Linux-Advocacy Digest #304, Volume #29 Mon, 25 Sep 00 15:13:06 EDT
Contents:
Re: [OT] Bush v. Gore on taxes (was: Re: Would a M$ Voluntary Split ...) (STATIC66)
Re: Space Shuttle uses Windows software almost exclusively ("Matt")
Re: "Overclocking" Is A Bad Idea (Bruce Scott TOK)
Re: "Overclocking" Is A Bad Idea
Re: Space Shuttle uses Windows software almost exclusively (mark)
Re: How low can they go...? (The Ghost In The Machine)
Re: The Linux Experience
Re: Id Software developer prefers OS X to Linux, NT (The Ghost In The Machine)
Re: Never tell me again that Windows is easy to install!!! It's a lie! (The Ghost
In The Machine)
Re: "Overclocking" Is A Bad Idea (Bruce Scott TOK)
Re: Yeah! Bring down da' man! ("Simon Palko")
Re: Never tell me again that Windows is easy to install!!! It's a lie! (The Ghost
In The Machine)
Re: So did they ever find out what makes windows98 freeze up all the time? (mark)
----------------------------------------------------------------------------
From: STATIC66 <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy
Subject: Re: [OT] Bush v. Gore on taxes (was: Re: Would a M$ Voluntary Split ...)
Date: Mon, 25 Sep 2000 18:03:45 GMT
On Tue, 29 Aug 2000 01:23:28 GMT, ZnU <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>, "Aaron R. Kulkis"
><[EMAIL PROTECTED]> wrote:
>
>> ZnU wrote:
>> >
>> > In article <[EMAIL PROTECTED]>, "Aaron R. Kulkis"
>> > <[EMAIL PROTECTED]> wrote:
>>
>> > And then pay to have them sit in jail. Which is much more
>> > expensive. Smart.
>>
>> Why should we pay for their incarceration. Put them to work doing
>> something to earn their keep.
>
>Even if I go along with the idea that people who misbehave in school
>should be expelled and forced into slave labor, it still doesn't solve
>the problem. Do you have any idea what it costs to keep someone in
>prison? You can go to college for less.
So criminals working to earn their keep is slave labor???
>
>> > > > > The public schools can expel those who disrupt the school,
>> > > > > but they refuse to do so.
>> > > >
>> > > > And then where do they go? Again, if you don't pay for their
>> > > > schooling now, you'll be paying for their incarceration later.
>> > >
>> > > Where they will serve as an example to others.
>> >
>> > If it were that simple, there wouldn't be kids causing trouble in
>> > school right now.
>> >
>> > > > > Hoisted by their own petards.
>> > > > >
>> > > > >
>> > > > > Actually, it's the tax-paying public who is paying for an
>> > > > > K-12 "education" but the money is instead being used for
>> > > > > left-wing indoctrination.
>> > > >
>> > > > What "left-wing indoctrination" would this be? Teaching kids
>> > > > about
>> > >
>> > > Global warming and other Eco-leftism
>> >
>> > Global warming is not "Eco-leftism" any more than quantum mechanics
>> > is.
>> >
>> > > Pro-homosexuality propaganda
>> >
>> > Examples?
>>
>> "Daddy has a roommate" "Heather has two mommies"
>
>They also teach kids about tribal life in Africa. Do you think this is
>intended to encourage kids to go join African tribes? If kids aren't
>exposed to the different ways in which people live they often turn into
>bigots. Which is what you seem to want.
>
>> > > Socialism
>> >
>> > Examples? Most school history texts I've seen have an
>> > American/capitalist bias, actually.
>>
>> Your information is at least 10 years out of date.
>
>Examples?
------------------------------
From: "Matt" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy
Subject: Re: Space Shuttle uses Windows software almost exclusively
Date: Mon, 25 Sep 2000 11:07:16 -0700
> If it is in the Intel section, it is Intel software which may have been
> ported to the Mac. It is not Mac only software. Moreover, those dual os
> programs are virtually all games. A game machine is not a computer and it
> doesn't take a computer to run games.
> The Mac is an insignificant game machine incapable of real world computer
> tasks such as running DB2, running cash registers, inventory systems,
> point of sale systems, etc. all of which are done in hundreds of thousands
> of businesses all over the world by Intel platform hardware.
>
Now that is just a DUMB statement. The latest games require a TON of
computing power. Just because a machine doesn't do POS or database apps, it
doesn't mean that it isn't a computer. Now quit talking out of your ass and
make some sense.
Matt
------------------------------
From: [EMAIL PROTECTED] (Bruce Scott TOK)
Subject: Re: "Overclocking" Is A Bad Idea
Date: 25 Sep 2000 20:11:35 +0200
In article <[EMAIL PROTECTED]>,
Jim Broughton <[EMAIL PROTECTED]> quotes Donal Fellows...
>"Donal K. Fellows" wrote:
>>
>> In article <[EMAIL PROTECTED]>,
>> Jim Broughton <[EMAIL PROTECTED]> wrote:
>> > What can be more arithmetic than calculations involved in ploting
>> > and then rendering a full motion scene in todays fast paced 3d FPS
>> > games.
