Linux-Advocacy Digest #408, Volume #31 Fri, 12 Jan 01 02:13:05 EST
Contents:
Re: RPM Hell (BradyBear)
Re: Linux is easier to install than windows (Lewis Miller)
Re: Linux is easier to install than windows (Lewis Miller)
Re: Linux is crude and inconsistant (Lewis Miller)
Re: Linux 2.4 Major Advance (J Sloan)
Re: Linux 2.4 Major Advance (J Sloan)
Major culture clash, was Re: KDE Hell (TTK Ciar)
Re: Linux is crude and inconsistant (Lewis Miller)
Re: Linux 2.4 Major Advance (J Sloan)
Re: KDE Hell (J Sloan)
Re: TCO challenge: [was Linux 2.4 Major Advance] (Shane Phelps)
Re: kernel problems(new) (J Sloan)
Re: The real truth about NT (Pete Goodwin)
Re: Linux 2.4 Major Advance ([EMAIL PROTECTED])
Re: The real truth about NT (Pete Goodwin)
Re: You and Microsoft... ("Tom Wilson")
----------------------------------------------------------------------------
From: BradyBear <[EMAIL PROTECTED]>
Subject: Re: RPM Hell
Date: Fri, 12 Jan 2001 06:14:27 GMT
On Thu, 11 Jan 2001 10:10:15 GMT, "Tom Wilson"
<[EMAIL PROTECTED]> wrote:
>
>I wasn't refering to your post.
>
>I shouldn't have jumped in, though. Sorry to have upset you.
Apology graciously accepted.
_
BradyBear
------------------------------
From: [EMAIL PROTECTED] (Lewis Miller)
Crossposted-To: alt.linux.sux
Subject: Re: Linux is easier to install than windows
Date: 12 Jan 2001 06:15:43 GMT
Kyle Jacobs was heard ranting about
<_mt76.42644$[EMAIL PROTECTED]> in alt.linux.sux on 11 Jan
2001
>XFree86's autoprobe system doesn't seem to find USB devices very well
>(LIKE MICE?!)
Ok here we go, WTF do you need a USB mouse? that's as stupid as a USB
keyboard. Those devices have no need of a faster port than what they're
attached to. I hate all these devices they're attaching to a USB port just
because they can. A camera? That make sense. a camera can take advantage
of it. A mouse? please the mouse has not changed much in the past 10
years. What you need a boomslang?
NT can't use ANY USB devices.
>I've only seen XFree86 4.*'s autoprobe on my 3dfx card, which IS one of
>the top 10% of hardware distributed in theUS, I'd love to see what it
>does on a system with an ATI Radeon chip, or an old nVidia Riva 128.
>Does it "support" it through the generic svga "driver"?
umm. it detected an old STB Velocity 128. besides write your own driver.
:)
>Can it even probe for those devices AND accomidate WITHOUT prompting the
>user?
I would hope not, when my computer stops asking me before it does anything
I didn't tell it to do directly, I will be upset. Thanks I like to knoe
what is happening on my machine.
>Well, it sure can't get that USB mouse down.
>
><[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> On Thu, 11 Jan 2001 23:57:48 GMT, Kyle Jacobs
>> <[EMAIL PROTECTED]>
>wrote:
>> >I don't call the autoconfiguration system of Xfree86 4.0 to be
>substantial.
>> >It could be better, and yet, it isn't. WHY?
>>
>> How? It handles all aspects of the hardware in question.
>>
>> >
>> >Because they don't care? Possible...
>> >
>> >I mean it can sniff out the top 10% of peoples interface hardware,
>> >(with
>>
>> Even Xfree 3.3 can manage more than that. You are just
>> plain indulging in slander at this point.
>>
>> >exclusion of USB peripherals, even with backported kernels).
>>
>> Why would/should an X server care if something is USB even?
>>
>> >
>> >But the point is it still could be better. Find more devices and
>> >WORK
>with
>> >them!
>> >
>
>> This issue still remains addressed by you.
>>
>> --
>>
>> Freedom != Anarchy.
>>
>> Some must be "opressed" in order for their
>> actions not to oppress the rest of us.
>>
>> |||
>> / | \
>
>
>
--
l8r
-LJM
a.k.a. Jaster Mereel
a.k.a. MrBobaFett
"Little things used to mean so much to Shelly. I used to think
they were kind of trivial. Believe me, nothing's trivial. "
-- Eric Draven, The Crow
------------------------------
From: [EMAIL PROTECTED] (Lewis Miller)
Crossposted-To: alt.linux.sux
Subject: Re: Linux is easier to install than windows
Date: 12 Jan 2001 06:19:26 GMT
was heard ranting about <[EMAIL PROTECTED]> in
alt.linux.sux on 11 Jan 2001
>On Fri, 12 Jan 2001 01:47:38 GMT, "Kyle Jacobs"
><[EMAIL PROTECTED]> wrote:
>
>>XFree86's autoprobe system doesn't seem to find USB devices very well
>>(LIKE MICE?!)
>
>It can't even identify my PS/2 port Logitech Wheelmouse correctly
>(Mandrake 7.2).
Found mine fine, Red Hat 7. Tip don't use crapy distribs like Mandrake,
which I'm sure that will start a war. But yes I know Mandrake is based on
Red Hat, but they fuck the installer.
>>I've only seen XFree86 4.*'s autoprobe on my 3dfx card, which IS one of
>>the top 10% of hardware distributed in theUS, I'd love to see what it
>>does on a system with an ATI Radeon chip, or an old nVidia Riva 128.
>>Does it "support" it through the generic svga "driver"?
>
>It thinks my Matrox card has anywhere from 4 meg to 16 meg depending
>on which version of Xfree I use. The card has 8 meg BTW.
again, it identified my Matrox fine... didn't auto detect the 3D labs
card, but I tossed in a cheap STB got the drivers then installed it. Fine.
>>Can it even probe for those devices AND accomidate WITHOUT prompting
>>the user?
>
>A concept totally foreign to Linux users.
A concept totaly unwelcome.
--
l8r
-LJM
a.k.a. Jaster Mereel
a.k.a. MrBobaFett
"Little things used to mean so much to Shelly. I used to think
they were kind of trivial. Believe me, nothing's trivial. "
-- Eric Draven, The Crow
------------------------------
From: [EMAIL PROTECTED] (Lewis Miller)
Crossposted-To: alt.linux.sux
Subject: Re: Linux is crude and inconsistant
Date: 12 Jan 2001 06:26:47 GMT
Kyle Jacobs was heard ranting about
<eGr76.42226$[EMAIL PROTECTED]> in alt.linux.sux on 11 Jan
2001
>24/7 tinkering is a DESKTOP problem for me. I don't think I care about
>Linux as a server anymore,
WHAT?! Linux is totaly a Server OS. Notice how it comes with Apache (Web
Server), SendMail (Mail Server), Wu-FTP (FTP Server), TelnetD (Telnet
server), NameD (DNS Server), etc.. it's all about being a server.
Windows is a workstation OS.. and that's about the extent of what I want to
use it for.
>as that distro makers are beginning to realize that hard-customization of
>Linux for server purposes just works out better for admins and uptime
>anyway.
Right, well this is why I dislike ditros like Caldera, it's all about aimed
at being a workstation OS, for everybody. Please, Linux is NOT for
everyone. Nor should it be.
>Kudzu is RedHat based Linux's only, and Kudzu is USELESS if you upgrade
>ANY software component under Kudzu (i.e., new kernel, or new Xserver, or
>new ANYTHING that isn't directly from RedHat).
AH.. I see.. so get with it. :) Use Red Hat. one of the best Distros ever.
RPMs are great.
And I'm pretty sure I updated my Kernel before I did the video switch. and
Kudzu was still working.
--
l8r
-LJM
a.k.a. Jaster Mereel
a.k.a. MrBobaFett
"Little things used to mean so much to Shelly. I used to think
they were kind of trivial. Believe me, nothing's trivial. "
-- Eric Draven, The Crow
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux 2.4 Major Advance
Date: Fri, 12 Jan 2001 06:36:54 GMT
Jan Johanson wrote:
> I've been reading about noticable slowdowns with journalling in linux - ymmv
I've done more than just read - I've run reiserfs, logical volume
manager, etc under Linux.
> I'll have to take your word for it, I can't find it. Regardless, you didn't
> address the non-production portion or limited funcionality - face it, it's
> not in use in the real world.
Why wouldn't you want a fast, free web server in the real world?
Do you think maybe the fact that it's not in widespread use
may have something to do with the fact that it's brand new?
>
> Tunes? yes. Designed from the ground up to do nothing but beat SpecWeb? no,
> just Tux.
Please back up that assertion - this is the first I've heard
of such an idea - you admit that you can't find any information
about tux, but then you make this odd assertion about it.
Tux was written to be fast, period.
> > Again, you get the facts all wrong - it is not an "off the shelf" iis,
> > but a specially designed and tuned version unavailable outside of
> > microsoft labs. It is a special "service pack" which make the web
> > server run as part of the "kernel" - that's how desperate they are
> > to beat Linux, and they came close.
>
> Um, wrong. #1) show me where they did that?
This was discussed in this newgroup -
> #2) we discussed tuned, of
> course it was tuned, just as linux was tuned for this test to. However, the
> only thing "special" about the service pack is that it's service pack 2
> beta - it's not even fully tuned itself yet, and it'll be available to the
> public in Feb, free. But they most definately did NOT run in the kernel -
> that's reserved for a mode of IIS6. And that's what they didn't use here.
>
> ALSO, do you realize how you sound? Tux beat IIS - yep, by a whole
> smoldering 2.7% woo hoo!!! Wow!! 2.7% -
Tux beat iis by 250% -
This new, all out microsoft web effort is what almost matched tux,
not the iis we all know and love. Or did you think that iis suddenly
started running 300% faster thanbefore?
>
> >
> > Would I build my web server with a free Unix that's the world
> > speed champ, or would I shell out an exorbitant fee for a windows
> > pc webserver that can't quite match it in performance?
>
> Ummm... again, you don't stop to think before writing do you? Sure, linux
> itself was free to download --- but are you forgetting the hardware? Ooooh,
> you didn't quite notice that the price of the OS was almost insignificant
> compared to the price of the hardware? Anyone who can afford that hardware
> isn't going to blink for the price of the OS, especially with the savings
You're right about TCO, there's the hidden cost of windows
biting you in the ass again - thanks for bringing that up.
> Yes, despite being free, designed as a benchmark buster without any
> real-world proof of viability, lack of functionality, no security or
> journaling getting in the way of performance,
You clearly don't understand or didn't bother to read the
machine setups. microsoft pulled out all the stops - you
should go back and look at the disclosures a bit more -
Don't be such a knee jerk microsoft cheerleader,
> the barely released tux
> screamed past the vetern IIS by a WHOLE 2.7% - yahoo!
Nope, it screamed past the "veteran iis" by 250% - what
part of that are you having trouble understanding?
IOW, after months of frantic effort, the microsoft "benchmark
buster" finally managed to come close to the Linux result -
The kernel is still being readied - the Linux results will get
better, since thereal optimization hasn't started yet -
jjs
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux 2.4 Major Advance
Date: Fri, 12 Jan 2001 06:40:20 GMT
Jan Johanson wrote:
> SWC is not a web server. Get it through your head!
>
> AND, remember, Tux beat windows by a whopping 2.7%
You got the order of events wrong. Tux beat iis by more like 250% -
then, after months of frantic, all-out effort, the best microsoft could
do is come close to the Linux result with their new "benchmark buster"
product.
> - woo hoo!!! A whole 2.7%
> and they had to go into kernel space to do it.
Nope, tux ran in userspace for the specweb tests.
> I have never seen Tux in production, IIS (and SWC) is out there.
I've never seen swc, but Tux is available, for free - today.
jjs
------------------------------
From: TTK Ciar
Crossposted-To: alt.linux.sux
Subject: Major culture clash, was Re: KDE Hell
Date: 12 Jan 2001 03:55:17 GMT
Once upon a time, "Kyle Jacobs" <[EMAIL PROTECTED]> said:
>
>Here is an idea. If the individual has ANY desire to begin programming in
>ANYTHING they are...
>
>1.Serious; They are, and should be ready and willing to pay for programming
>tools. If they are programming for UNIX, they should have UNIX testing
>facilities available. Same with Windows, same with Mac.
How does a desire or ability to pay confer seriousness?
>2.Knowledgeable. This usually means enrolling at a college to take a
>programming course, which satisfies the educational value of a discount on
>Windows programming tools.
Why must someone know how to program before they can start learning
how to program? This strikes me as rediculous.
>3.Realistic. Programming something costs money.
How do these two terms go together? Programming something costs
time, and potentially scant resources. I am a senior Unix software
engineer, and a nontrivial chunk of my time is spent twiddling with
a pen and pad of paper.
I suspect that a very basic cultural difference is at the root of
our disagreement. I have no doubt that if you had siezed me when I
was seven years old (just before I started to program) and taught me
your doctrine of "if thou wants to program, thou must..", I would be
convinced of it and incapable of conceiving of things working any
differently. However, since I have been programming, both privately
and professionally, with free (ie: emacs, gcc) or very low-cost (ie:
pen, paper) tools for >75% of my life now, and since I taught myself
how to program long before taking any formal programming class, I am
utterly incapable of believing the things you have said here.
If an individual has ANY desire to begin programming in ANYTHING,
they need to:
* Find a book or a friend who can show them programming basics.
* Be willing to keep an open mind.
* Code like a rabid monkey -- experience breeds excellence. It
doesn't matter so much what projects they work on, as long as they
keep working on them.
* Keep learning -- keep talking to friends, keep reading books.
When they're old enough for formal programming education to become
available, they really should take advantage of it. For most people
this opportunity doesn't arise until highschool, though. A formal
education provides rigorous tools of analysis and representation, and
provides access to concepts which are not generally intuitive, and
thus tend to be passed over by the self-taught (dynamic programming,
for one).
Reading your post, where you exalt the tools of the trade over the
skills to use those tools reminds me of an anecdote from Chuck Yeager's
autobiography. He was staying at an airbase, flytesting the first
russian MiG the United States had captured, and he overheard one of
the other pilots brag about how he and his american Thunderchief could
out-dogfight the russian MiG any day of the week. Yeager made him a
small wager, and the two pilots took their jets up to test each others'
skills.
In the MiG, Yeager "waxed the tail" (ie, stayed up close behind the
other plane, which in combat would mean he could shoot the other plane
down) of the Thunderchief all over the sky. Then they landed and
swapped planes, and Yeager in the Thunderchief "waxed the tail" of the
MiG.
The moral of the anecdote, he said, is that piloting is nine parts
man and one part machine. I believe very strongly that programming is
the same way -- nine parts programmer, one part tools. The fancy toys
are fun, and can sometimes be useful, but they don't change the
fundamental task of formulating good code. That task is NOT done on
the computer with a text editor or integrated development environment;
it is done only in the human mind.
-- TTK
------------------------------
From: [EMAIL PROTECTED] (Lewis Miller)
Crossposted-To: alt.linux.sux
Subject: Re: Linux is crude and inconsistant
Date: 12 Jan 2001 06:39:43 GMT
Kyle Jacobs was heard ranting about
<8tt76.42646$[EMAIL PROTECTED]> in alt.linux.sux on 11 Jan
2001
>> Besides, anyone is free to fork a new distro based on the core
>> of Redhat anytime they like.
>
>It's not my job to provide the world with a better Linux.
It IS if you want a better version of Linux! That's the point of it. You
have to take responsiblity for you own OS, make it as pretty or Ugly, or
fast or slow as you want. Look if you don't want to do that. There are
plenty of other OSes out there you can pay for and they'll come all pre
packaged, and with phone support. Go get that and stop stumbling around in
Linux. It's a wonderfully powerfull and flexible OS. No it's not Easy.
It's not supposed to be. It's stick shift to automatic. More control more
possiblity for better preformance, yet also the chance you could easily
destroy the transmition comes with stick, or get an automatic and just
drive. If you don't like stick don't try to get rid of standard
transmitions, just get an automatic and let those who can or want, drive
stick.
