Linux-Advocacy Digest #442, Volume #34 Fri, 11 May 01 23:13:02 EDT
Contents:
Re: Justice Department LOVES Microsoft! (GreyCloud)
Re: Justice Department LOVES Microsoft! (GreyCloud)
Re: Justice Department LOVES Microsoft! (GreyCloud)
Re: Microsoft standards... (was Re: Windows 2000 - It is a crappy product) ("Chad
Myers")
Re: Richard Stallman what a tosser, and lies about free software (Isaac)
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
----------------------------------------------------------------------------
From: GreyCloud <[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: Fri, 11 May 2001 19:57:29 -0700
Ayende Rahien wrote:
>
> "T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Said Ayende Rahien in comp.os.linux.advocacy on Thu, 10 May 2001
>
> > >Seriously, though, I think that it's a primitive API.
> >
> > Is there some taxonomy of APIs within the engineering community which I
> > am unaware of? If not, you're just begging the question, I think.
> >
> > Which is fine, as long as you say "I do not think it is really an API,
> > although it is a primitive form of API". At least then we know the
> > metaphysical ground you are standing on; where API's "in the wild" can
> > be captured and domesticated and categorized.
> >
> > Let me ask you something; did anyone ever call DOS interrupts "an API"
> > at the time DOS was prevalent? Or is this just hindsight that enables
> > you to ascertain the morphology of APIs?
> >
>
> No one called the first automobils cars, but they are certainly primitive
> sort of a car.
> It fits into the defination of API, so it's an API.
> It doesn't fit into the same category as most APIs today, so I called it
> primitve API.
Pretty soon you guys are going to call your hands and feet APIs.
--
V
------------------------------
From: GreyCloud <[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: Fri, 11 May 2001 19:59:30 -0700
Dave Martel wrote:
>
> On Fri, 11 May 2001 00:15:03 -0700, GreyCloud <[EMAIL PROTECTED]>
> wrote:
>
> >Dave Martel wrote:
> >>
> >> On Thu, 10 May 2001 17:21:53 -0700, GreyCloud <[EMAIL PROTECTED]>
> >> wrote:
> >>
> >> >> Let me ask you something; did anyone ever call DOS interrupts "an API"
> >> >> at the time DOS was prevalent? Or is this just hindsight that enables
> >> >> you to ascertain the morphology of APIs?
> >> >>
> >> >> --
> >> >> T. Max Devlin
> >> >> *** The best way to convince another is
> >> >> to state your case moderately and
> >> >> accurately. - Benjamin Franklin ***
> >> >
> >> >I've looked all over my documentation sets from MS dating back to 1987.
> >> >No such wording back then about APIs. I'm beginning to thing MS has
> >> >been re-painting their old horse a new color is all. Somehow, the
> >> >semantics are being changed.
> >>
> >> Don't know how official it was, but I remember a time about 12 years
> >> ago when "DOS API" was a common term. I've long since thrown it out
> >> but also used to have a book entitled something like "A Programmer's
> >> Guide to the DOS API".
> >
> >That is unfortunate for me. I have never seen the term "API" until long
> >after I retired. I thought when I saw it that it meant the GUI form of
> >a routine call.
> >I guess I'm just out of date is all.
>
> I think the current term for people like us is, "obsolete old
> dinosaurs". <g>
Well, if it weren't for us 'dinosaurs' these kids wouldn't have their
PCs. :-)
--
V
------------------------------
From: GreyCloud <[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: Fri, 11 May 2001 20:04:06 -0700
"T. Max Devlin" wrote:
>
> Said GreyCloud in comp.os.linux.advocacy on Thu, 10 May 2001 17:21:53
> >"T. Max Devlin" wrote:
> >>
> >> Said Ayende Rahien in comp.os.linux.advocacy on Thu, 10 May 2001
> >> >"Tom Wilson" <[EMAIL PROTECTED]> wrote in message
> >> >news:xEtK6.83$[EMAIL PROTECTED]...
