Linus Torvalds wrote:
$B!V(BI'm distrustful of projects that do not have well-defined goals, and
well-defined interfaces. They tend to bloat and do "everything" over
time. This is what gives us horrors like GNU emacs and Mach: they
don't try to do one thing well, they try to do _everything_ based on
some loose principle ("LISP is good" or "microkernels make sense" or
"GGI should do graphics")$B!W(B
--------------
I'm in general of a anti-hero attitude... and so i haven't much read
any writings by Linus. However, in the past decade since i'm in the
unix industry (since 1998), I have ran into a few of Linus's writings.
My impression is that he's one of the rather more pretentious type of
character, prone to use words and rhetoric to smite different opinions
in a nonchalant way. (but i think he's overall a good human animal)
Yesterday i was checking out what's the current status of the
vapourware Fresco windows system, and in turn i was reading about GGI
( http://en.wikipedia.org/wiki/General_Graphics_Interface )
Then i ran into one of Linus's post: regarding it.
List: linux-kernel
Subject: Re: GGI Project Unhappy On Linux
From: torvalds () transmeta ! com (Linus Torvalds)
Date: 1998-03-26 6:52:04
http://marc.info/?l=linux-kernel&m=89089527200744&w=2
------------
I don't know much about low-level programing. The article seems
reasonable and level headed, but except that gratis emacs lisp
vilification (and Mach). What a practical asshole.
Xah
···@xahlee.org
$B-t(B http://xahlee.org/
On Sun, 23 Dec 2007 20:48:45 -0800, Xah Lee wrote:
> I don't know much about low-level programing. The article seems
> reasonable and level headed, but except that gratis emacs lisp
> vilification (and Mach). What a practical asshole.
Have you seen the infamous Google Tech Talk on Git? That is enlightening
in more ways than one!
--
Sohail Somani
http://uint32t.blogspot.com
On Mon, 24 Dec 2007 05:04:12 +0000, Sohail Somani wrote:
> On Sun, 23 Dec 2007 20:48:45 -0800, Xah Lee wrote:
>
>> [quoted text muted]
>
> Have you seen the infamous Google Tech Talk on Git? That is enlightening
> in more ways than one!
Do you have a link?
On Mon, 24 Dec 2007 10:32:33 -0800, Jeff Cunningham wrote:
> On Mon, 24 Dec 2007 05:04:12 +0000, Sohail Somani wrote:
>
>> On Sun, 23 Dec 2007 20:48:45 -0800, Xah Lee wrote:
>>
>>> [quoted text muted]
>>
>> Have you seen the infamous Google Tech Talk on Git? That is
>> enlightening in more ways than one!
>
> Do you have a link?
http://www.youtube.com/watch?v=4XpnKHJAok8
--
Sohail Somani
http://uint32t.blogspot.com
On Sun, 23 Dec 2007 20:48:45 -0800, Xah Lee wrote:
> Linus Torvalds wrote:
> $B!V(BI'm distrustful of projects that do not have well-defined goals, and
> well-defined interfaces. They tend to bloat and do "everything" over
> time. This is what gives us horrors like GNU emacs and Mach: they
> don't try to do one thing well, they try to do _everything_ based on
> some loose principle ("LISP is good" or "microkernels make sense" or
> "GGI should do graphics")$B!W(B
I'd hardly call this a vilification of Lisp. He just seems to be saying
that "Lisp is good" is a very vague principle to base everything on. I
don't have the faintest idea how he applies that to GNU Emacs (afaict, its
problems, such as they are, aren't caused by Lisp, and the main idea is
about extensibility, not Lisp), but I can't really disagree with the
general idea. Nowhere does he say that Lisp is bad. Only that "Lisp", by
itself, is a poor philosophy.
PD> I'd hardly call this a vilification of Lisp. He just seems to be saying
PD> that "Lisp is good" is a very vague principle to base everything on. I
PD> don't have the faintest idea how he applies that to GNU Emacs (afaict,
PD> its problems, such as they are, aren't caused by Lisp, and the main
PD> idea is about extensibility, not Lisp), but I can't really disagree
PD> with the general idea. Nowhere does he say that Lisp is bad. Only that
PD> "Lisp", by itself, is a poor philosophy.
doesn't "Lisp philosophy" mean extensibility? "programmable programming
language". programming language that can be extended _from inside_ in
natural way.
so is Emacs, btw, it can be extended or customized _from inside_ in natural
way.
certainly that doesn't fit "unix" philosophy -- they like small rigid bricks
with well-defined contours glued to each other with scripting glue making
construction somehow "sort of works".
so idea of a living, flexible program seems alien to people who get used to
rigid programs, no wonder.
many Windows developers who work mainly with C++ will find bash shell a "a
project that do not have well-defined goals", "based on some loose
principle: 'scripting is good'". living in a world where only few programs
have working command-line interface, but many do support COM RPC, they will
wonder why anyone needs that scripting glue if he can call with application
function directly via COM interfaces.
this reminds me a joke: "military general asks civilian journalist: -- if
you, civilians, are so smart, why do not you march in formations then?".
