Linux-Advocacy Digest #124, Volume #26           Fri, 14 Apr 00 10:13:05 EDT

Contents:
  Re: uptime -> /dev/null (Opinionated)
  Re: The truth is often painful... (The Ghost In The Machine)
  Re: Linux for ex-Windows users (long story) (Bloody Viking)
  Re: uptime -> /dev/null (The Ghost In The Machine)
  Re: Be & Linux & Microsoft... (abraxas)
  Re: The truth is often painful... (CG)
  Re: What GUI development tools are there for Linux? (abraxas)
  Re: No Microsoft Certification = NO JOB! ([EMAIL PROTECTED])
  Re: The truth is often painful... (CG)
  Re: benchmark for speed in linux / windows ("Drestin Black")
  Re: Sendmail refuses connection (Hans Dumbrajs)
  Re: Linux vs. Windows Benchmark ("Drestin Black")
  Re: uptime -> /dev/null (CG)
  Re: uptime -> /dev/null (CG)
  dvwssr.dll (Tesla Coil)
  Re: Be vs. Linux ("Jim Ross")
  Re: What GUI development tools are there for Linux? ("Mike")
  Re: What GUI development tools are there for Linux? (Christopher Browne)
  Re: Corel Linux Office 2000 dubious at best? (Gary Hallock)
  Re: BSD & Linux (Peter da Silva)

----------------------------------------------------------------------------

From: Opinionated <[EMAIL PROTECTED]>
Subject: Re: uptime -> /dev/null
Date: Fri, 14 Apr 2000 07:25:29 -0500

Pedro Ballester wrote:

>    A new thought, have you ever heard about ecologism, global temperature
> raising, energy missspending and the such ? Yeah yeah, it sounds far, but
> it is here.

Have you ever thought about how much energy savings and far less damage to
the ecology there would be if everyone had their computers on 24hrs./day and
telecommuted to work (ie - no Cars on the road to contribute to the global
temperature raising).




------------------------------

From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: The truth is often painful...
Date: Fri, 14 Apr 2000 12:34:12 GMT

In comp.os.linux.advocacy, mh <[EMAIL PROTECTED]>
 wrote on Fri, 14 Apr 2000 04:01:39 +0000 <[EMAIL PROTECTED]>:
>A lot of half-truths and diatribes here, but the simple truth is this: 
>when it comes to business oriented desktop software Linux applications
>are AT LEAST TWO GENERATIONS BEHIND the stuff Microsoft produces.

And twice as reliable. :-)

>
>Please note that I am a Linux advocate.  My home network is Linux.  I
>believe in the principles, the ethics, of open source and admire the
>people who have made GNU/Linux a reality.  I use Netscape 4.7, both for
>email and news, not because I ACTUALLY PREFER Netscape, but because
>there is nothing better available on Linux.

Indeed.

>
>Personally, I think that's pathetic, given the quality of the Linux
>OS--but it's true.  Applixware?  StarOffice?  Get real.  If I had a
>choice, based purely on functionality, I'd take Office 95 over either. 

Office 2000, you mean.

>In fact, I'd take Word 2.0, Excel 5.0 and Access 2.0 over any similar
>programs currently available for Linux.

You haven't seen LaTeX or TeX do its thing, then.  It's
extremely elegant. :-)

>
>I use Linux because I love its open nature, its flexibility, its
>empowerment, and because I believe in the philsophy/ethics of the Free
>Software Foundation.  I cannot contribute as a programmer, because I
>don't possess the skills.  But I can offer an honest evaluation based on
>years of experience with a wide range of business oriented desktop
>software in a business environment, and the truth is that in this
>particular realm Linux SUCKS.  I fervently hope to see the day when that
>is no longer true.

I will merely throw in the suggestion that perhaps we should switch
to a tagged-messaging protocol internally in the operating system --
whatever OS becomes mainstream in the next few years -- similar to
my understanding of the Tandem OS, built on my understanding of
the AmigaDOS, which made very heavy use of shared memory message passing.
(In the case of Tandem, no memory is shared; in the case of the Amiga,
memory was shared for performance reasons; whoever held the message
held the memory, conceptually speaking.)

