  Long live Linux!  Development would fork, there would
be a period of confusion, a lot of good people would flock to the
money, and the open source Linux would rise from the ashes and mutate
into something totally unusable to the average user again because only
the hackers would use it, and they will worry more about features and
standards than about stability of this or that patchlevel.  (Maybe it
would be overtaken by the *BSD groups, although I don't think so; that
would require losing Linus and a civil war among the remaining leaders
at the same time; the development process used by Linux is more
effective.)

But the unusability of open source Linux[tm] wouldn't matter, because
you'd have MS Linux, with Word and Excel and fairly shortly NetObjects
Fusion.  And you'd still have Apache and Emacs and Perl and all (if
you wanted them, but despite their availability for Windows few
Windows users I know have them), although possibly not Aladdin
Ghostscript and I'm pretty sure not qmail and xv.  It would be
expensive, at least more so than TurboLinux.  Not unreachably so.

But that's what you really want, isn't it?  It would give a hundred
million users, maybe more, permanent freedom from the Blue Screen of
Death.  Sure, MS Linux users would likely lose the benefit of future
development of open source Linux, because it would be more profitable
for MS to develop inhouse (integrating new kernel features would be
costly, because they would have to avoid conflicts with whatever
device MS used to keep their proprietary parts of the kernel
proprietary).

Theoretically it's possible that MS would use MS Linux as a
springboard, and never look back, and Linux and the other free OSes
would all die.  I think that's unlikely; MS will find some nice
internal contradictions and the MS process will choke itself, just
like NT has.  But Linux itself is already Good Enough[tm], you don't
need it to get better.  You need apps, or that's what you said.

And of course MS would track the kernel, and the next time it
stabilizes they would do the whole process again.  _This is exactly
what RMS wants to see_:  an MS that stops wasting resources on junky
proprietary OSes, and does what they're good at: monopolizing the app
market, to the benefit[sic] of the hundreds of millions to whom
computers are just a tool.  Well, maybe he'd ask that they abandon
Word in favor of WYSIWYG extensions to GNU Emacs, too ;-)

The only protection I want from the GNU GPL is the freedom to use the
published sources without worrying about getting sued (GIF is the
classic example of the worry); I'd be unhappy and frustrated if MS
developed something _really good_ for the kernel and found a legal way
to keep it proprietary, but forcing MS to publish is not really what
the GPL is for.

You see, I want to keep Linux, and open source in general, _for_
myself, not _to_ myself.

I also object to the non-compliant garbage that a more powerful MS OS
would likely spew all over cyberspace, but that's going to happen
anyway.  But I refuse to support, and will argue against, processes
that lead to open source products spewing garbage.  I don't expect MS
to listen.

I expect _you_ (tjh, that is) _will_ listen, and probably decide I'm a
paranoid maniac, and be on the other side of the argument.  That's OK
with me.  I don't _need_ to win.  (Or even be right: you may even
convert me.)  I have the GPL to guarantee there will always be a
Linux; I may not have the most advanced version of a Linux-derived OS, 
but I'll have source.


Footnotes: 
[1]  Emacs is the closest thing to a GUI (Grand Unified Interface) Linux
has.  I know at least one person (not me) whose /etc/passwd contains

juser:*encrypted*:1000:1000:J. User,Earth:/home/juser:/usr/bin/xemacs

I'm not about to recommend that as the default shell to TurboLinux,
though ;-)

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences        Tel/fax: +1 (298) 53-5091
---------------------------------------------------------------
Next Meeting: 10 October, 12:30 Tokyo Station Yaesu central gate
Featuring the IMASY Eng. Team on "IPv6 - The Next Generation IP"
Next Nomikai: 20 November, 19:30  Tengu TokyoEkiMae 03-3275-3691
---------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp