Linux-Advocacy Digest #630, Volume #34 Sat, 19 May 01 21:13:03 EDT
Contents:
Re: Linux posts #1 TPC-H result (W2K still better) ("Les Mikesell")
Re: Rather humorous posting on news.com commentry forum: (Charlie Ebert)
Re: Linux posts #1 TPC-H result (W2K still better) ("Les Mikesell")
Re: Linux posts #1 TPC-H result (W2K still better) (Charlie Ebert)
Re: EXTRA EXTRA MS ADMITS!!!! (Charlie Ebert)
Re: Linux still not ready for home use. ("Mad.Scientist")
Re: Justice Department LOVES Microsoft! ("Daniel Johnson")
Re: Justice Department LOVES Microsoft! ("Daniel Johnson")
Re: W2K/IIS proves itself over Linux/Tux ("Les Mikesell")
Re: My plan worked! ("Mad.Scientist")
----------------------------------------------------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux posts #1 TPC-H result (W2K still better)
Date: Sun, 20 May 2001 00:20:21 GMT
"Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
news:9e6r4u$3ul$[EMAIL PROTECTED]...
>
> "Les Mikesell" <[EMAIL PROTECTED]> wrote in message
> news:1%yN6.917$[EMAIL PROTECTED]...
>
> > I don't think you've looked at PostgreSQL recently. It is too bad
that
> > MS and other commercial DB vendors prohibit publishing benchmarks
> > that they don't control or you would probably know how well it compares.
>
> Do you've any idea how *easy* it is to cheat in such a benchmark?
> Twist one parameter, see how performance drop.
Of course. That's why the vendors insist that they are the only
ones allowed to do the cheating. However, what the customers
want to know is how it will work when they run it themselves.
> The only way you could have anything close to a fair benchmark is when
each
> DB company sumbit their highest results.
> Something like TPC does.
The TPC was really designed for commercial vendors and has requirements
that are unfair to free programs. Does that mean we shouldn't be able
to find out how the performance compares?
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Subject: Re: Rather humorous posting on news.com commentry forum:
Reply-To: [EMAIL PROTECTED]
Date: Sun, 20 May 2001 00:22:45 GMT
In article <hO8N6.14172$[EMAIL PROTECTED]>,
Flacco wrote:
>> I HATE YOU FLACCO!
>
>That's a little extreme, isn't it?
Oh well, I was just covering the last exchange we had.
HA. Hello Flacco!
--
Charlie
=======
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux posts #1 TPC-H result (W2K still better)
Date: Sun, 20 May 2001 00:23:58 GMT
"Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
news:9e6sc2$568$[EMAIL PROTECTED]...
>
> > >
> > > > > > > Perl & Python from activestate.com (free). C#, VB.NET comes
with
> > > .NET
> > > > > > > beta, and there are also other languages that you can hook
> there,
> > I
> > > > > > > believe.
> > > > > >
> > > > > > Sounds better than it was, though with UNIX, you can use an
> > arbitrary
> > > > > > executable as the interpreter.
> > > > >
> > > > > You can do the same in Windows, what is your point?
> > > >
> > > > How do you make a .bat file interpret itself with perl and pass some
> > > > arguments as it starts? Under unix, making the first line:
> > > > #!/usr/bin/perl -w
> > > > would make perl execute it and turn on warnings.
> > >
> > > I'm not sure what you meant here. Can you be clearer?
> >
> > Unix looks for
> > #!/path/to/interpreter args_to_interpreter
> > at the top of an executable file so you can control
> > which shell or other interpreter parses it and force
> > it to always have certain arguments. This is a
> > general facility that works with any program that
> > reads from stdin. Any executable file can use this
> > to automatically have itself read by any other program.
> > For example:
> > ===
> > #!/bin/cat
> > test
> > ===
> > as an executable file would invoke /bin/cat, which simply
> > copies stdin to stdout and would thus print the file contents
> > to your screen (just to show that no special concepts are
> > involved for the program invoked).
> >
> > How do you do the same in Windows?
>
> Usually, it's not needed, because you use extentions to mark where it
should
> go.
> I suppose you can build an interaptor that would interapt the first line
> without too much trouble, is should take only couple of minutes to do it
in
> C.
>
> Again, that is now how you would do it in Windows.
> The usual thing to do is to register an extention and put the
> \path\to\interpreter args_to_interpreter as the openner of this file.
How do you make this setting follow the file as you copy
it to different machines or execute it over a network? Or
provide arguments to the interpreter that might be unique to
each file?
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux posts #1 TPC-H result (W2K still better)
Reply-To: [EMAIL PROTECTED]
Date: Sun, 20 May 2001 00:29:51 GMT
In article <[EMAIL PROTECTED]>, Chris Ahlstrom wrote:
>Jan Johanson wrote:
>> >
>> > AFAIK, no (except maybe Perl)
>> > There is a regex COM object, RegExp I think, that is accessible to just
>> > about anytihng in Windows, inclusing WSH lanaguages.
>>
>> Actually - yes it does:
>> http://msdn.microsoft.com/library/dotnet/cpguide/cpconintroductiontoregulare
>> xpressions.htm
>
>What a great link! Nice marketing, no meat.
>
>--
This is the exact same problem I just had with Mathew Gardiner.
As a trial attorney this kind of evidence would suck.
Yet there is no point on arguing with the blind.
It's kind of like shooting rabbits with a shotgun.
No sport in it.
--
Charlie
=======
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Subject: Re: EXTRA EXTRA MS ADMITS!!!!
Reply-To: [EMAIL PROTECTED]
Date: Sun, 20 May 2001 00:40:43 GMT
>
>After some considerable reading at HP and other web sites... it appears
>that the PA-RISC can execute IA-64 instructions unmodified. HP has had
>a large hand in its development with Intel. That's about all I could
>find out about it. There will always be development chips available
>from Intel in small quantities, but from past experience the properties
>may change or come more closely to what Intel wanted all along...
>currently the IA-64 has some speed problems with certain programs having
>to do with the predictive jump feature... can't remember all of what
>I've found on other web sites, but it appears to be its biggest hurdle.
>
>--
Oh don't do this. You don't want to hurt poor little
Erik Fuckenbush and Matthew Gardiners feelings.
I mean, after all, if you go read thru Intel's own web site
they badmouth RISC in every article.
You might also notice that both EF and MG are not disputing
that HP-9000 is currently using an Intel chip.
If Intel makes a risc chip then why are they badmouthing
RISC all over the Itanium link? They really trashed it
bad.
When you start talking about 64 bit processors from Intel
you get people refering to 3 seperate family lines from
Intel. All of these lines came from the IA-64 idea.
To be honest with you at this very moment I've lost
some of my confidence in this issue and I'm going
to drop it. Not because I believe EF or MG is right
and not because I feel I'm right either.
I just simply don't seem to understand why it is
EF and MG keep hollering it's a RISC chip from
the WEB site at HP and yet Intel claims they've
never made a RISC chip and that RISC CHIPS are
garbage?
These two ideas here don't JIVE. Then there's
this issue of instructions set compatibility
as the instruction sets are the same....
To my knowledge, Motorola and IBM were the only
ones to do the RISC business. They were the
ones who started it, not Intel.
--
Charlie
=======
------------------------------
From: "Mad.Scientist" <[EMAIL PROTECTED]>
Subject: Re: Linux still not ready for home use.
Date: Fri, 11 May 2001 19:17:32 -0500
I use Linux (LM 8.0 to be exact) and it works very well as a end user OS.
"Chaparral" <[EMAIL PROTECTED]> wrote in message
news:9daq5c$9bs$[EMAIL PROTECTED]...
> We can all say what we want about how Windows sucks and that Linux is the
> end-all-be-all, but after trying almost every Linux version to date, the
> bottom line folks is that Mr Gates has made operating a home computer easy
> enough for my great uncle to run. The Penguin still doesnt come close!
>
> What Linux is VERY good at is the handling of servers... this is stuff
that
> you are expected to fiddle with and fine tune. Home users don't want to
> fart around all day trying to figure out what to click and then having
> barely predictable responses.
>
> So, Linux sucks hard for the home user but beats the hell out of WinBlows
on
> the server farm... especially when you can tell a client that full-blown
> server software will only cost him $75 compared to $2000 plus for
> 2000Server!
>
> Microsoft will rule the home front for many years I think, but their
> exorbitant pricing and draconian licensing policies will soon cause the
> server market to dry up.
>
> Im done now.
>
>
------------------------------
From: "Daniel Johnson" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Date: Sun, 20 May 2001 00:44:53 GMT
"Rick" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Daniel Johnson wrote:
[snip]
> > Seriously, they may well buy an app
> > because their co-workers use it and
> > they want to share data or otherwise
> > collaborate.
>
> ... they use what everyone else uses.
Well, what their co-workers use anyway.
> > But they buy the OS so they can run that
> > app.
>
> ... they use what everyone else uses.
... and for a reason. It's the reason that matters,
but you seem averse to discussing it.
Why is that?
[snip]
> > Anyway, it's kind of hard for me to search for
> > them if you can't even give me the machine's
> > model number, or an author, or something.
>
> You profess to be such an expert on things and you dont even know the
> TRS 80 model numbers? I, II, III, 4, 4P
Those guys weren't CP/M machines. They
had a thing called TRS-DOS, I believe.
They were kinda weak even for 8-bit computers;
they certainly were not better than IBM PCs.
> > And the chance that such an old article would
> > appear on the web is pretty small. I'd have
> > to, like, use a library or something. :D
>
> Too lazy to look, huh?
I really can't, not without more than what
you've given me.
[snip]
> Damned typos. I have told you this beofer, but now you seize on the
> typo.. here's the correction:
> m$-DO$ was very close to CP/M. Thats why IBM paid Killdal $800,000
> dollar so he wouldn't sue over the CP/M code in it.
Oh.
Sure, they probably thought that if he hadn't
patented something in CP/M, we surely could.
MS-DOS and CP/M were just awful close.
Same problem with Windows NT and VMS.
[snip]
> > On the 68000. The instruction set of this
> > computer was admirably forward looking, and
> > made it very much easier to move to 32 bits
> > in, oh, 1987- years before the PC did it.
>
> IIRC the The Mac OS was 24 bit from the beginning.
24 bit address spaces, yes, but that was typical
for 16-bit computers. The 8086 was unusual in
having only 20 bits. The 80286 fixed it, and
other chips like the 65816 had 24 bit addressing
too.
What was unusual about that 68000 was that
the instruction set made it look like you had
32-bits of addresses. You didn't, and the MacOS
took advantage of that, but it was good thing in
terms of forwards compatibility.
> > But from a performance and memory standpoint,
> > the 68000 was a 16-bit computer. True, it
> > was a far less awful one than the 8086; it could
> > access 16MB of memory not just 1MB; but
> > at the time few could afford even 1MB, and
> > early Macs were well below that line.
>
> Well, if the 68000 could only access 1 meg of memory, why was the Plus
> able to use 4 megs - not constrained by the 68000, but by the Mac
> hardware?
The 68000 could access 16 megs, not 1. That was
par for the course for 16 bit CPUs of that era.
What was unusal was that it did it with
no segments.
The reason early Macs had so much less
memory was that it was expensive back
then.
[snip]
> > > What does that matter? It worked well for me.
> >
> > That's what I'm trying to explain: what
> > matters is what works well for developers.
>
> Users dont care what worls well for developers. And developers will do
> what it takes to make money. When will you get that through your head?
Developers know that users will follow the apps.
They can and they do program for non-dominant
platforms- when they gain an advantage thereby.
> > You may love AppleWorks, but if Apple
> > hadn't writen it nobody else would have;
> > too much effort compared to doing the
> > same things on the PC.
>
> Hmmm. IIRC, Appleworks was commissioned. The Apple II also had
> AppleWriter, which was a programable word processor.
The very first on a PC, as I recall. Very primitive stuff,
but groundbreaking in concept.
How was it "programmable"? I didn't think it had
a macro system.
> BTW, you filed to
> answer the question that you snipped. What was missing from Appleworks
> that others DID have... in THAT time frame?
In 1987? I think I covered that. If you have
some other timeframe you'd like to talk about,
I need to know what it is.
> Rupert Lissner wrote it. Apple marketed it,
Put some new life in the Apple II, it did, but
that kind of thing is the exception, not the rule.
Even then it was limited by the Apple II's
capabilities.
[snip]
> > > It wasnt horrid AT THE TIME.
> >
> > Yes it was. "Real" computers had been
> > doing *vastly* better since the '60s.
>
> The sould you here is passive-aggressive dDaniel moving the damned goal
> posts again. "real computers"?? "Real ones"?? Like WHAT. C64? Tandy
> Model II? Color Computer? SOL? Altair? We are talking "personal
> computers" here. So answer the question about Appleworjs in that light.
No, compared to *real* computers. Even compared to
little ones like PDP-11s. Compared to a System 360s,
PCs were jokes.
The PCs got better. The introduction of the IBM PC
was a big step in the right direction.
I'm not moving the goalposts. I'm telling you that
the problems faced by programmers were
recognizable even at the time.
> > It was awful. But it was cheap.
>
> It was NOT awlful. It addressed over 1 meg of ram. It had integrated WP,
> SS and DB.
It certainly did not *address* 1 meg of anything; the 6502
had an address space of 64k, period. It used bank switching
to get beyond that.
AppleWorks is a neat product, but it's not really a
first class anything. That's the problem integrated
suites always have.
But then, AppleWorks wasn't terribly integrated.
The modules didn't work together very much.
[snip]
> > > What substantial features did Appleworks need in 1987, that Appleworks
> > > didnt have?
> >
> > It didn't have anything like WYSIWYG;
> > the Macintosh was already well established
>
> No the Mac wasnt already "established" when Appleworks shipped.
> Appleworks shipped in 1984.
You said 1987. In 1987 the Mac was well
established.
In 1984 the Mac existed and could do things
AppleWorks could not, already.
> > with this feature by then. AppleWorks couldn't
> > compete.
> >
> > Its database was a poor joke. Flat file, and only
> > one file at a time. No programming to speak of.
> > No external access.
>
> So what? there is nothing wrond with flat file DBs if you dont need
> realtional tables.
It's weak. It's like an address book.
The Apple II really couldn't do much
better, of course. But that doesn't
mean it wasn't weak. Better stuff
was available by 1987 on the PC;
DBase, I think, was out.
> > It didn't provide much of anything in terms of
> > integration. You couldn't put a spreadsheet in
> > a word processing document, not even statically.
>
> Yes, you could. And Mail merges. And reports.
I know you could do mail merge. I doubt
it could do reports, but if you say so.
I think you are mistaken about embedding
spreadsheets, though.
> > It had no graphics module at all.
>
> IIRC, third party ADD-on.
Oh? What was it called? I may
look it up.
Apple IIs had really seirous issues
with supporting graphics. The
display hardware was not much,
not even compared to other 8 bit
computers.
(Though at release the IBM PC actually
was *worse*, believe it or not. It's one reason
why the Mac, which was very good at that, got
products like Photoshop, and the PC didn't.)
> > It would have to hit the disk switching
> > between modules as I recall, and those
> > disks were not real fast. Integrated
> > packages on better computers did not
> > have to do that.
>
> What "better" computers?
Macs. PCs. Amigas.
> > Compare with the later Macintosh ClarisWorks
> > program. ClarisWorks simply makes AppleWorks
> > look like a joke. It shows how big a difference
> > chosing the right platform makes.
>
> Apples/oranges... goal posts moving.
No, really ClarisWorks is a integrated
package that fills a market nice very
similar to that of AppleWorks; a lightway
productivity suite that is very easy to
use.
ClarisWorks is *much* better at it.
[snip]
> > > Except I WAS running Appleworks with 1 meg of RAM and almost all of it
> > > was accessible to Appleworks.
> >
> > Not directly. AppleWorks was one of the few
> > programs that would do bank switching to get
> > to it.
>
> Except I WAS running Appleworks with 1 meg of RAM and almost all of it
> was accessible to Appleworks. And the IIRC, there was a utility to allow
> other apps to accesss the bank switched memory. Pinpoint?
There is no way to make apps use back-switched
memory unless they already know how, not at the
Apple II with its rigid address space map.
Other apps did use it, but it wasn't common.
The same situation as on PCs at that early
stage, except that they could get up to 640k
before needing to bank switch. And they
had a bigger bank of memory to switch.
[snip]
> > It wasn't used much. USCD Pascal worked, but it worked
> > by emiting what we now call bytecodes to be interpreted
> > at runtime; the interpreter could overcome the memory
> > limits to some extent by paging to disk!
> >
>
> Pascal wasnt used much? For the Apple II? HAhahahahhahahah...
> You+credibility=0
Well, it was good for playing around, and
there was a famous game written using it-
the original Wizardy.
But it was *visibly* slowed by using
Pascal. Other similar games came after
that were dramatically faster and better
looking- and they used assembly to do it.
It's a competitive advantage thing; if
you use assembly (on the 8-bit computers)
you produced visibly better product.
[snip]
> > Rick, you are a piece of work, you are.
>
> YEah, Well, at least I am capable of changing my mind when presented
> with facts.
When did that last happen? :D
------------------------------
From: "Daniel Johnson" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Date: Sun, 20 May 2001 00:50:50 GMT
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Daniel Johnson in comp.os.linux.advocacy on Sat, 19 May 2001
> >> Yup. People buy "what everybody else has". micro$oft made sure what
> >> everybody else had is micro$oft.
> >
> >You sure it isn't to deny consumers any alternative
> >choices?
>
> Whatever pretend grammar mistake you had to make to pretend the point
> wasn't made obviously requires some explanation if you expect anyone
> else to repeat it, Daniel.
I see no grammar mistake here. Rick seems to
feel that Microsoft has founded a monopoly,
so that they can, er, have a monopoly.
Not founded for power, or to increase
prices, or any of the other nefarious things
I have heard people accuse MS of.
I think he has overstated his position.
> In point of fact, people do not buy "what everybody else has", they buy
> what is best for them. Sometimes that is the same choice as others,
> sometimes it is not. Unless there is illegal monopolization going on
> (and, yes, the fact that this happens alone is sufficient evidence for a
> conviction), then everybody makes the same choice.
I do not give your legal weight much credit, but I find
your reasoning here quite provocative.
You seem to understand that end-users are not
idiots; why do you think they react to this
"monopolization" by all buying Microsoft?
> Unreasonable restraint of trade is unethical and illegal, Daniel; no
> amount of trolling changes that, or could possibly refute the
> correctness of the law in this regard.
We haven't really tried to argue about that yet; I
suspect we can't do so until you tell us all what
you mean by "unreasonable restraint of trade".
Coming from you, I expect it to be quite creative;
please don't disappoint me. :D
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Sun, 20 May 2001 00:53:33 GMT
"Jon Johansan" <[EMAIL PROTECTED]> wrote in message
news:3b06dfb3$0$12213$[EMAIL PROTECTED]...
>
> >
> > There is that MS commercial on TV about servers that haven't been
> > touched in 'days' as though that should be a surprisingly long time.
> > Real OS's run for years without any attention. And they don't
> > pop up dialog boxes and stop and wait like IIS 5.0 does when
> > an error occurs.
> >
>
> Amazing that I've never ever seen an IIS box do that, 4 or 5 - and yet you
> seem to make it sound like they all do... agenda ??
All of mine do it since I moved what a pair of Linux boxes used to do
with Apache/mod_perl and some custom programming over to a set of 8
Win2k/IIS5 boxes handling the same job with ASP pages that use
xml/xsl formatting. The reason for the switch is mostly that the
xml data comes from an in-house product and service that we sell
and we wanted to use the same technology on our own web site.
I had run a pair of Win2k/IIS5 boxes serving only static images
for a few months before attempting the dynamic part and they
ran fine then. Why would I need an agenda to describe what
happens in actual operation? I'd much prefer that it wasn't
happening - it is wasting a lot of my time.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: "Mad.Scientist" <[EMAIL PROTECTED]>
Subject: Re: My plan worked!
Date: Fri, 11 May 2001 19:35:07 -0500
Ummm, how are you going to killfile yourself?
"Public <Anonymous_Account>" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Now that I know who all the fags are in this group, I have you kill filed,
and you don't know who I am!
>
> :)
>
>
>
>
>
> ---
> This message did not originate from the Sender address above.
> It was posted with the use of anonymizing software at
> http://anon.xg.nu
> ---
>
>
------------------------------
** 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
******************************