Tandem's architecture is extremely scalable (up to 4,096 processors,
and probably even more if they can remove that 12-bit tag that they've
obviously got somewhere).  Unix is not, although it performs much
better than NT in this regard.

Of course, Tandem's a 20-year-old operating system, too.  Does this
make it two generations behind? :-)

-- 
[EMAIL PROTECTED] -- Unix originally didn't have shareable libraries,
                    sockets, and windowing display.  Think on that.

------------------------------

From: Bloody Viking <[EMAIL PROTECTED]>
Subject: Re: Linux for ex-Windows users (long story)
Date: Fri, 14 Apr 2000 12:36:08 GMT

[EMAIL PROTECTED] wrote:

: I never got converted to Windows, being a UNIX bigot.  I went straight
: to Linux.  But I now have the same problem with UNIX that some people
: have with Windows.  I've been spoiled by Linux.  Linux has made amazing
: progress in five years, but HP-UX and AIX are still about the same as
: they were.

I guess the progress with Linux is from having to try to compete with
Windows, not other UNIXes. The other UNIXes were never really designed for
trying to compete on the desktop. Instead, they were and are marketed to
be used on servers, minicomputers, and in the case of UNICOS, Crays. 

Oddly, as my pet story depicts, it was this "mainframe" appeal that helped
attract me toward Linux. My mentality about computers with DIY programming
came very early on. With that old Commodore, at the time all the apps,
such as they were, were on floppy and I couldn't afford the drive. So, I
had the tape deck and took up programming. 

Given how Linux has to try to compete on the desktop, it has a lot more
"normal" apps than other UNIXes, so Linux is an awful good all-around
performer. It's a desktop OS that'll please the office worker who gets
used to Wordperfect and for the Admins. And if an old mainframer used UNIX
mainframes in the past uses it, he's happy too! This could be a possible
down point. That's becuse it's anything you want it to be, but never quite
optimal for any one environment. 

That could cause some fragmenting of Linux for different uses. An affluent
computer enthusiast could get Linux Extreme and have a homebrew
supercomputer. But that distribution probably would not be good for office
apps. It was meant for math modellers. And of course, a Linux distribution
optimised for a desktop would not be good for hardcore math modellers. 

-- 
CAUTION: Email Spam Killer in use. Leave this line in your reply! 152680
 First Law of Economics: You can't sell product to people without money.

4968238 bytes of spam mail deleted.           http://www.wwa.com/~nospam/

------------------------------

From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: uptime -> /dev/null
Date: Fri, 14 Apr 2000 12:58:58 GMT

In comp.os.linux.advocacy, Bastian <[EMAIL PROTECTED]>
 wrote on 13 Apr 2000 21:32:27 GMT <[EMAIL PROTECTED]>:
>On 13 Apr 2000 19:04:05 GMT, matts wrote:
>>>
>>> Thought:
>>>       Maybe the reason most people turn their computer on and off
>>>       all the time is that they are running M$ Windows?
>>
>>Or maybe they don't feel like butching their hardware for no
>>purpose???????  Using it all day to do work is a useful thing,
>>keeping it on to do nothing is not.  Just my (Canadian)$0.02.
>>
>
>Especially harddisks are known to be very "touchy". You can let a drive run
>for years without a shutdown, and they'll work just fine. If you switch it
>on and off too often (official recommendations are AFAIK 3000 times give or
>take), it'll screw up soon. Monitors behave similiarly.
>A light bulb hardly ever screws up when it's running. The dangerous point
>(as with every electrical device) is the time when you switch it on. It has
>something to do with Ohm... (don't ask me about the details please :-)

Actually, it's fairly simple, in some cases.  In the case of the light
bulb, it's because the filament actually stretches slightly during
initial warmup (I've seen high-speed video of it pulsating, but it's
fairly obvious when one remembers that things expand upon heating
and contract upon cooling -- and that filament gets *hot*). Also, I
would guess the tungsten sublimates slowly over the life of the bulb
while lit (obviously, it's not quite hot enough to melt during
operation), making it more fragile.  (This would explain why burnt-out
bulbs appear darker than new ones, anyway.)

In the case of a disk drive, similar thermal effects ensue, although
on a lesser scale.  There's also the issue of the head sitting on
the drive when cold (in a "parking spot", somewhere off the actual
active area of the disk), the increased torque required during startup,
presumably, the non-optimal configuration of the bearings just as power
is applied (I can't say I know precisely what type of bearing is used
on a modern disk drive, though), and a few other things.

All in all, I'd think a disk would just want to keep sucking juice
if it could (and if it had a brain :-) ).  A light bulb probably
might, too -- although the bulb might worry about its tungsten
stealing away... :-)

As for uptime -- it's not clear that uptime is as important
for a desktop as it would be for a server, but one issue that
is annoying is the death of a download because something croaked.
(Not as big of an issue in this age of high-speed bandwidth.)
Another is that development tools (including Word, which develops
documents, in a fashion) shouldn't die and lose all of the user's
hard-won typing and/or thinking.  I've had Visual C++ do that once,
and, while losing only a few minutes of work (Visual C++ saves every
time a build is done, because the compiler, unlike Borland C++ 4.51,
can't read directly from the "pad" -- this is actually an advantage
since it means the compiler can be independent of the GUI; Word has
an autosave feature), it can be disconcerting.

In other words, reliability is the important concept here, not uptime,
although the one is a measure of the other.  Recoverability might
be, too -- if a tool *does* die, the user should be able to reinvoke
it and pick up where he left off, and modern tools are getting
more complicated; vi is looking like a bit of an anachronism
(although it's also looking like a simple, but heavy, metal vacuum
cleaner, compared to the cheesy light ones -- less inclined
to break :-) ).

One especially reliable item should be the OS kernel.  If that
dies, it takes everything else with it -- hence the jokes about
the Blue Screen Of Death, the NT equivalent of a Unix panic().
(Of course, part of the joke is that BSODs occur so often.)

Linux has a very reliable kernel. :-)  To be fair, though, no
operating system -- or anything else -- can be expected to function
well on substandard hardware -- and this includes the power supply.
Might as well expect a bulb to function optimally with 40 V AC when
it was designed for 110.  Also, drivers within a kernel, if not
given especial care during development, can cause all sorts of
whackiness (after all, in a sense they're part of the kernel).
In an ideal OS, drivers would be separate, but a driver usually
needs special privileges to function -- access to kernel semaphores
and DMA, if nothing else.

>
>Bastian
>

-- 
[EMAIL PROTECTED] -- insert random misquote here

------------------------------

From: [EMAIL PROTECTED] (abraxas)
Subject: Re: Be & Linux & Microsoft...
Date: 14 Apr 2000 12:59:40 GMT

Erik Funkenbusch <[EMAIL PROTECTED]> wrote:

> You guys amaze me.  On one hand, you clamor that Microsoft is an unstoppable
> Monopoly that needs government intervention in order to allow the world to
> compete, then you spout that MS's dominance is fading fast and that it's
> only a matter of time before they're gone.

Apparantly you havent taken a logic class yet.

You really should.

No one ever called microsoft "an unstoppable monopoly".  It is a "monopoly
that must be stopped"---and it is being stopped by the US govt.

> Which is it?  Are they an unstoppable 

No.

> omnipotent 

No.

> monopoly 

Yes.

> that needs
> regulating, 

Yes.

> or are they a soon to be has-been that can't stay in business?

Hopefully.

Twisting your targets words and taking them out of context is a sure sign
of baseless and diluted argument.  




=====yttrx




------------------------------

From: [EMAIL PROTECTED] (CG)
Subject: Re: The truth is often painful...
Date: 14 Apr 2000 09:00:14 EDT

On Fri, 14 Apr 2000 04:01:39 +0000, mh <[EMAIL PROTECTED]> wrote:

>
>Please note that I am a Linux advocate.  My home network is Linux.  I
>believe in the principles, the ethics, of open source and admire the
>people who have made GNU/Linux a reality.  I use Netscape 4.7, both for
>email and news, not because I ACTUALLY PREFER Netscape, but because
>there is nothing better available on Linux.
>

I smell troll.  No linux user could actually form the opinion that
there is nothing better in linux for email or news than Netscape.  I
can think of loads of applications that do these jobs better.

------------------------------

From: [EMAIL PROTECTED] (abraxas)
Subject: Re: What GUI development tools are there for Linux?
Date: 14 Apr 2000 13:02:00 GMT

Pete Goodwin <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] (abraxas) wrote in <8d5imn$1p4j$[EMAIL PROTECTED]>:

>>Wow.
>>
>>Suddenly my response is alot less funny.  You should make it your
>>business to know what plan-9 was...in fact EVERY person who is 
>>interested in computers should make it their business to know what
>>plan-9 was.

> Why? What is plan-9 (for the second and last time)?

I cant believe you asked twice without going and looking for 
yourself.

It was something of an operating system.  It was scrapped.  Inferno
(which has also been scrapped) was loosely based on its architecture.




=====yttrx


------------------------------

From: [EMAIL PROTECTED]
Subject: Re: No Microsoft Certification = NO JOB!
Date: 14 Apr 2000 09:02:51 -0400

Charlie Ebert <[EMAIL PROTECTED]> writes:

> YET it is MY GOVERNMENT who has found MICROSOFT GUILTY.  And at the same
> time are continuing to develop
> their projects using ONLY VISUAL BASIC in EVERY job I've applied for
> from the City, thru the County, thru the State, thru
> the Federal Government itself!

So move to Massachusetts.

http://www.magnet.state.ma.us/hrd/ceo/

Plenty of state jobs here don't require MSFT experience.  I only noticed
one listing where MCSE was mentioned ("definitely a plus"), but it
wasn't a requirement even for those jobs that involved NT.  Caveat:  I
didn't look at every listing, just 5 or 6.

------------------------------

From: [EMAIL PROTECTED] (CG)
Subject: Re: The truth is often painful...
Date: 14 Apr 2000 09:09:21 EDT

On Fri, 14 Apr 2000 04:01:39 +0000, mh <[EMAIL PROTECTED]> wrote:


>I use Linux because I love its open nature, its flexibility, its
>empowerment, and because I believe in the philsophy/ethics of the Free
>Software Foundation.  I cannot contribute as a programmer, because I
>don't possess the skills.  


duh, where do you think the skills come from.  Pick up a book and
start reading.  I can't wait to see your contributions.


>But I can offer an honest evaluation based on
>years of experience with a wide range of business oriented desktop
>software in a business environment, 


oh, I get it, you're a bidni man.

>and the truth is that in this
>particular realm Linux SUCKS.  I fervently hope to see the day when that
>is no longer true.

troll.


------------------------------

From: "Drestin Black" <[EMAIL PROTECTED]>
Subject: Re: benchmark for speed in linux / windows
Date: Fri, 14 Apr 2000 09:02:03 -0400


"abraxas" <[EMAIL PROTECTED]> wrote in message
news:8ctvd5$1j53$[EMAIL PROTECTED]...
> Drestin Black <[EMAIL PROTECTED]> wrote:
> > Oh my god - and you think that your version is easier or better?
>
> > People - don't you even realize what you are typing? How in the WORLD do
you
> > consider that drivel easier or better? It takes longer to type in and
can be
> > typed wrong and ... oh never mind, you'll never learn.
>
> The point is, you only have to type it once.
>

what if you only need to use it once?
what if it's just too hard for a non-computer expert to
create/understand/use it?

honestly - my comment is: it's not that difficult for an expert but what
about the majority of computer users out there - they'll never have the
patience or desire to write such a line. Remember, PCs didn't take off until
Windows made them easier/accessible for everyone - the CLI held back
computing.



------------------------------

From: Hans Dumbrajs <[EMAIL PROTECTED]>
Crossposted-To: 
comp.mail.sendmail,comp.infosystems.www.servers.unix,comp.os.linux.security
Subject: Re: Sendmail refuses connection
Date: Fri, 14 Apr 2000 16:16:28 +0300

[EMAIL PROTECTED] wrote:
> 
> Could you kindly share your experience with me?
> I have some questions about sendmail on a DEC Unix server.
> 
> Why would sendmail stoped accept connections?
> I was wondering how to force sendmail to start accepting connections
> again.
> 
> I look forward to hearing from you. Thank you.
> 
> --
> Auction Booth:
> http://page.auctions.yahoo.com/booth/acunet3278
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Maybe the spool disk is full.