>> Alternately, anyone is free to adapt kudzu to their particular
>> distribution.
>
>Kudzu seems to be pretty RedHat dependent, contact us when you get Kudzu
>for a Debian distro, or slack.
Read what he said. It can be adapted, YOU might have to do some work. You
don't want to work? Then you have to Pay, go buy Windows. or Irix, or
BSD/OS.
>> Furthermore, I have infact upgraded kernels and xservers and kudzu
>> continued to be useful. The really useful, core parts of what kudzu
>> does are not horribly version dependent.
>
>Aparently you manged to configure your new kernel exactly the same as
>the old one. Kudzu can only recognize what it THINKS the kernel
>supports, and act approperately with scripts. Those scripts tend not to
>work when a new revision of SOMETHING uses a completely different file
>format.
Or maybe, just maybe you don't know exactly how Kudzu works, since I doubt
you've gone out and grabed the source and looked at it. Because that's 2
ppl so far in here that it's worked for.
>> Besides, the need to run Redhat to get certain creature comforts
>> does not negate their existence. At most it would imply that
>> Redhat is ahead of the compeition.
>
>Wonderful. Except RedHat has some of the worst design problems found in
>a commercial distro. Remember "BugHat"?
Blah. :P I'm stickin with me Red Hat. I've got an Irix system too.. it's
cool, I've used Slackware, not that impressed.. Never liked Mandrake
>> So, all you've really demonstrated is that genuine choice and
>> competition are a good thing.
>
>Not really, so far they only seem to have lead to a load of incompatible
>Linux distro's with random and idiotic functionality problems, and
>general compatibility problems (FORKING!)
incompatable linux distros? how do you mean? I've never seen send mail
have a problem tossing between distros or anything like that. since that's
how the distros will see each other, over the network, they're not going to
be on the same machine.....
>> Common Standards, Common Ownership.
>>
>> The alternative only leads to destructive anti-capitalist
>> and anti-democratic monopolies.
>> |||
>> / | \
>
>
>
--
l8r
-LJM
a.k.a. Jaster Mereel
a.k.a. MrBobaFett
"Little things used to mean so much to Shelly. I used to think
they were kind of trivial. Believe me, nothing's trivial. "
-- Eric Draven, The Crow
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux 2.4 Major Advance
Date: Fri, 12 Jan 2001 06:44:20 GMT
Adam Warner wrote:
> Check out the statement here:
> http://www.microsoft.com/windows/serverappliance/webserver/overview.asp
>
> ---Begin Quote---
>
> Industry-Leading Hosting Performance
>
> Based on SPECWeb 99, the leading benchmark for Web server performance, Web
> Server Appliances powered by Windows 2000 deliver significantly greater
> hosting capacity than alternative Web server solutions. Recent SpecWeb test
> results showed single-processor machines built on Windows 2000 handled 75%
> more simultaneous connections than servers on non-Windows platforms.
>
> ---End Quote---
Talk about misleading!
Notice how they misspelled "couldn't match Linux performance"?
jjs
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Crossposted-To: alt.linux.sux
Subject: Re: KDE Hell
Date: Fri, 12 Jan 2001 06:46:39 GMT
Kyle Jacobs wrote:
> I guess you haven't read what Wired, AND CNET both said about the new
2.4
> kernel's SMP support.
I guess you haven't read what specweb 99 said about the
new 2.4 kernel's SMP support -
BTW real world results count more than any cluless IT
"pundits" you may wish to trot out.
jjs
------------------------------
From: Shane Phelps <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: TCO challenge: [was Linux 2.4 Major Advance]
Date: Fri, 12 Jan 2001 17:47:15 +1100
Jan Johanson wrote:
>
> "J Sloan" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
[ snip ]
> > Would I build my web server with a free Unix that's the world
> > speed champ, or would I shell out an exorbitant fee for a windows
> > pc webserver that can't quite match it in performance?
>
> Ummm... again, you don't stop to think before writing do you? Sure, linux
> itself was free to download --- but are you forgetting the hardware? Ooooh,
> you didn't quite notice that the price of the OS was almost insignificant
> compared to the price of the hardware? Anyone who can afford that hardware
> isn't going to blink for the price of the OS, especially with the savings
> down the road in TCO.