> >> >>
> >> >> "Ayende Rahien" <[EMAIL PROTECTED]> wrote in message
> >> >
> >> >> > > Tell us then of *your* criteria a body of functions must meet to be
> >> >> > > classified as such...
> >> >> >
> >> >> > I asked him that before, couple of times, so far he refused to answer.
> >> >>
> >> >> I simply take him at his word that he's not a programmer when these little
> >> >> asides occur. I enjoy the arguments, though.
> >> >>
> >> >> Oh what the hell, what's *your* definitive opinion, as a programmer, as to
> >> >> 21h calls?
> >> >
> >> >That I'm glad to get rid of them :-)
> >> >
> >> >Seriously, though, I think that it's a primitive API.
> >>
> >> Is there some taxonomy of APIs within the engineering community which I
> >> am unaware of? If not, you're just begging the question, I think.
> >>
> >> Which is fine, as long as you say "I do not think it is really an API,
> >> although it is a primitive form of API". At least then we know the
> >> metaphysical ground you are standing on; where API's "in the wild" can
> >> be captured and domesticated and categorized.
> >>
> >> Let me ask you something; did anyone ever call DOS interrupts "an API"
> >> at the time DOS was prevalent? Or is this just hindsight that enables
> >> you to ascertain the morphology of APIs?
> >>
> >> --
> >> T. Max Devlin
> >> *** The best way to convince another is
> >> to state your case moderately and
> >> accurately. - Benjamin Franklin ***
> >
> >I've looked all over my documentation sets from MS dating back to 1987.
> >No such wording back then about APIs. I'm beginning to thing MS has
> >been re-painting their old horse a new color is all. Somehow, the
> >semantics are being changed.
>
> No, not at all. The term API simply didn't exist until "platforms"
> (really, the roots of middleware) became more commonly the basis of an
> application, as general purpose computers went from specialty use to
> consumer use. It *would* have been entirely appropriate at the time to
> call DOS an API, if the phrase meant anything to enough people. It
> didn't, though; program developers were still writing to "bare metal",
> on DOS, essentially, and "on the OS" everywhere else.
>
> This is the root of the FSF's position on library linking, in fact; it's
> all well and good to talk about "writing to an API" when there's
> actually a replaceable library on the other side of the interface. But
> when the only thing on the other side is just the bare metal or bios or
> ONE SPECIFIC AND PARTICULAR library, talking about APIs is really just
> nonsense outside the question of which function call to use for what,
> and that is certainly nothing that anyone outside the program author
> really cares about at all. Whether it is described as "calls in/to the
> API" or "calls to/functions of the library" is just semantic quibbling.
>
> --
> T. Max Devlin
> *** The best way to convince another is
> to state your case moderately and
> accurately. - Benjamin Franklin ***
That is my consensus as well. But, the new generation of kids are now in
charge and its what they like to call it. I won't quibble with them
just as long as they do it right.
--
V
------------------------------
From: "Chad Myers" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Microsoft standards... (was Re: Windows 2000 - It is a crappy product)
Date: Sat, 12 May 2001 02:51:04 GMT
"Joseph T. Adams" <[EMAIL PROTECTED]> wrote in message
news:9dgabr$2a0$[EMAIL PROTECTED]...
> In comp.os.linux.advocacy Chris Ahlstrom <[EMAIL PROTECTED]> wrote:
> : Jan Johanson wrote:
> :>
> :> Perhaps when you graduate from the hell that telnet is - I mean, how
> :> limiting! - into the year 2000 and beyond you'll realize how silly your
> :> hanging on to the past is. I mean, all this worship of telnet and the
> :> command line. You sound like an Amish person swearing off technology! I
> :> can't remember the last time I needed the command line or even felt the
urge
> :> to time travel into days of old...
>
> : I'd fire up X-Windows over ssh if I could find a decent free (ie.
> : no corporate timed-out software or shareware)
> : X-Windows server for Windows.
>
> Have you tried VNC?
Oh you mean the bloaty, resource pig that is a gaping security hole
and slowly degrades the performance of any machine its on by leaking
memory like a seive?
Yeah, I tried it. It sucks =)
-c
------------------------------
From: [EMAIL PROTECTED] (Isaac)
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Date: Sat, 12 May 2001 02:52:17 GMT
On Thu, 10 May 2001 23:03:40 GMT, T. Max Devlin <[EMAIL PROTECTED]> wrote:
>I agree with your description of the situation; the FSF, with no
>evidence of support in copyright law, claims that they can demand that
>others comply with their political agenda. I'm willing agree to that
A sentence like this months ago would have probably ended much of the
discussion. Is this a change in position or is it somehow consistent
with your previously posted positions?
I don't have any issues with the FSF's political agenda, so I guess
I don't have anything left to discuss... Maybe some more copyright
law that I disagree with...
Wait a minute, what's that down below?
>
>The limits are copyright. You are not allowed to use an author's work
That word "use" is just too slippery. We need to be more precise if
we are trying to say that copyright is the limit. You agree that use
doesn't mean read but why wouldn't you be allowed to "use the code" in all
ways other than those reserved by statute for the copyright holder. Those
uses are broad in one sense, but in a real sense they are very narrow
because both the copyright holders rights and the limitations of those
rights are enumerated in the copyright statute.
>It is this definition of 'derivative' which is what you claim to be an
>extra-legal extension of copyright. This is a conceptual glitch, I know
>realize. I was committing it myself, for quite a while, until very
>recently. It is the assumption that the term 'derivative' in the
>license is necessarily identical to the term 'derivative' within the
>statute of copyright. This could well be a mistake, though.
You are near the heart of the matter here. I think the GPL uses
the term "based on", and while the GPL says that it is equivalent
to 'derivative works' as defined in copyright law, it appears to me
that the usage appears to include derivative works and compilations
with some exceptions. The problem is really that the works we
are discussing are not derivative works according to copyright law,
and if they are compilations, copyright holders can only restrict the
distribution of their portions of a compilation.
>The GPL is a private contract. Although it conforms to copyright law in
>being a license agreement for copyright material, it is NOT copyright
>law itself. It is a private contract. So your observation that the
And how is the contract formed? You seem to agree that you have not
become obligated to the contract simply by reading the source code. I
don't see how there can be any sign of agreement or obligation until
you either attempt to do something which law prohibits and the license
allows, or until you manifest your assent and acceptance to the
agreement in some other fashion.
Creating and distributing a derivative work as defined in copyright law
certainly does the trick, as would calling up RMS and telling him you
agree to his terms. But if neither of those things are done, how
does the license become binding? There isn't even the highly
questionable shrinkwrap to discuss here. Consideration is even
more problematic.
IMO it's as if I wrote out an IOU with your name on it and then tried
to collect. I wouldn't expect it to work. Similarly, the FSF can
print up all the contracts they want, but somehow they need to bind
parties to them before those parties need to care about the contents
of the contract.
Isaac
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Fri, 11 May 2001 17:38:37 +0200
"Jan Johanson" <[EMAIL PROTECTED]> wrote in message
news:3afc9ad7$0$41685$[EMAIL PROTECTED]...
>
> "Matthew Gardiner" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > > Here are a couple of Win2K servers that stayed up for a long time.
> > >
> > > http://uptime.netcraft.com/up/graph?site=partnering3.microsoft.com
> > > 244
> > >
>
http://uptime.netcraft.com/up/graph?site=download.windowsbeta.microsoft.com
> > > 216
> > > http://uptime.netcraft.com/up/graph?site=msdnisv.microsoft.com
> > > 189
> > >
>
http://uptime.netcraft.com/up/graph?site=corporate.windowsupdate.microsoft.c
> > > om
> > > 189
> > > http://uptime.netcraft.com/up/graph?site=esl.one.microsoft.com
> > > 184
> > They are all clusters. Now, get, one, lone server loaded with Win2k
> > Server, and then see the uptime.
> >
> > Matthew Gardiner
>
> I have one lone server loaded with W2K Server that has been running
non-stop
> since Feb 17th 2000. It was rebooted one single time when SP1 was
released,
> intentionally obviously, and never since. It has 100% uptime during the
> first period and continues 100% at this time.
>
> Are you starting to understand? W2K is reliable.
Time for another reboot, SP2 is here.
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Fri, 11 May 2001 17:40:31 +0200
"Jan Johanson" <[EMAIL PROTECTED]> wrote in message
news:3afc9bb7$0$41698$[EMAIL PROTECTED]...
>
> "T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Said Ayende Rahien in comp.os.linux.advocacy on Fri, 11 May 2001
> > >"Matthew Gardiner" <[EMAIL PROTECTED]> wrote in message
> > >news:[EMAIL PROTECTED]...
> > >> > Can't speak of uptime, because it's usually to expensive (and long)
> to
> > >> > benchmark those.
> > >> > But according to TCP.ORG, in the unclustered category, Win2K win on
> > >> > price/performance.
> > >> > On unclustered/clustered category, Win2K wins *both*
> price/performance &
> > >> > performance.
> > >>
> > >> Support is more expensive on UNIX boxes, however, that is off set my
> the
> > >> reduced amount of downtime, hence the reason why the New Zealand
> > >> financial system runs on big irons.
> > >
> > >Interesting, I keep hearing about TCO for Unix being lower than TCO for
> > >Windows.
> > >I can't comment about downtime, I know that any Win2K box that I've
seen
> was
> > >up, and *stayed* up, as long as its owner wanted it to.
> > >The only exceptions were driver problems.
> >
> > *Ding* *Ding* *Ding*
> >
> > We have a winner.
>
> Had you ever used a computer before you'd realize how stupid that sounds.
> Cause once you get the right driver in place, guess what... it works and
> stays working. So, by installing a certified driver FIRST you never have a
> driver problem.
>
> Are you trying to suggest that other OSes are _immune_ to driver problems?
> Never seen a unix box hosed due to horrible drivers? I have ...
HURD is supposed to be immune to bad drivers, isn't it?
I know that you can easily hose a linux with misconfiguring your sound card.
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Fri, 11 May 2001 17:44:00 +0200
"mlw" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Ayende Rahien wrote:
> >
> > "Matthew Gardiner" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > Jan Johanson wrote:
> > > >
> > > > "Donn Miller" <[EMAIL PROTECTED]> wrote in message
> > > > news:[EMAIL PROTECTED]...
> > > > >
> > > > >
> > > > > Jan Johanson wrote:
> > > > >
> > > > > > Is there really any doubt that W2K rox the house?
> > > > >
> > > > > Yes, because unix systems stay up longer. Remember the "awesome"
MTTF
> > > > > that Windows 2000 exhibits? LOL.
> > > >
> > > > Yes, I do. And W2K stays up every bit as long as unix systems.I know
you
> > > > won't admit it or can't imagine it but that's your problem not ours.
> > > Why does Microsoft rely so heavily on clustering technology? when you
> > > can get a big iron like a SunFire w/ 16 x Sparc III CPUS, or an z900
> > > mainframe that can stay up for years, requiring little or no
> > > maintainance. I would be quite interested in a Windows 2000 Server
vs.
> > > SUN Sunfire midframe, without clustering technology, and see the
uptimes
> > > of them.
> >
> > Can't speak of uptime, because it's usually to expensive (and long) to
> > benchmark those.
> > But according to TCP.ORG, in the unclustered category, Win2K win on
> > price/performance.
> > On unclustered/clustered category, Win2K wins *both* price/performance &
> > performance.
> If I've said it once, I've said it a hundred times.
>
> TPC is not a universal benchmark. People must be paying members to submit
> results. Because of this only certain configurations ever get listed, and
thus
> is not usable as a fair and equal benchmark.
Oh, you've to pay, how does this make it an unfair & unequal benchmark?
