Linux-Advocacy Digest #912, Volume #34 Sat, 2 Jun 01 14:13:05 EDT
Contents:
Re: The nature of competition (The Ghost In The Machine)
Re: Justice Department LOVES Microsoft! ("Daniel Johnson")
Re: Justice Department LOVES Microsoft! (Rick)
Re: RIP the Linux desktop (Jesse F. Hughes)
Re: Justice Department LOVES Microsoft! ("Daniel Johnson")
Re: Linux beats Win2K (again) ("Edward Rosten")
Re: Linux beats Win2K (again) ("Edward Rosten")
Re: Why should an OS cost money? (Marc Schlensog)
Re: The beginning of the end for microsoft (Marc Schlensog)
Re: ease and convenience (*long* and possibly boring;-) (Marc Schlensog)
Re: Justice Department LOVES Microsoft! ("Daniel Johnson")
Re: The _one_ thing that pisses me off about Linux (The Ghost In The Machine)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: The nature of competition
Date: Sat, 02 Jun 2001 17:01:49 GMT
In comp.os.linux.advocacy, Matthew Gardiner
<[EMAIL PROTECTED]>
wrote
on Wed, 23 May 2001 15:02:03 +1200
<9ef98o$dv7$[EMAIL PROTECTED]>:
>Surprise, surprise, all of them run a veriant of UNIX. I wonder why? Chad,
>maybe you could answer that question.
Maybe it's because they haven't discovered Microsoft's
bodginess^H^H^H^H^H^H^H^H^Hclever solutions yet....? :-)
Or maybe it's because Microsoft is such a good
monopoly^H^H^H^H^H^H^H^Hsoftware provider, and they
haven't been exposed to Microsoft's indoctrination^H^H^H^H^H^H^H^H^H^H^H^H^H^H
advertising yet?
It could be because Microsoft has
cornered^H^H^H^H^H^H^H^Hexpanded the market for Web browsers
on non-Unix machines.... :-)
Or maybe it's the big fat mess^H^H^H^H^H^H^H^H^H^H^H^Hall-in-one
integration of their software products?
Or maybe it's....I could go on, but why? :-)
>
>Matthew Gardiner
>
>"Edward Rosten" <[EMAIL PROTECTED]> wrote in message
>news:9edm6r$gkf$[EMAIL PROTECTED]...
>> >> >> TPC is just a benchmark, not a real world measure. In the real
>> >> >> world, it is Linux, not Win2K that shows up at the top end of
>> >> >> acalibility and price/performance.
>> >> >
>> >> > Then why aren't we seeing any real world measurements?
>> >>
>> >> But we have. Linux is used at the top end of scalibility, where
>> >> price/performance is really critical, since the costs are so high, ie
>> >> supercomputers. There are several Linux machines in the top500, there
>> >> are no Win2K or NT amchines in the top 500. A real supercomputer in
>> >> real use is the real world. A benchmark setup is not.
>> >
>> > My understanding is that all of those "supercomputers" are "in
>> > development", and not being used in the real world. The ones that are,
>> > are clusters, not single machines.
>>
>> No, thewre are some real ones being used now. Check out the list at
>> www.top500.org
>>
>> Yes, it is true, the computers are clustered in a tightly coupled network
>> (by clusetering standards), but it still rates as a supercomputer.
>> Besides, one measure of scalibility is the ability to cluster. In this
>> area, Linux thrashes Win2K. If Win2K was so much better, we'd see Win2K
>> clusters in the top500. We don't; we see Linux ones instead.
>>
>>
>>
>>
>> >> >> >> Linux has been proven to be more stable.
>> >> >> >
>> >> >> > It has? How? I've seen no verifiable studies that show Linux's
>> >> >> > uptime to be greater than anything else.
>> >> >>
>> >> >> 120 day MTTF, *with* nightly reboots.
>> >> >
>> >> > Really? There's Linux uptime studies that show this? Or did you
>> >> > forget the question?
>> >>
>> >> Linux's uptime isn't at the top end (Only OS/390 and VM are with a
>> >> guarnteed uptime of 35 years), but I'd wager that Linux can beat 120
>> >> days with nightly reboots (ie Win2K's verified MTTF).
>> >
>> > Again, how come there are no studies?
>>
>> Who's going to pay?
>>
>>
>> --
>> (You can't go wrong with psycho-rats.)