--
Hans Dumbrajs / [EMAIL PROTECTED]
Tel: +358-9-88176060
Fax: +358-3-31390898
GSM: +358-5-05941341
ICQ: 16859609

------------------------------

From: "Drestin Black" <[EMAIL PROTECTED]>
Subject: Re: Linux vs. Windows Benchmark
Date: Fri, 14 Apr 2000 09:08:48 -0400

"C't" = German "The Register"

'nuff said

"JEDIDIAH" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Mon, 10 Apr 2000 19:49:25 -0400, Drestin Black
<[EMAIL PROTECTED]> wrote:
> >
> >"Chris Lee" <[EMAIL PROTECTED]> wrote in message
> >news:8cl3u2$cei$[EMAIL PROTECTED]...
> >> In article <8ck0i7$q3b$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> >> >
> >> >In article <[EMAIL PROTECTED]>,
> >> >  [EMAIL PROTECTED] wrote:
> >> >> Ha ha...You can't!
> >> >>
> >> >> The Linux crowd runs and hides ever time the word "benchmark" is
> >> >> mentioned.
> >> >
> >> >This is a benchmark using NetBench 5.01 in a typical configuration
> >> >of both Linux and NT.
> >> >http://www.zdnet.com/sr/stories/issue/0,4537,2196106,00.html
> >> >
> >> >Microsoft spent 6 months trying to come up with a benchmark
> >> >that favored Microsoft and came out with the mindcraft benchmarks.
> >>
> >>
> >> We are going to be seeing fewer and fewer of these benchmark
comparisons
> >> bettween Linux and Windows in the future as Linux ports on hardware
like
> >the
> >> S/390's become widespread. Companies like Mindcraft and magazines like
PC
> >> Mag don't have the knowlege or the skills to deal with Linux and this
kind
> >> of hardware.
> >>
> >
> >oh - that is funny - that is REALLY funny. And, say, I thought linvocates
> >tell us that any system running Linux is so much cheaper (cause according
to
> >them the cost of the OS makes up 92% of the TCO of the system) that I'm
sure
> >PC Mag and Mindcraft can afford the hardware. And as for the knowledge or
> >skills - the last time Mindcraft invited the best Linux people to it's
own
> >labs linux STILL lost huge. Who do you need to tune linux before it'll
work;
> >a personal visit from linus?
>
> Then again, when C't decided to be remotely scientific about this
> whole sort of process, the picture radically changed.
>
> It really doesn't matter who gets on the playing field  when Microsoft
> is in the position to play Umpire. They set up the test to skew results
> in their favor.
>
> --
>
> It is not the advocates of free love and software
> that are the communists here , but rather those that        |||
> advocate or perpetuate the necessity of only using         / | \
> one option among many, like in some regime where
> product choice is a thing only seen in museums.
>
>               Need sane PPP docs? Try penguin.lvcm.com.




------------------------------

From: [EMAIL PROTECTED] (CG)
Subject: Re: uptime -> /dev/null
Date: 14 Apr 2000 09:20:00 EDT

On Fri, 14 Apr 2000 11:25:27 GMT, "Pedro Ballester"
<[EMAIL PROTECTED]> wrote:

>> No.  All electronic devices (and even things like car engines) will
>> (theorectially) last much longer if they are left on continuously (and
>properly
>> cooled/maintained of course, which in the case of an engine means you have
>to
>> turn it off to change the oil, filters, etc).
>> The act of turning devices on causes stress on an order of magnitude
>higher then
>> the stress it gets simply running. Why do you think flipping the power on
>and
>> off rapidly is bad for a machine?  Because you're giving it the surge of
>power
>> when it's turned on.  The act of turning your machine off each night just
>> spreads it out over time, but it still does the damage in smaller
>increments.
>> Someone else mentioned light bulbs.  How often have you seen a light just
>> suddenly go out?  How many times does the light go right when you flip it
>on?  I
>> know for me (and everyone I know) lights always (or almost always) die
>right
>> when you flip it on, due to the surge of electricity that is forced
>through it.
>
>   I'd like to see a great scale testing of computers being left on and
>computers
>switching on and off that gets this point clearer to me. I understand
>completly
>what you say and even agree in some points, but real personal experience
>tells
>me other thing; besides, don't forget that server hardware is more robust,
>while
>desktop/home users computers are not sold neither designed for continuous
>use.
>
>


ok, let's collaborate.  You can turn your computers on and off all day
long.  I'll leave mine on.  We'll compare notes in three years.  No
cheating please.

