Linux-Advocacy Digest #511, Volume #28 Sun, 20 Aug 00 02:13:04 EDT
Contents:
Re: Anonymous Wintrolls and Authentic Linvocates - Re: R.E. Ballard says Linux
growth stagnating ("Mike Byrns")
Re: Would a M$ Voluntary Split Save It? (Stephen S. Edwards II)
Re: Anonymous Wintrolls and Authentic Linvocates - Re: R.E. Ballard says Linux
growth stagnating ("Mike Byrns")
Re: Linus says Mindcraft was accurate (Tim Hanson)
Re: Whats a usenet troll? (Tim Hanson)
Re: refrigerator using Linux? (Tim Hanson)
Re: Windows blows (Tim Hanson)
Re: Anonymous Wintrolls and Authentic Linvocates - Re: R.E. Ballard says Linux
growth stagnating ("Mike Byrns")
----------------------------------------------------------------------------
From: "Mike Byrns" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Anonymous Wintrolls and Authentic Linvocates - Re: R.E. Ballard says
Linux growth stagnating
Date: Sun, 20 Aug 2000 05:33:18 GMT
"Craig Kelley" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Mike Byrns" <[EMAIL PROTECTED]> writes:
>
> > "Craig Kelley" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > "Mike Byrns" <[EMAIL PROTECTED]> writes:
> > >
> > > [snip]
> >
> > Not suprisingly, those points that cannot be rebutted were snipped.
I'll
> > declare victory on those. :-) Not trying to be an ass but there were
alot
> > of advantages there that were not answered...
>
> If it makes you feel better, by all means. :) This isn't a war after
> all.
Agree. Thank you. Claim of victory retracted. You are a gentleman and a
scholar. I apologize.
> > > > In many cases programmers forgo implementing memory protection
between
> > > > threads but that's not the fault of the OS. It's the fault of the
> > > > programmer. It's there if you want to take advantage of it.
> > >
> > > Yes, but we have a choice under Linux of whether we want to
> > > significantly add to our program's bulk, or to just use the one-line
> > > fork() call.
> >
> > CreateThread is a one line call. Once you wrap the VirtualProtect et.
al.
> > functions in accessor functions they are one line calls. It's personal
> > choice.
>
> Well, all calls are one line; it's the _number_ of one-line calls that
> you have to make that is a bother.
I use my wrapped calls constantly. They are in MSBThreadUtils.lib. It's
copyrighted :-) I know that goes against the GPL but I make enough from it
to take a cruise every couple years. You can license it, ah well
nevermind...
> > I find the advantages of easy interthread local procedure calls,
> > thread pooling and functional encapsulation to result in much less
bloated
> > code. I also like using structured exception handling to catch
unhandled
> > thread exceptions in a manager thread instead of in the OS. I guess
it's
> > really all in how you learned it. I've been a Win32 programmer since
the
> > first beta SDK shipped on a single 1.2MB (5 1/4 inch remember those :-)
> > floppy. Before that I did <gasp> DOS and Windows API (as it was called
> > then). I've had threads as long as I've had pre-emptive multitasking.
It's
> > just a natural choice for me.
>
> That's great; everyone has their own comfort zone when it comes to
> programming. I try to push mine, but It's always easier to fall back
> on what I know. I started programming on the Commodore Pet, of all
> things :)
And I on the C64. Welcome brother! I did assembler with HESMon and FORTH
as well as the nasty standard BASIC. I had a my Dad's stock photo service
running on a C64 with 3 1541s as storage. I've even contributed code to
KMail to fix their leave mail on server bugs. I've programmed for Windows,
SUN, PL/SQL Oracle and Macintosh. I always came back to Windows because it
was a richer environment.
> > > Forking is fairly scalable, but not as scalable as
> > > threads in most situations.
> >
> > I'll say. Windows thread stacks exist in the apps virtual address space
so
> > you can have over 2000 per app at the default stack size. You should
never
> > need to though. Optimal design on Windows warrants optimized code in a
pool
> > of worker threads equal in number to the number of CPUs on the server.
Plus
> > whatever manager or UI threads you might need. Since I can accept and
> > schedule multiple incoming TCP/IP connections on a single thread and
> > dispatch them to a worker thread as the queue allows, I can clear
> > connections faster than with the overhead of the standard UNIX way of
> > forking for every connection.
>
> Perhaps, depending on the machine's architecture. A Linux forking
> daemon can outperform an NT threadded daemon quite easily on a
> single-processor machine. The threaded daemons start wiping the floor
> when more processors are added, though.
That's called scalability. Linux beats the pants off Windows 2000 on a 386.
So what? :-) OK... I concede that too :-)
> So, if you're writing a quick app to do some network connections, it
> really isn't worth coding in manager threads passing off connections
> to worker threads when you can just fork()... (see my point?)
Yah, I do. I just use VB for prototypes like that. Heck, if it's going to
production I'll have to recode it anyway...
> > That's OK for the first couple dozen
> > connections but then you start really using the system resources up
after
> > that. My connections might wait slightly longer to be serviced but they
get
> > serviced so much faster that Windows tends to come out ahead. For
lengthy
> > streams processes like FTP I don't block in the thread since Windows
allows
> > me to "connect" a file handle to a pipe and forget about it unless it's
done
> > or there's an exception. This makes implementing FTP or HTTP servers
that
> > support command pipelining *so* simple. :-)
>
> I think we are violently agreeing on all of this...
Cool. So call it even?
------------------------------
From: [EMAIL PROTECTED] (Stephen S. Edwards II)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy
Subject: Re: Would a M$ Voluntary Split Save It?
Date: 20 Aug 2000 05:39:46 GMT
[EMAIL PROTECTED] (Christopher Smith) wrote in
<8nnp7k$lqr$[EMAIL PROTECTED]>:
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Christopher Smith in comp.os.linux.advocacy;
>> >"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>> >> I sincerely doubt everyone in any country agrees on any point.
>> >> What's your point? Yes, governments (and citizens) around the
>> >> world are telling MS to "ESAD". Why is it you can't seem to handle
>> >> the fact that this is because Microsoft is a criminal organization
>> >> which has retarded development and innovation in the PC world to
>> >> their own private ends
>for
>> >> many years?
>> >1. The citizens, thus far, are voting with their feet in great
>> >favour of Microsoft.
>>
>> I'm sorry, that's delusional. The citizens are suffering under a
>> monopoly, which is why the case was brought.
>
>The "citizens", as you put it, are buying Windows because there is no
>other viable alternative. It is not Microsoft's fault no-one else has
>been able to develop an OS desirable to customers.
It never ceases to amaze me how many people fail
to realize that if Microsoft went away, so would
personal computing.
Why hasn't any other software company done what
Microsoft has done with software engineering?
Using the old lame-assed worn-out excuses like,
"because their software is buggy, insecure, and
bloated" isn't sufficient. Someone please explain
to me how these other whiney software companies
expect to do the same kind of business that Microsoft
does when they don't even offer the same kind of
products.
It's exactly like a bunch of people in the desert
selling sand, getting pissed off at the one vendor
who is smart enough to be selling water.
8<SNIP>8
>> But you can't hand-wave MS's
>> attention-grabbing activities by insisting blindly that all other
>> companies "do the same thing". It is nothing more than an unsupported
>> assertion at best, and at worst it is an unfalsifiable contention.
>
>At least one example of another company springs instantly to mind -
>Apple and the Mac clones.
And it's uncanny how easily they seem to forget
about Intel and IBM in all of this "monopoly" speak.
Facts and observations of convenience are anything but.
8<SNIP>8
>> That is a post-modernist bit of clap-trappery, hiding an
>> argument from ignorance behind an incredibly thin veil of objectivity.
>
>Your constant assertions that I argue from a position of ignorance are
>amusing me greatly, thinking back to previous discussions about
>multitasking and file permissions.
Is it just me, or do these guys seem to operate
out of some sort of universal anti-Microsoft
handbook?
>> It is why so many people recognize your position as either entirely
>> incorrect or purposefully wrong.
>
>"So many" ? I'd be fairly willing to bet you'll find no more who
>disagree, than agree.
And yet, he referred to your stance as "delusional".
Please Max, point out these alleged "many people"
who are recognizing Christopher to be wrong.
>> >Then again, what else does the law have if not circumstantial
>> >evidence,
>open
>> >to different interpretations ?
>>
>> Reasoned argument, free from general ignorance of the circumstances.
>
>Pfft. The law is chock full of poor reasoning and just plain stupidity.
Laws of any kind are brought about "because" of stupidity in
the first place. It's an attempt to provide a low-level means
for those who have no common sense or ethic to operate in a
civil society.
>Remember the laws that determined how to decide whether a woman was a
>witch ? I'm sure they were considered reasoned and free from ignorance
>at the time.
Just as were the laws that allowed hospitals to
deny black people medical care of any kind.
What I find hypocritical is that the "hacker"
types who propose to be against the government
for the most part, all of a sudden think that
Janet Reno is so highly commendable, simply
because she is going after the one thing that
they seem to hate the most... Microsoft.
These people are just flakey, and their positions
are just as fleeting as their arguments are.
--
.-----.
|[_] | Stephen S. Edwards II | http://www.primenet.com/~rakmount/
| = :| -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
| -| "You are a waste of space; a disgrace to your profession;
| | both the one you claim and the kindergarten student you
|_..._| act like..." -- Robert Moir to Aaron R. Kulkis in COMNA
------------------------------
From: "Mike Byrns" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Anonymous Wintrolls and Authentic Linvocates - Re: R.E. Ballard says
Linux growth stagnating
Date: Sun, 20 Aug 2000 05:40:44 GMT
"Craig Kelley" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Mike Byrns" <[EMAIL PROTECTED]> writes:
>
> > "Craig Kelley" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > <[EMAIL PROTECTED]> writes:
> > > >
> > > > > Yes, but we have a choice under Linux of whether we want to
> > > > > significantly add to our program's bulk, or to just use the
one-line
> > > > > fork() call. Forking is fairly scalable, but not as scalable as
> > > > > threads in most situations. The problem is, 90% of the time (my
time,
> > > > > anyway) you don't care if the process is extremely scalable and
you
> > > > > can ditch a bunch of complexity by using processes instead.
> > > >
> > > > And if using fork() ends up consuming too many resources, you can
just
> > fork
> > > > out for a more powerful platform! <g> Sorry, just couldn't resist.
> > >
> > > Yep, Linux does provide a good path to powerful OSes like AIX.
> >
> > Would this be the same AIX that I've got on my WebSphere farm?
>
> I have no idea. I'm talking about the AIX that with 24 450Mhz
> processors got half the TPC-C performance that Windows 2000 did with
> 96 700Mhz processors.
Price comparison please? :-) I really like AIX it's at least got a semi
consistant GUI and some decent management tools. I just think I could do
better with Windows for less
> > The same one that performs 90% slower than the single CPU Windows
> > 2000 development box? The same AIX that I've had IBM on site TWICE
> > in the last month to fix so it retains eCommerce session data when
> > OSE protocol app server load balancing is enabled? AIX may be
> > powerful but it sure doesn't look like it on MY RS/6000 rack...
> >
> > BTW, you wouldn't have any clue as to what's wrong with it would
> > you? ;-)
>
> Nope, we still run AIX 3 over here. :)
>
> Something about paying $20k for an OS upgrade got under our skin.
Yah. We did it though for other reasons. Can WebSphere run on AIX3? DO
you know? I think we are licensed for older OSs still. Maybe that's a way
out.
------------------------------
From: Tim Hanson <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linus says Mindcraft was accurate
Date: Sun, 20 Aug 2000 05:50:36 GMT
"Stephen S. Edwards II" wrote:
>
> [cross-posted to comp.os.linux.advocacy
> as a point of possible interest]
>
> >From an interview in the Wall Street Journal:
> >
> >http://interactive.wsj.com/public/current/articles/SB966551209980435711.h
> >tm
> >
> >What's more, Microsoft last year began to take notice of Linux. In the
> >spring, for example, it released a study showing that Windows NT
> >outperformed Linux in some common computing tasks.
> >
> >The study caused an uproar in the Linux camp, with Mr. Torvalds and
> >others suggesting at the time that they had somehow been rigged.
> >
> >Subsequent tests, though, showed Microsoft was right, and in his
> >interview, Mr. Torvalds conceded that he initially had been "in denial"
> >on the matter.
> >
> >"We had been arrogant," he said, adding it was painful for him to admit
> >that Windows was better than Linux, at least in the areas covered by the
> >Microsoft test.
>
> I don't even think "better" is a proper term, because
> "better" is subjective. I think "performed faster"
> is adequate, and it helps alleviate the whole
> "penis comparison" that benchmarks tend to degrade
> into.
>
> I think that this clearly makes Linus the better
> man, as it's not easy for anyone to admit when
> they were incorrect, especially when they have
> a lot of admiration and integrity on the line.
>
> Such humility is quite deserving of grand respect, IMHO.
> It was this kind of attitude that I respected Linus for
> in the past.
>
> I think what people need to understand, is that these
> different operating systems will never ever be the one
> true OS. Just use what suits you, and suggest to others
> that they might like what you like, but if they don't
> then leave them alone. And if they find reason to
> list a bunch of convoluted reasons why "your choice
> sucks", then most likely, they'll be full of crap,
> and you can pick them apart, and make them into
> fools at your leisure.
>
> I think it's perfectly acceptable to state that you
> don't like a certain software product, or OS, and
> leave it at that, but I think it's childish for
> someone to go out of their way to explain why
> something is empiracally inferior, just so that
> they can feel better about themselves.
>
> >Now, will the linux apologists finally admit they were wrong?
>
> They don't even have to do that, IMHO. I would've
> just liked to have heard them say "gee, that's
> interesting. I'd like to verify that, and if it's
> true, find out why it turned out that way."
I think it's unfortunate Linus gave the Mindcraft "study" as much
credence as he did, given the fact that it was a benchmark designed to
exploit the strengths of NT and the weaknesses of Linux from the gitgo.
As such it was merely a marketing exercise, and a sure loser for Linux
advocates. I wasn't there, but I would have held out for a more
balanced set of tests, as well as a balance of relative cost.
--
The 80's -- when you can't tell hairstyles from chemotherapy.
------------------------------
From: Tim Hanson <[EMAIL PROTECTED]>
Subject: Re: Whats a usenet troll?
Date: Sun, 20 Aug 2000 05:56:08 GMT
mark wrote:
>
> In article <399e3ce4$[EMAIL PROTECTED]>, bloke wrote:
> >hey people,
> >im a bit of a news group newbie and was wondering what is a troll in the
> >context of news groups?
> >im not losing any sleep or anything.....just curious
> >cya's
> >
> >
> >> Someone who tries to start an argument
>
> > No it's not, you piece of sh*t
>
> Yes it is - and I can prove it (but I won't now)...
Mostly, it's people who resort to gratuitous name-calling, like you,
idiot. :-)
--
The 80's -- when you can't tell hairstyles from chemotherapy.
------------------------------
From: Tim Hanson <[EMAIL PROTECTED]>
Subject: Re: refrigerator using Linux?
Date: Sun, 20 Aug 2000 06:01:00 GMT
Dan Jacobson wrote:
>
> What's this I read that even a refrigerator could be using Linux? To help me
>visualize this I've seen how a PC can control a
> modem, printer, GPS [Global positioning system]... so let's say the PC was shrunk
>down to fit into the refrigerator control panel
> brains; I suppose upon booting [i.e., when the AC cord is plugged in] there would
>be an /etc/profile type script that would
> immediately throw the computer into a loop controlling the refrigerator... ok, got
>that, now how is it that Linux is superior to
> Micro$oft stuff here? couldn't the same be achieved with autoexec.bat? Or is there
>some kind of scalability impossible with the M$
> stuff? Is the bonus to controlling wristwatches with Linux found in portability to
>different watches hardware, where in the past
> one would program for a single watch type? Again is what is there miraculous in
>this new Linux way other than portability vs.
> assembler code specific solutions? Could watch companies have equally banded on to
>a "Micro$oft standard", or does M$ stuff die
> due to being hardcoded to find C: and not finding it? etc.
> --
> www.geocities.com/jidanni ... fix e-mail address to reply; ¿n¤¦¥§
> Tel:+886-4-5854780; starting in year 2001: +886-4-25854780
Actually, what I'd like would be for it to replace the notes under all
those magnets on the door.
--
The 80's -- when you can't tell hairstyles from chemotherapy.
------------------------------
From: Tim Hanson <[EMAIL PROTECTED]>
Subject: Re: Windows blows
Date: Sun, 20 Aug 2000 06:02:14 GMT
root wrote:
>
> Windows sucks dick.. every single Microsoft operating system is complete
> garbage.. Anybody stupid enough to bet their business on Microshaft
> products deserves what they get.
>
> -Barry
This might be an example for the "What is a troll" thread.
--
The 80's -- when you can't tell hairstyles from chemotherapy.
------------------------------
From: "Mike Byrns" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Anonymous Wintrolls and Authentic Linvocates - Re: R.E. Ballard says
Linux growth stagnating
Date: Sun, 20 Aug 2000 06:09:00 GMT
"Gary Hallock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Mike Byrns wrote:
>
> >
> > Would this be the same AIX that I've got on my WebSphere farm? The same
one
> > that performs 90% slower than the single CPU Windows 2000 development
box?
> > The same AIX that I've had IBM on site TWICE in the last month to fix so
it
> > retains eCommerce session data when OSE protocol app server load
balancing
> > is enabled? AIX may be powerful but it sure doesn't look like it on MY
> > RS/6000 rack...
> >
> > BTW, you wouldn't have any clue as to what's wrong with it would you?
;-)
>
> Comparing AIX and Windows 2000 performance is meaningless without knowing
what
> kind of hardware you are running on. Is AIX running on an RS/6000 model
320
> with 32MB of ram or on a dual processor model 260 with 1GB of ram, or
perhaps a
> 4-way 604e base machine?
Eight 2 way 604e's, B50's with 512MB each to be exact. Behind dual
fault-tolerant Cisco LocalDirectors and dual PIX. App servers are H70's
with a GB or more. I've tested and found that they can outperform the 300GL
(Windows 2000 box) for static web serving but...
Why can't they do what a $1000 workstation do? I know this is a IBM problem
but this kinda thing plauges IS depts worldwide. Windows works out of the
box more times than UNIX does in my experience.
And the cost differential is starting to raise heads in corporate. Less
cost for more performance is compelling!
------------------------------
** 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
******************************