>(u98ejr)(@)(ecs.ox)(.ac.uk)
>>
>> /d{def}def/f{/Times-Roman findfont s scalefont setfont}d/s{10}d/r{roll}d f
>5 -1
>> r 230 350 moveto 0 1 179{2 1 r dup show 2 1 r 88 rotate 4 mul 0
>rmoveto}for/s{15
>> }d f/t{240 420 moveto 0 1 3 {4 2 1 r sub -1 r show}for showpage}d pop t
>
>
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 33d:10h:01m actually running Linux.
[select one]
This is a voluntary signature virus. Send this to somebody.
Hi. What's your sign? Mine's "Out To Lunch".
Life's getting too complicated, even listening to the radio.
[ ] Check here to always trust monopolistic software.
The Internet routes around censorship.
[ ] Do you want this message to be private? Oops, too late.
This is a .sig.
The EAC doesn't exist, but they're still watching you.
------------------------------
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: Sat, 02 Jun 2001 17:11:01 GMT
"Chris Ahlstrom" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Daniel Johnson wrote:
> > > You're half right. GIMP uses GTK+, which has been
> > > ported to Windows. It would have been hell to port
> > > if it had used Xlib directly.
> >
> > Why? Surely it is simply a matter of running
> > an X Server, if it did do so?
>
> Sure. If you can find a decent free X server for
> Windows.
Free? Why does it have to be free? This is
Windows- the people using it are used
to paying for stuff.
> It seems to me that the GIMP authors
> don't want to have to depend on your average Windoze
> user finding, downloading, installing, and configuring
> an X-Server just to run a paint program.
Surely the GIMP installer could handle
installing the necessary X-server.
Heck, theoretically it could be embedded
into the Windows GIMP code. Though
I don't know that anyone has written an
X server than can be so used, it should be
workable on a technical level.
> > It's easier, anyway. But ignoring end users
> > desire for a consistant, familiar user interface
> > is not usually wise.
>
> Try for consistent spelling first.
:D
> Anyway, you
> used the key word, "usually". Anyone who has mastered
> one set of conventions can usually master another.
> That is why I can edit documents with vi, emacs,
> and your "standard" [not all Microsoft functions
> adhere] CUA interface editor.
Users can do that- once upon a time they
*had* to.
But the product that does not ask them to
has an advantage.
> > Hmmm. I take it you feel that they
> > should have slavishly copied the
> > Macintosh Toolbox, despite its not
> > inconsiderable warts, then?
>
> I don't know shit about Macintosh. I'm talking about
> Microsoft ignoring some pretty useful UNIX conventions.
What conventions are those, and why should
Microsoft have prefered them to the useful
Macintosh conventions, say?
[snip]
> > That is not true if you include Unix.
>
> Again I say unto you, what you say here is utter
> boolshit. Apart from the exact keystroke or
> mouse sequences, I've not found anything in
> Windows functionality that UNIX lacks.
> But, hey, I've still got a couple more decades
> of computer experience to come, don't I!
I think you misunderstood me; I am talking
about the featureset used by application
developers.
For instance, Unix has no equivalent of
MS's "Windows Scripting Host" or
Apple's "Open Scripting Architecture".
[snip]
------------------------------
From: Rick <[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: Sat, 02 Jun 2001 13:25:31 -0400
Daniel Johnson wrote:
>
> "Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
> news:9fa6c7$bgt$[EMAIL PROTECTED]...
> > "GreyCloud" <[EMAIL PROTECTED]> wrote:
> > > That is not true in regards to the widget set. I ported the source of
> > > xgdb from linux (KDE) to Solaris 8 (Motif) and just compiled the
> > > source. When I ran xgdb it was not using the KDE widget set but Motif
> > > widget set. Everything ran as it should.
> >
> > When you are talking about ported, what are you talking about?
>
> What he means, I think, is that some programs are
So.. you are still telling people what other people think, huh?
--
Rick
------------------------------
Subject: Re: RIP the Linux desktop
From: [EMAIL PROTECTED] (Jesse F. Hughes)
Date: Sat, 02 Jun 2001 17:22:28 GMT
drsquare <[EMAIL PROTECTED]> writes:
> On Sat, 02 Jun 2001 05:45:14 GMT, in comp.os.linux.advocacy,
> ([EMAIL PROTECTED] (Jesse F. Hughes)) wrote:
>
> >drsquare <[EMAIL PROTECTED]> writes:
>
> >> >Scotland Yard has been fudging the data for years, by Only reporting
> >> >CONVICTIONS, not crimes reported.