I have a couple of old P5-60s (gateways, when gateways were made
really well and came with intelligent docs) that have been running
more or less continuously since 1994.  They're (now) running linux, of
course, with no crashes (of course) for a year.

We reboot our windows machines daily as preventative maintenance.  I
also reboot my windows machine every time I change my mind to make
sure that the changes take effect.

------------------------------

From: [EMAIL PROTECTED] (CG)
Subject: Re: uptime -> /dev/null
Date: 14 Apr 2000 09:23:30 EDT

On Thu, 13 Apr 2000 22:58:32 GMT, "Pedro Ballester"
<[EMAIL PROTECTED]> wrote:

>
>   A new thought, have you ever heard about ecologism, global temperature
>raising, energy missspending and the such ? Yeah yeah, it sounds far, but
>it is here.
>
>


that's a valid point, but I save energy where it makes sense.  I don't
watch T.V. (I think the average american watches enough T.V. to power
a cray) and I take public transportation to work.  Also I don't own an
SUV, and I turn off the water when I'm brushing my teeth and shaving
my face.  :-)

------------------------------

From: Tesla Coil <[EMAIL PROTECTED]>
Subject: dvwssr.dll
Date: Fri, 14 Apr 2000 07:27:33 -0500

"NEW YORK (AP) Microsoft Corp. engineers included a
secret password in Internet software that could be used
to gain illegal access to hundreds of thousands of Web
sites, The Wall Street Journal reported today.
    "The rogue computer code was discovered in a three-
year-old piece of software by two security experts, the
newspaper said.  Contained within the code is a derisive
comment aimed at a Microsoft rival: ''Netscape engineers
are weenies!''
   "The file, called ''dvwssr.dll'' is installed on
Microsoft's Internet-server software with Frontpage 98
extensions.  A hacker [sic] may be able to gain access
to key Web site management files, which could in turn
provide a road map to such things as customer credit 
card numbers, The Journal reported."

ZDnet story at:
http://www.zdnet.com/zdnn/stories/news/0,4586,2543490,00.html?&_ref=1884210358


------------------------------

From: "Jim Ross" <[EMAIL PROTECTED]>
Subject: Re: Be vs. Linux
Date: Tue, 11 Apr 2000 00:16:19 -0400


JoeX1029 <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Be does own it's own OS.  BeOS 5 (the one for download) is a persoanl
edition
> and can't be used for commerical stuff.  If you want to use it
commerically you
> have to buy a license from Be.

I wonder how many do that?
I have a feeling the number is low.
The included browser isn't that great.

Jim Ross



------------------------------

From: "Mike" <[EMAIL PROTECTED]>
Subject: Re: What GUI development tools are there for Linux?
Date: Fri, 14 Apr 2000 13:47:02 GMT


"Donovan Rebbechi" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Fri, 14 Apr 2000 09:55:02 GMT, [EMAIL PROTECTED] wrote:
>
[...]
>
> No kidding. They essentially seem to blame Qt because Qt uses the same
> methods to print and draw ( you print by "painting" to a QPrinter ).
> This is one area where GNOME gets my vote -- if you saw the slashdot
> interview with Miguel, I asked him a question about this. His answer
> was interesting, and it's clearly an issue he ( and some other GNOME
> developers ) have really given some thought.

Donovan,

OK, how does GNOME handle printing? Last time I looked was quite some time
ago, and there didn't appear to be any printing system. I've always
preferred using the same methods to print and draw, since I only have to
code things once. If I'm reading you correctly, you're saying that GNOME has
(or will have) different methods. How is this better?

