Re: tlug: Re: kernel modules (was: Kernel rebuild problem)

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

>>>>> "Rex" == Rex Walters <rex@example.com> writes:

    Rex> On Wed, Sep 30, 1998 at 05:01:37PM +0900, Stephen J. Turnbull
    Rex> wrote:
    >> I really don't understand your point (nor with respect to the
    >> other three [mis-]quotes).  To be melodramatic about it, we,
    >> _Ye Olde Garde_, see a real threat to our currently very
    >> satisfying community.

    Rex> My admittedly knee-jerk reaction was to the "us and them"
    Rex> mentality.

It's pretty natural for "us" to want "them" to be more like "us."

Not nice, I guess, but natural....  :-)

    Rex> I certainly agree that bad protocols, bad API's, bad
    Rex> programming practices, bad implementation, bad administrative
    Rex> practices, etc. could _become_ a threat, but I see a clear
    Rex> distinction between users and implementors.

I don't:  the demands of the users will be reflected in what the
implementors implement.  Microsoft has done an honorable thing in
putting a computer on my mother's desktop.  Wouldn't be there without
Windows.  (She had a Mac once; choice was too slow or too expensive.)

But that's been at the expense of introduction of bad (all that stuff
you list), because it's good business for Microsoft.  It's alleged
that MS does it on purpose on occasion (and I can't see any good
reason for registering the Windows-1214 character set with the IANA,
ISO-8859-whatever should be good enough for email), but mostly it's
because the market doesn't care if Outlook Express sets its
Content-Type to ASCII for pure ASCII messages, but does care about
ease of use.

    Rex> In other words, I'm all for Ye Olde Garde protecting the
    Rex> protocols and good programming practices, but doing this by
    Rex> keeping the community small and self-contained is throwing
    Rex> the baby out with the bath.

And how do you propose to protect good practice without imposing
limits that will frustrate users?

People who code for personal pride aren't a problem; I suspect John De
Hoog will not get a decent Linux environment in his lifetime if he has
to wait for them.  But there's money to be made coding to satisfy him.
And bragging rights over market share.  (OK, Larry Wall blanches when
called "alpha hacker", but look at Apache's continual reference to its
installation stats.)

Consider the De Hoog vs. Byrne discussion on WYSIWYG HTML editors.
JDH is someone I'd like to see using Linux, but his page, as pretty as
it is, makes nsgmls spit error messages like crazy.  He is not going
to rate that as important, I suspect, as long as he doesn't get
complaints from non-standard-bigots.  And he is going to continue to
(financially, I assume) support that bad practice, because he doesn't
pay the price for it.  He does pay the price for ugly pages, he does
pay the price if he has to switch to a tool lacking the "site
management feature".

_Most business users are going to take the same attitude._  No?

    Rex> No, I was just waving my Linux-advocacy flag.  I believe it's
    Rex> too easy to continually make "in" references to Serdar Argic,
    Rex> "n.a.p", and such, without realizing the harm it can do.

I realize the harm it can do.  I don't care; I _like_ ObRefs.  I think
it's a good thing; I like it when other people do it.  And you
_personally_ must bear a fair part of the burden for me continuing to
do so.  :-)

That admission made, in this particular case, I didn't really have an
alternative.  Did I?  Those weren't "in" references; I doubt most of
the people on this list knew what I was talking about.  That was
history, crucial to my argument.

    Rex> (btw -- it's been so long since I've read news regularly that
    Rex> I don't remember.  What was "n.a.p" --
    Rex> "news.abuse.pffthtth"?)

news.admin.policy.  See?  There you go again, encouraging me.  :-)

    Rex> I agree with your sentiments with regard to developers, but
    Rex> what is wrong with novice users?

Nothing.  I go to some lengths to help them join the community _as it
stands, with its current customs_.  I don't see anything wrong in
principle with changing the customs either; I support modules and
package managers, for example.

But emphasizing ease of use comes with a great danger of deemphasis of 
standards; baby and bathwater....

    >> I'm sorry, but it just so happens that a lot of the best
    >> (worst) examples of standards-noncompliance are M$-related.
    >> Does that make me a bigot?

    Rex> Yikes!  Have I somehow miscommunicated so badly that you
    Rex> think I'm defending Microsoft standards compliance?

No.  However, along the forms of bigotry _you_ mentioned, there is
also knee-jerk anti-M$ bigotry, which others have taken Ye Olde Garde
(and linuxers and linux MLs) to task for.  I shouldn't have lumped
them together so abruptly.

It's just that while we're talking about what we would like Linux to
_be_, we also need to talk about what it _can't be_ or _shouldn't
be_.  For that latter, M$ is a big fat target, _especially_ since
they've b(r)ought so many good ideas to market in a half-assed way.

    Rex> Personally, I'd like to see Linux as much more than that.
    Rex> "World Domination", as Linus jokes, would actually be a good

And RMS, for that matter (and far earlier than Linus).  Except that
RMS doesn't see the humor in it (at least not where I've seen his
comments on the matter).

    Rex> thing in my book.  People would get more work done, and would
    Rex> spend less time and money on "infrastructure" and
    Rex> administration.

"Patience, my son.  May the Source be with you."  (Sorry, Yoda.)

Anyway, XEmacs will get there first (^^).  (I wish.... :)

    Rex> Linux is too good to keep to ourselves.

True, I'm just not in that great a hurry.  I introduce/advocate it to
my friends and students.

But people _like_ their inefficiencies (well, actually, some feature
that is linked to the inefficiency, via product or practice).  I don't
see any point in getting hated for pointing them out and suggesting
they do real, hard, work getting them smoothed out.

    Rex> The balance is important, and I don't want to see it change
    Rex> abruptly, but self-interest if nothing else makes me want to
    Rex> see Linux on as many desktops running as many real "work I'm
    Rex> paid for" applications as possible.

That's fine by me.  I am definitely in favor of incrementally adding
ease of use _for users_ so that you guerrilla admins can extend your
domains [pun?].  How fast to add it, we differ quite a bit, I guess.

So it basically comes down to the fact that you're more optimistic
than I am about the robustness of the "Linux balance" to a massive
influx of new users.

I hope you're right.  It will be a lot more fun that way.  Other
things equal, more _is_ merrier.

-- 
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