>
> >> If they're not convicted, then you cannot count it as a crime. Do you
> >> not have "innocent until proven guilty" in your country?
>
>
> >Consequently, murder/suicide is not a crime. After all, if I kill
> >myself after offing whoever is annoying me, I will not be convicted.
> >
> >Brilliant. No convictions, no crimes.
> >
> >I'm not entering the debate here, but that kind of tortured reasoning
> >deserves comment. Clearly, some *crimes* go unsolved (or unpunished),
> >but no one has any doubt that a crime has been committed.
>
> Well, that's relevant.
It looks relevant to me. If we look back at the quotes above, we see
that you were denying that there is a difference between convictions
and number of crimes reported. My response is intended to refute that
denial. Seems to be relevant to the quotations above.
Perhaps you mean that *your* statement was irrelevant, and hence so
was my refutation? I don't know if that's what you intended, but
looks to me like my response must be as relevant as the statement to
which I was responding here.
Well, no matter.
--
Jesse Hughes
"There's a thrill that's gone that I'll probably not have in quite the
same way again. After all, FLT was a unique animal, and we had a
great dance." -J.S. Harris on "proving" Fermat's last theorem
------------------------------
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: Sat, 02 Jun 2001 17:29:13 GMT
"Chris Ahlstrom" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Daniel Johnson wrote:
> > That does let you do a *small* part of what
> > Kalidioscope does- you can change the way
> > window frames look, and the widgets in the
> > title bar.
> >
> > That's something. But it's not very much.
>
> Can you tell me what Kaleidoscope does? (Feel
> free to tell me to f*ck off and research it
> on my own, I won't be offended).
What it does is dynamically switch the
widgets the Macintosh offers with other
widgets- there are many widgets available
to use.
Applications still access the widgetset
just as they would if they were accessing
the standard ones. But while the controls
they get look like standard controls to the
apps, to the users they can look like
anything.
[snip]
> > Linux is *exceptional* as a "desktop OS" for
> > not having this feature- and quite a few others
> > too.
>
> What do you mean? Linux has bash and many other
> script processors. And any app can use OS
> functions to spawn them! What you say makes
> no sense!
Windows has an equivalent to bash-
COMMAND.COM. (Okay, so it's not
much of an equivalent :D ). This is different.
There are actually several technologies
at work here.
OLE Automation is a protocol by which
a scripting system (VB, say) can discover
what an application can do, and can use
its capabilities.
With this, a script can manipulate documents
using the desktop apps that the users
would. It is not necessary to interpret
the document's storage format. If you are
scripting Excel, say, you can manipulate
areas of the spreadsheet directly, because
those are the kinds of things Excel understands,
and it exposes them to you.
The scripts can be run in separate processes
ala VB programs, but they can also be run
'internally' using an ActiveX control. When used
this way a program can control the scripts so that
they only see the objects it wants them to. This
is what Internet Explorer does, for instance.
The Macintosh has most of this stuff too, and
it is collectively called the "Open Scripting
Architecture". I used the Windows example
mainly because spliting it into pieces makes
it a little easier to understand.
> Perhaps our internal maps of the
> world differ too much right now for us to
> have any meeting of minds. What you're saying
> just sounds incredible to me as I sit here
> using Linux.
I think calling this stuff "scripting" may
be problematic; we don't really mean quite
the same thing.
I do not mean "scripting" in the sense
of "lightweight programming" alone,
but in the sense of controlling *other*
apps.
You know how Java^H^H^H^HECMAScript
can "see" things in a web-page and manipulate
them?
It's like that. By having this be a feature of
the OS, any application can do likewise.
------------------------------
From: "Edward Rosten" <[EMAIL PROTECTED]>
Subject: Re: Linux beats Win2K (again)
Date: Sat, 02 Jun 2001 19:33:55 +0100
In article <[EMAIL PROTECTED]>, "Pete
Goodwin" <[EMAIL PROTECTED]> wrote:
> In article <9f91ep$rkk$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>> > Where is this example?
>>
>> I love your tactic of simply ignoring every piece of evidence and when
>> the poster won't post it for the 25th time, you accuse them of lying.
>>
>> Now this is a prime example.
>
> Nobody recently has posted a single shred of evidence.
^^^^^^^^
sounds like a concession that evidence has been posted, in which case why
should I have to post more?
If so, what is wrong with the first lot. Perhaps you could post it here
and say exactly what is wrong with it.