-- Mike --




------------------------------

From: [EMAIL PROTECTED] (Christopher Browne)
Subject: Re: What GUI development tools are there for Linux?
Reply-To: [EMAIL PROTECTED]
Date: Fri, 14 Apr 2000 13:54:46 GMT

Centuries ago, Nostradamus foresaw a time when abraxas would say:
>Pete Goodwin <[EMAIL PROTECTED]> wrote:
>> [EMAIL PROTECTED] (abraxas) wrote in <8d5imn$1p4j$[EMAIL PROTECTED]>:
>
>>>Wow.
>>>
>>>Suddenly my response is alot less funny.  You should make it your
>>>business to know what plan-9 was...in fact EVERY person who is 
>>>interested in computers should make it their business to know what
>>>plan-9 was.
>
>> Why? What is plan-9 (for the second and last time)?
>
>I cant believe you asked twice without going and looking for 
>yourself.
>
>It was something of an operating system.  It was scrapped.  Inferno
>(which has also been scrapped) was loosely based on its architecture.

For those so inept or disinterested that they are incapable of going to
<http://www.google.com/> and type "plan-9" into the form...

See <http://www.cs.bell-labs.com/plan9/faq.html>

This isn't rocket science; you go to a search engine, type in the name of
whatever it is you want to know _something_ about, and see what pops up.

It is quite distressing how so many seem so helpless to do their own
research...
-- 
Rules of the Evil Overlord #31. "No matter how well it would perform, I
will never construct any sort of machinery which is completely
indestructible except for one small and virtually inaccessible
vulnerable spot." 
<http://www.eviloverlord.com/lists/overlord.html>
[EMAIL PROTECTED] - - <http://www.hex.net/~cbbrowne/lsf.html>

------------------------------

Date: Fri, 14 Apr 2000 10:12:34 -0400
From: Gary Hallock <[EMAIL PROTECTED]>
Subject: Re: Corel Linux Office 2000 dubious at best?

Itchy wrote:

> My name did stay the same.
>
> Steve
>

I'm not talking about your signature.  I'm talking about the name the appears next
the post (heather69, Itchy).  It doesn't have to be the same as your user id.

Gary


------------------------------

From: [EMAIL PROTECTED] (Peter da Silva)
Crossposted-To: 
comp.unix.bsd.freebsd.misc,comp.unix.bsd.openbsd.misc,comp.unix.bsd.misc,comp.unix.bsd.netbsd.misc
Subject: Re: BSD & Linux
Date: 14 Apr 2000 13:59:49 GMT

In article <8d5tda$2jtk$[EMAIL PROTECTED]>,
Timothy Murphy <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] (Peter da Silva) writes:
> >> Minix was absolutely horrible.  Message passing like that is just wrong.

> >The problem is that Minix wasn't ever intended to be a particularly high
> >performance OS. 

> Minix worked pretty well, in my experience.

For some purposes it did. If you were running a single program, or otherwise
avoided contention for resources, it was fine. The problem was that the tasks
were pretty much all all single-threaded single-loop event handlers, so if
the file system task was blocked it quit accepting requests... even if it
would have been able to service those requests without blocking.

This is fine for a teaching system... going to multiple loops (via threads
or internal queues) would have significantly complicated the code.

That's where the monolithic kernel model wins big. A simple monolithic kernel
like Linux or traditional UNIX (including BSD) automatically gets a fresh
execution context for every blocking request (since the kernel executed in
the processes execution context on its behalf) so you don't get everyone
blocked behind a disk read.

It's a pity Tannenbaum couldn't or wouldn't completely open-source Minix back
then... removing the bottlenecks in the Minix kernel would have produced a
much more interesting system than just duplicating traditional UNIX did.

-- 
In hoc signo hack, Peter da Silva <[EMAIL PROTECTED]>
 `-_-'   Ar rug tú barróg ar do mhactíre inniu? 
  'U`    "Hint for long-term survival: be tasty, and farmable." -- Tanuki
         "And that's the real message of 'The Matrix'." -- Abigail

------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.advocacy) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Advocacy Digest
******************************

Reply via email to