>
Alright - I'll bite. I continually see assertions about the lower TCO
of NT x compared to unix, but I'm yet to see a credible study.
Put up or shut up (and that includes Erik, Chad Nulligan, et al)
Note that does not include MS's NT vs Linux TCO sham, which is actually
a poll of Solaris on SPARC sites vs NT sites - no details of any sort.
BTW, I don't especially trust any vendor's TCO fgures. Sun's are dubious
... but MS's are farcical!
and as a parting shot - the NT box should have had *much* better disk
performance, but a kernel-mode webserver strikes me as a benchmark
special. Read that as you may.
[ snip ]
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Subject: Re: kernel problems(new)
Date: Fri, 12 Jan 2001 06:49:46 GMT
"Joseph T. Adams" wrote:
> mlw <[EMAIL PROTECTED]> wrote:
>
> : I have attached both my Makefile and a config file. It builds damn near
> : everything and it boots. Copy the Makefile to your kernel directory, and
> : the load the config with xconfig. I am using RedHat Linux 7.0 and using
> : gcc 2.96.
>
> : No guarantees as to if it will work for you, or even produce a stable
> : kernel.
>
> In RH7, aren't you supposed to use kgcc for compiling the kernel?
I've been compiling 2.4 kernels with Red Hat's gcc with no
problems - the systems include a celeron 600, an amd K6/2
450, and a quad pentium pro.
jjs
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: The real truth about NT
Reply-To: [EMAIL PROTECTED]
Date: Fri, 12 Jan 2001 07:01:56 +0000
Aaron R. Kulkis wrote:
> Write once....and then it can't be used again.
> Capacity ... less than 1G.
>
> 4mm DAT
>
> Write once....it's still good for several hundred RE-recordings
> Capacity... 2G - 25G
CD-R has two big advantages:
(i) You can use it on any machine with a CDROM
(ii) It has a random access file system
Tapes generally aren't available across several machines, unless they've
all got the same tape drive attached.
--
Pete, running KDE2 on Linux Mandrake 7.2
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux 2.4 Major Advance
Date: Fri, 12 Jan 2001 06:52:51 GMT
Jan, what you say might be partially true, but Microsoft's own SpecWeb99
submission page at
http://www.spec.org/osg/web99/results/res2000q4/web99-20001211-00082.html
tells us a different story. Check out the "dynamic API" field, it says: "TWC
3.0". IIS 5.0's dynamic API is ISAPI, so Microsoft did not even use IIS 5.0
for the dynamic portion of the test! SWC 3.0 is a kernel-space webcache, with
its own dynamic API. The more I look at it, the worse it gets: this
submission tells tales about Microsoft's (non-) confidence in the performance
and scalability of their own web products: IIS 5.0 and ISAPI. In fact this
tells tales about the (non-) scalability and design of Windows 2000 WIN32
system-calls, Microsoft had to move a new webserver (available in March) into
the kernel to (not quite) reach Linux 2.4 (available today) performance
levels! Another question bothers me. SWC 2.0 has been available for more than
four years, but why are there no websites using SWC 2.0? Is there any problem
with it? Not even Microsoft uses SWC on their own website! Maybe it does not
integrate well with IIS environments? (No shared logfiles maybe?) A search
for "SWC" on www.microsoft.com gives 4 (!) hits only. No Knowledge Base hits
at all, which means that either SWC is Microsoft's most bugfree product ever,
or that nobody (including Microsoft) tests and uses SWC.
Thomas
> SWC 2.0 is afront-end cache for Web request operations with IIS running on
> Windows 2000 Server. The design goals are based on IIS's basic data transfer
> algorithm with certain performance principles that take significant
> advantage of Windows 2000 Server scalability in a symmetric multiprocessor
> (SMP) environment. SWC 2.0 aims at complete partitioning in an SMP
> environment. The concept of a "mapping" is used to bind a network address to
> certain threads that run on specific processors. While SWC 2.0 was designed
> for SMP environments, it also provides significant performance gains on
> single-processor servers.