-Ed
--
(You can't go wrong with psycho-rats.) (u98ejr)(@)(ecs.ox)(.ac.uk)
/d{def}def/f{/Times-Roman findfont s scalefont setfont}d/s{10}d/r{roll}d f 5 -1
r 230 350 moveto 0 1 179{2 1 r dup show 2 1 r 88 rotate 4 mul 0 rmoveto}for/s 15
d f pop 240 420 moveto 0 1 3 {4 2 1 r sub -1 r show}for showpage
------------------------------
From: "Edward Rosten" <[EMAIL PROTECTED]>
Subject: Re: Linux beats Win2K (again)
Date: Sat, 02 Jun 2001 19:35:02 +0100
>> > No wonder you can't post a single shred of evidence!
>>
>> Why should I reprat myself, hmm?
>
> Yes, why should you re-prat yourself, exactly!
LOL! Good typo, that one :)
-Ed
--
(You can't go wrong with psycho-rats.) (u98ejr)(@)(ecs.ox)(.ac.uk)
/d{def}def/f{/Times-Roman findfont s scalefont setfont}d/s{10}d/r{roll}d f 5 -1
r 230 350 moveto 0 1 179{2 1 r dup show 2 1 r 88 rotate 4 mul 0 rmoveto}for/s 15
d f pop 240 420 moveto 0 1 3 {4 2 1 r sub -1 r show}for showpage
------------------------------
From: Marc Schlensog <[EMAIL PROTECTED]>
Subject: Re: Why should an OS cost money?
Date: Sat, 02 Jun 2001 18:08:20 +0200
Ayende Rahien wrote:
>
> "Nick Condon" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Stuart Fox wrote:
> >
> > >
> > >"mlw" <[EMAIL PROTECTED]> wrote in message
> > >news:[EMAIL PROTECTED]...
> > >> If one thinks about the history of man, and the nature of invention,
> > >> one
> > >must
> > >> ask themselves why an OS costs any money.
> > >>
> > >Wrong question, the real question is why shouldn't an OS cost money?
> > >
> > >An OS should cost money, because it is derived from effort, which most
> > >humans expect to be paid for. Total up the amount of time that Linux
> > >has taken to develop, and then try and recover that cost. Linux of
> > >course is a special example, as it is allegedly developed by people on
> > >their "free" time. Of course, this doesn't include people who are paid
> > >to develop it (Linus by Transmeta, Alan Cox by Redhat). I don't see too
Hmm... isn't Linus paid by Transmeta for doing something totally
different?
> > >many Linux companies actually making money - because they can't recover
> > >the costs of their effort.
OK, I will pay a rational price for software, a price that this software
is worth.
Paying $250 for a CD and a booklet (Win9x) or $700 for an office-suite
is not
priceworthy, if you ask me. I really don't want to take my month of work
and buy
software with it, when I can have all that and even more for about 10%.
[Snipped]
--
They're only trying to make me LOOK paranoid!
------------------------------
From: Marc Schlensog <[EMAIL PROTECTED]>
Subject: Re: The beginning of the end for microsoft
Date: Sat, 02 Jun 2001 18:21:04 +0200
Terry Porter wrote:
>
> On Fri, 01 Jun 2001 22:32:41 GMT, KSG <[EMAIL PROTECTED]> wrote:
> >
> > "Michael Vester" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> >> Ben Franchuk wrote:
> >
> >> I find a network card install in Linux quite simple. It has been over 4
> >> years since I used a dial-up ISP. A Linux install with a network card with
> >> my ISP requires setting the IP address, gateway and name servers. I never
> >> experienced a Linux install that did not identify the NIC and load the
> >> correct drivers.
> >
> > Maybe I could get you to install Linux on my computer.
>
> Maybe you should state your offer to do so in dollars?
>
> > I've tried Debian
> > and Red Hat, neither found my NIC, and I could never get networking to work.
> > Win2k came up just fine with networking and all.
Hmm, just great. Let me tell you my experience with NICs/SCSI-adaptors.
WinME was the first M$-OS that was able to recognize and install drivers
for my
NIC (Realtek8139). They didn't work, nevertheless. Neither Win9x nor
WinNT or even
Win2k were able to find this card. But my SuSE Linux 6.0 was able to. I
booted from
my SCSI-CD-drive and the module for my SCSI-adaptor and the NIC were
loaded.
I also tried to install W9x, WME, WNT and W2K from CD with the same
SCSI-adaptor
(oh yeah, it's a SYM53c875). Guess what? WNT/W2K bluescreened and the
W9x series
wasn't able to find my CD-drive.