> Second, the OS has little to do with TPC results. It is mostly database
and
> configuration.
Really?
What would you say if I run a TCP test on a one of Sun's big iron boxes,
using Linux 2.0 & Oracle?
Do you think that it would give the same result as Solaris? Even close?
After all, "the OS has little to do with TPC results", according to you.
> TPC results have no place in an OS discussion.
Yes, they have.
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Fri, 11 May 2001 18:04:46 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Ayende Rahien in comp.os.linux.advocacy on Fri, 11 May 2001
> [...]
> >Can't speak of uptime, because it's usually to expensive (and long) to
> >benchmark those.
> >But according to TCP.ORG, in the unclustered category, Win2K win on
> >price/performance.
>
> Ironic, given the monopoly pricing, eh? Just goes to show you why "high
> prices" aren't the be-all-and-end-all of why monopolies are illegal.
> Obviously the numbers carefully avoided the zero price for Linux and the
> predatory licensing of Windows. Could be MS just competed for 'cheapest
> system' in a way that TCP.ORG is rather blind to. Windows runs on
> cheaper hardware *would* be the lesson learned, but for the fact that we
> know this is not true in the real world, beyond a certain limit. (Linux
> can't run on a 286, but then DOS can't compare to Linux to begin with.)
DOS, actually, is possibly the best platform for a RDBMS.
It stays *out* of the DB way, period. A RDBMS does pretty much everything on
its own.
I think that the only reason that main reason that RDBMS use OS is drivers &
maintaince over hardware platforms.
Otherwise, they would be just as happy have full control of the computer.
> >On unclustered/clustered category, Win2K wins *both* price/performance &
> >performance.
>
> By pitting only clustered Windows against only unclustered Linux,
> mostly.
There isn't a *single* Linux on the TPC.ORG list.
There are non-MS benchmarks with Windows , IBM with DB2 best result, for
example, guess what OS they were running them on? Win2K.
They couldn't make it fast enough on *their* OS.
Best result of DB2(IBM) on OS/400(IBM) 163775 @ 58.88 US $
DB2(IBM) on Win2K(MS) 440879 @ 32.28 US $
Check here for performance table:
http://www.tpc.org/tpcc/results/tpcc_perf_results.asp?resulttype=all
Win2K has the first *six* places.
Check here for performance table for clustered results:
http://www.tpc.org/tpcc/results/tpcc_perf_results.asp?resulttype=cluster&ver
sion=3
Win2K has the first *nine* places:
Price/Performance?
http://www.tpc.org/tpcc/results/tpcc_price_perf_results.asp?resulttype=all
Win2K/NT4 (both non-clustered & clustered) owns this place, with all *ten*
places taken by them.
Performance on non-clustered system:
http://www.tpc.org/tpcc/results/tpcc_perf_results.asp?resulttype=noncluster&;
version=3
Is the only place where Unix seems to shine, largely because it run on more
hardware than Win2K has a market to, it seems.
> I'm willing to be if you give me the url, it will be the very same page
> of numbers I looked at months ago. You're just unaware of how sound a
> spanking Erik got when he brought it up back then.
http://www.tpc.org/
Now, *that* is a hard to rememeber URL, isn't it?
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Fri, 11 May 2001 18:07:35 +0200
"Jan Johanson" <[EMAIL PROTECTED]> wrote in message
news:3afc9e01$0$41652$[EMAIL PROTECTED]...
>
> "mlw" <[EMAIL PROTECTED]> wrote in message
> > 6% difference, at these performance levels, can be anything and is not
> even
> > relevant. Not to mention we have different disk subsystem
configurations.
> That
> > 15K RPM disk? What did it do? Presumably it was the system disk. It
could
> have
> > been the disk to which the logs get written, which would have an impact
on
> > performance.
>
> I don't think 6% is much - but when tux beat IIS in the first round by 7%
it
> was linvocate heaven so... this is just a little form of payback :)
2.7%, not 7%.
Just a reminder. :-D
------------------------------
** 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
******************************