>
> SWC 3 is an upgraded version which is why the release date for the HTTPD
> software is listed as available.
>
> It's a simple web server caching product - except that it's designed for
> really high performance servers. What's wrong with that? IIS is the web
> server, SWC is a cache, nothing more.
>
> <[EMAIL PROTECTED]> wrote in message
> news:93l7m9$ql8$[EMAIL PROTECTED]...
> > Adam, the answer is simple - Microsoft did not use IIS 5.0 at all,
> Microsoft
> > used 'SWC 3.0'. This is clearly visible in the submission form. SWC I
> believe
> > means "Scalable Web Cache", which is a kernel-space HTTP accelerator
> (sounds
> > familiar?). I'm not aware of any sites running SWC (although there are a
> few
> > SPECweb96 results done with SWC). I can only see a March 2001 availability
> > date. SWC is probably a non-production, benchmark-only, closed-source,
> > specialized kernel-space webserver - IIS 5.0 is just the bumper sticker.
> This
> > submission is Microsoft's admission that IIS 5.0 cannot scale. I'm not
> aware
> > of any production sites using SWC 3.0. (or 2.0.) Thomas
> >
> > >
> > > > Have you even seen the 4Q results of SPECWeb99 ?
> > > > Tux just barely got the higher score, and it's a web server *no one*
> use.
> > > > IIS got just behind Tux, and it's a commonly used webserver.
> > >
> > > Thank you for alerting me to those results.
> > >
> > > I see the results you are referring to are here:
> > > http://www.spec.org/osg/web99/results/res2000q4/
> > >
> > > Specifically the Dell PowerEdge 8450/700. Windows 2000 Datacenter Server
> > > only loses to Red Hat : TUX 2.0 by a small margin.
> > >
> > > Do you know how Microsoft improved the results so much? (Also possibly
> > > helped by the fact that Linux still might not scale as well as the
> number of
> > > processors is increased?)
> > >
> > > Regards,
> > > Adam
> > >
> > >
> >
> >
> > Sent via Deja.com
> > http://www.deja.com/
>
>
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: The real truth about NT
Reply-To: [EMAIL PROTECTED]
Date: Fri, 12 Jan 2001 07:03:40 +0000
[EMAIL PROTECTED] wrote:
> Can't belief? I have a Cheapbytes copy of RedHat 6.2. Everything I need
> is there except StarOffice, which I have on another CD and on the
> Network.
You still don't mention the list of applications. Therefore I consign your
replies to the dustbin.
--
Pete, running KDE2 on Linux Mandrake 7.2
------------------------------
From: "Tom Wilson" <[EMAIL PROTECTED]>
Subject: Re: You and Microsoft...
Date: Fri, 12 Jan 2001 07:03:03 GMT
"Charlie Ebert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> In article <08f76.2258$[EMAIL PROTECTED]>,
> Tom Wilson wrote:
> >
> >"Erik Funkenbusch" <[EMAIL PROTECTED]> wrote in message
> >news:uGd76.288$[EMAIL PROTECTED]...
> >> "Nigel Feltham" <[EMAIL PROTECTED]> wrote in message
> >> news:93in2m$adklg$[EMAIL PROTECTED]...
> >> > >The Windows setup files are all 8.3 conformant. We were talking
about
> >> > using
> >> > >a network card, not a modem.
> >> >
> >> > I thought we were talking about installing from the internet so both
> >> > netcards and modems are relevent here.
> >>
> >> As if installing Linux via modem is feasible.
> >
> >I have great admiration for anyone with patience enough to install Linux
> >from a dial-up connection. I know I couldn't do it. I got impatient
> >downloading the Windows 98 Beta image from a T1 connection.
> >
> >--
> >Tom Wilson
> >Sunbelt Software Solutions
> >
> >
>
> Well thanks but I mainly just had a beer and went to bed.
>
> Then we ate breakfast and when we finished we answered
> a few config questions which apt-get left us and I
> showed the farmer his new os.
One of my goals in life is to be that laid back.
Haven't been able to pull it off yet.
--
Tom Wilson
Sunbelt Software Solutions
------------------------------
** 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 by posting to comp.os.linux.advocacy.
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
******************************