What chip did the NIC have you were using?
>
> Then perhaps the $300 you paid for Win2k was worth it ?
What about additional software? Is a Windog system actually worth
thousands of bucks?
I don't think so. I'd rather save the money for software and get me a
nicely equipped
Alpha.
>
> >
> > KSG
> >
> >
>
> --
> Kind Regards
> Terry
> --
> **** ****
> My Desktop is powered by GNU/Linux.
> 1972 Kawa Mach3, 1974 Kawa Z1B, .. 15 more road bikes..
> Current Ride ... a 94 Blade
> Free Micro burner: http://jsno.downunder.net.au/terry/
> ** Registration Number: 103931, http://counter.li.org **
Just my 2c...
Marc
--
They're only trying to make me LOOK paranoid!
------------------------------
From: Marc Schlensog <[EMAIL PROTECTED]>
Subject: Re: ease and convenience (*long* and possibly boring;-)
Date: Sat, 02 Jun 2001 17:54:28 +0200
Todd wrote:
>
> "GreyCloud" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Todd wrote:
> > > Well, 9x seems to crash more than its fair share then... I've been
> spoiled
> > > by W2k. I think XP will be great for consumers... they are going to
> wonder
> > > why they don't have to reboot nearly as often.
> > >
> >
> > Nearly as often?? I'd rather not reboot at all.
>
> Obviously.
>
> But all OSes crash... and yes, even Linux does.
Hmm, yes, it obviously does.... I had about one crash a year
during the last... hang on... 6 years. That's even less than
Windows9x does per day and WinNT does per week....
>
==============8<====[Snippage]===========>8===================
Marc
--
They're only trying to make me LOOK paranoid!
------------------------------
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: Sat, 02 Jun 2001 17:41:23 GMT
"Rick" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Daniel Johnson wrote:
> > > When you are talking about ported, what are you talking about?
> >
> > What he means, I think, is that some programs are
>
> So.. you are still telling people what other people think, huh?
Yes. Any further questions? :D
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: The _one_ thing that pisses me off about Linux
Date: Sat, 02 Jun 2001 17:43:56 GMT
In comp.os.linux.advocacy, Edward Rosten
<[EMAIL PROTECTED]>
wrote
on Mon, 07 May 2001 09:35:35 +0100
<9d5j7e$m82$[EMAIL PROTECTED]>:
><snip>
>
>> whole story of CPU porting problems, but then I know sod-all about the
>> problems of different size integers from platform to platform.
>
>Assembler is by no means the only problem.
>
>There are a whole host of problems, such as making false assumptions
>about the sizeof an integer and char.
>
>sizeof(char)==1 by definition
>
>and CHAR_BITS >= 8
>
>and CHAR_MAX <=SHORT_MAX <=INT_MAX <=LONG_MAX <=LONG_LONG_MAX (now in
>C99)
>
>however, it is possible to have a stadards conforming compiler, where
>sizeof(long long)==1 (CRAY had one, for instance and many DSP processors
>have them as well).
That would be an awful big char. :-)
>
>I think quite a few programs assume that CHAR_MAX == 255 and will break
>if otherwise.
>
>
>Another problem is the endianness, for instance
>
>a<<1 is *2 on one platform and /2 on another platform.
??
Endianity has to do with the handling of things like:
union silly
{ int i;
char c[sizeof(int)];
} u;
and then looking at u.c[0] when u.i is set to 1. On a little-endian
machine such as the x86,
u.i = 1; cout << (int) u.c[0]
results in a 1; however, on machines sch as m68k and Sparc, it
will result in 0.
(Or one can set u.i = 0x11223344 and see whether u.c[0] == 0x11
or 0x44.)
I have yet to run into a machine where a<<1 *divides* by 2, although
I have run into machines where
union silly2
{ int i;
int b0:1;
} v;
u.i = 1; cout << u.b0;
produces results that are slightly unexpected (not that this is all that
well defined, anyway; programs depending on this will probably break :-) ).
Some very old machines might do peculiar things with negative numbers
and shifting, though, if they use one's complement arithmetic.
AFAIK, no modern machine does this, although the MiX emulator (Knuth)
uses undefined arithmetic means (presumably at the time of its design,
some machines were still using BCD) and therefore a<<1 on MiX will
not produce a useful result, arithmetically speaking.
[.sigsnip]
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 33d:14h:26m actually running Linux.
I am, you are, he, she, and it is, but they're not.
------------------------------
** 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
******************************