On Mon, 24 Dec 2007 12:03:48 +0200, Alex Mizrahi wrote:
> PD> I'd hardly call this a vilification of Lisp. He just seems to be saying
> PD> that "Lisp is good" is a very vague principle to base everything on. I
> PD> don't have the faintest idea how he applies that to GNU Emacs (afaict,
> PD> its problems, such as they are, aren't caused by Lisp, and the main
> PD> idea is about extensibility, not Lisp), but I can't really disagree
> PD> with the general idea. Nowhere does he say that Lisp is bad. Only that
> PD> "Lisp", by itself, is a poor philosophy.
>
> doesn't "Lisp philosophy" mean extensibility? "programmable programming
> language". programming language that can be extended _from inside_ in
> natural way.
I think so. But while extensibility may be Lispers' philosophy, Lisp
itself isn't what I would call a philosophy. More like the inspiration and
vehicle for a philosophy. "Extensibility" might be your philosophy, and
you might choose Lisp for that reason. "I don't want to fight with my
language" might be your philosophy, and you might choose Lisp for that
reason. But "lol its good" would not be a well reasoned philosophy.
And that, I think, is what Linus was talking about.
Not that I know of anyone who has picked Lisp for that reason, nor do I
think Lisp was poorly chosen for Emacs.
Paul Donnelly wrote:
> On Sun, 23 Dec 2007 20:48:45 -0800, Xah Lee wrote:
>
>> Linus Torvalds wrote:
>> $B!V(BI'm distrustful of projects that do not have well-defined goals, and
>> well-defined interfaces. They tend to bloat and do "everything" over
>> time. This is what gives us horrors like GNU emacs and Mach: they
>> don't try to do one thing well, they try to do _everything_ based on
>> some loose principle ("LISP is good" or "microkernels make sense" or
>> "GGI should do graphics")$B!W(B
>
> I'd hardly call this a vilification of Lisp. He just seems to be saying
> that "Lisp is good" is a very vague principle to base everything on. I
> don't have the faintest idea how he applies that to GNU Emacs (afaict, its
> problems, such as they are, aren't caused by Lisp, and the main idea is
> about extensibility, not Lisp), but I can't really disagree with the
> general idea. Nowhere does he say that Lisp is bad. Only that "Lisp", by
> itself, is a poor philosophy.
Yes, that's exactly how I interpreted what he said. I agree that nobody
should make serious engineering choices on the basis of a vague idea
like "Lisp is good".
As for whether Linus Torvalds is opinionated: I can hardly imagine
what life must be like for someone who has been so lionized and
canonized as Linus. After having people treat you like some kind
of deity, it's probably understandably hard to hold on to your
modesty. So I'm inclined to cut him some slack. Just from reading
this posting, I don't think he's being over-opinionated and I
certainly would not describe it as obnoxious or anything like that.
Daniel Weinreb <···@alum.mit.edu> writes:
>> I'd hardly call this a vilification of Lisp. He just seems to be saying
>> that "Lisp is good" is a very vague principle to base everything on. I
>> don't have the faintest idea how he applies that to GNU Emacs (afaict, its
>> problems, such as they are, aren't caused by Lisp, and the main idea is
>> about extensibility, not Lisp), but I can't really disagree with the
>> general idea. Nowhere does he say that Lisp is bad. Only that "Lisp", by
>> itself, is a poor philosophy.
>
> Yes, that's exactly how I interpreted what he said. I agree that nobody
> should make serious engineering choices on the basis of a vague idea
> like "Lisp is good".
>
> As for whether Linus Torvalds is opinionated: I can hardly imagine
> what life must be like for someone who has been so lionized and
> canonized as Linus. After having people treat you like some kind
> of deity, it's probably understandably hard to hold on to your
> modesty. So I'm inclined to cut him some slack. Just from reading
> this posting, I don't think he's being over-opinionated and I
> certainly would not describe it as obnoxious or anything like that.
From reading some other stuff he has said, even cutting he some slack
from the media attention, he doesn't come across as someone who thinks
before he speaks. He has said some really odd stuff, this not that
bad. I'd take anything he says with a truckload of salt.
/Andreas
--
A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?
[ comp.lang.lisp only; http://www.nhplace.com/kent/PFAQ/cross-posting.html ]
Daniel Weinreb <···@alum.mit.edu> writes:
> Yes, that's exactly how I interpreted what he said. I agree that
> nobody should make serious engineering choices on the basis of a
> vague idea like "Lisp is good".
Indeed. This is the very essence of dogma, and the thing I have
repeatedly opposed. Saying, about anything, "do x and you will ipso
facto be doing good" (or "... will be good"), without analyzing the
consequences, makes no sense to me.
I try to be careful not to say that using Lisp makes things Good or
failing to use Lisp makes things bad. I generally like Lisp, often
prefer using it to not using it, and often miss it when I'm not using
it. But I don't think someone who doesn't use it is doing bad... I'm
just happy to share my enthusiasm about it with people who are of
similar mind or interest, or with those who are just curious. I don't
feel a need to convert the world, nor do I want to be converted by
others.
I think evangelism, whatever the domain one is evangelism is about,
is very double-edged and one must take some care not to push too hard
on others lest they push back invasively in reverse.
Besides, since it's just obviously true that no one thing can be
automatically good for all purposes, it destroys one's credibility to
start being angered that anyone might say such an obvious truth.
When I was interviewed by Slashdot, a lot of people commented that they
took a second look in part because they expected me to take the position
that Lisp was better than everything, and I specifically took the position
that did not make it an us/them conflict, but rather "a good programmer
must have many arrows in his quiver" ... especially in the heterogeneous
programming environment that one finds today.
A somewhat related aside, just to make this personal to my own experience:
This seems to me to be the problem I have about free software as well.
I don't object to either free software or open source on a per se
basis, I object to the notion that the mere use of that specific tool
will make everyone using it and everything it touches Good and the
failure to use it will make you and everything you touch Bad. This is
more nuanced than simply liking or not liking free software--rather, I
reserve the right to judge the individual use. I have tried to
explain that I am not "anti-free-software" but rather
"anti-'free-software-qua-panacea'", but that doesn't seem to work well
for people as a bumper sticker, in part because many have to run to
the dictionary to look what "qua" and/or "panacea" mean... and many
don't bother and seem to just decide I'm attacking them just for not
agreeing with the simpler, more mindless phrasing of "free software is
ipso facto good". (Ok, they don't use "ipso facto" in the short form...
But you can hear it whispering in the wind there if you listen really
carefully... and it's that little bit of extra dogma that drives me nuts.)
And so, in the context of the main text of this thread, I can
appreciate Linus' probable desire to be given similar latitude. I
concur with DLW's remark:
> I can hardly imagine what life must be like for someone who has been
> so lionized and canonized as Linus.
I'm often misquoted myself, as is the case in the Wikipedia entry that
exists for me now. The part about my "opposition" to open source
appears to have resulted from someone (Pmetzger 30-Apr-2006) trying to
clarify my position, changing "his attitude towards" (which at least
allows me wiggle room) to the more clinical "his opposition to" (which
actually paints me considerably more rigidly, and sans reference, than
is appropriate). I only haven't fixed it because it seems like
Wikipedia doesn't really like people managing their own biographies,
and so I just kind of wince when I see this kind of thing. One of
these days I'll create a position page on my opinion, I suppose, and
then someone can reference that and fix the entry.
And I can only imagine Linus must do a fair amount of wincing, too.
In short: The failure to champion something is not vilification.
On Dec 23, 8:48 pm, Xah Lee <····@xahlee.org> wrote:
> Linus Torvalds wrote:
>
> $B!V(BI'm distrustful of projects that do not have well-defined goals, and
> well-defined interfaces. They tend to bloat and do "everything" over
> time.
You pulled that piece of opinion from a 1998 e-mail.
Since 1998, a cynic might remark, the above is pretty much what has
happened to the Linux kernel.
Just a few years after that, in 2001, we find Torvalds no longer
talking about having well-defined goals.
``Let's just be honest, and admit that [Linux] wasn't designed.''
``But I _am_ claiming that there is no common goal, and that most
development ends up being done for fairly random reasons - one persons
particular interest or similar. ... It's "directed mutation" on a
microscopic level, but there is very little
macroscopic direction.''
> This is what gives us horrors like GNU emacs and Mach: they
> don't try to do one thing well, they try to do _everything_ based on
> some loose principle ("LISP is good" or "microkernels make sense" or
> "GGI should do graphics")$B!W(B
Isn't it the case that the code written for the Emacs platform tries
(as a collective base) to do everything, not Emacs itself?
The Lisp just makes it integrated, so it looks like one program.
Some entity, AKA Kaz Kylheku <········@gmail.com>,
wrote this mindboggling stuff:
(selectively-snipped-or-not-p)
> Isn't it the case that the code written for the Emacs platform tries
> (as a collective base) to do everything, not Emacs itself?
>
> The Lisp just makes it integrated, so it looks like one program.
Some people just don't understand non black//white type of stuff, they just
cannot grok grey.
Cor
--
SPAM DELENDA EST HTTP://WWW.CLSNET.NL/MAIL.PHP
(defvar My-Computer '((OS . "GNU/Emacs") (IPL . "GNU/Linux")))
Alle schraifvauden zijn opsettelick, teneynde ieder lafaart de cans te
gevuh over spelingk te mekkuh instede de inhaut
> I'm distrustful of projects that do not have well-defined goals, and
> well-defined interfaces.
And that from the creator of Linux... split personality? Hilarious.