>>
>> Scientific simulation. Those guys use whopper supercomputers for a
>> good reason, and those computers don't use Intel CPUs for a good
>> reason too, namely floating-point performance. By way of some
>> illustration, I was in a discussion on another group on the fastest
>> way to shuffle (IOW randomly permute) a list, and there were several
>> algorithms proposed. However, it turned out that while recent Intels
>> were fastest when an algorithm was written using mainly integer math,
>> an UltraSparc with a nominally lower clock speed would kick the pants
>> off that Intel chip if the algorithm used floating point comparisons
>> instead of integer ones.
Don't look now but there are more and more Pentium chips in those Linux
clusters people are building. The ones I know are running plasma
physics codes to do such things as MHD stability, turbulence, or
detailed particle motion in complicated magnetic surface geometries.
Typical runs are on 16 or 32 chips. More commonly people use alphas but
that is changing with the speed increases of the pentiums. One
disadvantage of the alphas is that their caches are too small and you
take inordinate penalty for misses (which are inevitable in most
codes). Simple PC chips and ethernet parallelism (using LAM MPI) are an
excellent way to get a reasonable bang for a small buck.
Sparcs are faster, but not _that_ much faster, and they cost more. Same
for alphas.
--
cu,
Bruce
drift wave turbulence: http://www.rzg.mpg.de/~bds/
------------------------------
From: [EMAIL PROTECTED] ()
Subject: Re: "Overclocking" Is A Bad Idea
Date: Mon, 25 Sep 2000 18:19:14 -0000
On 25 Sep 2000 20:11:35 +0200, Bruce Scott TOK <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>,
>Jim Broughton <[EMAIL PROTECTED]> quotes Donal Fellows...
>
>>"Donal K. Fellows" wrote:
>>>
>>> In article <[EMAIL PROTECTED]>,
>>> Jim Broughton <[EMAIL PROTECTED]> wrote:
>>> > What can be more arithmetic than calculations involved in ploting
>>> > and then rendering a full motion scene in todays fast paced 3d FPS
>>> > games.
>>>
>>> Scientific simulation. Those guys use whopper supercomputers for a
>>> good reason, and those computers don't use Intel CPUs for a good
>>> reason too, namely floating-point performance. By way of some
>>> illustration, I was in a discussion on another group on the fastest
>>> way to shuffle (IOW randomly permute) a list, and there were several
>>> algorithms proposed. However, it turned out that while recent Intels
>>> were fastest when an algorithm was written using mainly integer math,
>>> an UltraSparc with a nominally lower clock speed would kick the pants
>>> off that Intel chip if the algorithm used floating point comparisons
>>> instead of integer ones.
>
>
>Don't look now but there are more and more Pentium chips in those Linux
>clusters people are building. The ones I know are running plasma
>physics codes to do such things as MHD stability, turbulence, or
>detailed particle motion in complicated magnetic surface geometries.
>Typical runs are on 16 or 32 chips. More commonly people use alphas but
>that is changing with the speed increases of the pentiums. One
>disadvantage of the alphas is that their caches are too small and you
>take inordinate penalty for misses (which are inevitable in most
>codes). Simple PC chips and ethernet parallelism (using LAM MPI) are an
>excellent way to get a reasonable bang for a small buck.
>
>Sparcs are faster, but not _that_ much faster, and they cost more. Same
>for alphas.
Sparcs are really not that much more expensive these days. Sure
you can get some pieces and parts for PC's cheaper than just about
anything else. However, for anything above the point of being your
own VAR, PC's aren't as cheap as they are represented to be.
[deletia]
--
A bird in the hand makes it awfully hard to blow your nose.
Caution: Keep out of reach of children.
------------------------------
From: [EMAIL PROTECTED] (mark)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy
Subject: Re: Space Shuttle uses Windows software almost exclusively
Date: Mon, 25 Sep 2000 07:00:43 +0100
In article <[EMAIL PROTECTED]>,
dc wrote:
>On Sun, 24 Sep 2000 19:16:19 GMT, "Joe R." <[EMAIL PROTECTED]>
>wrote:
>
>>In article
>><[EMAIL PROTECTED]>, dc
>><[EMAIL PROTECTED]> wrote:
>>
>>> On Sun, 24 Sep 2000 02:21:42 GMT, "Joe R." <[EMAIL PROTECTED]>
>>> wrote:
>>>
>>> >In article
>>> ><[EMAIL PROTECTED]>, dc
>>> ><[EMAIL PROTECTED]> wrote:
>>> >
>>> >> On Sat, 23 Sep 2000 21:26:04 +0100,
>>> >> [EMAIL PROTECTED] (mark) wrote:
>>> >>
>>> >> >In article
>>> >> ><[EMAIL PROTECTED]>,
>>> >> > dc wrote:
>>> >> >>On Sat, 23 Sep 2000 17:47:11 GMT, Timberwoof <[EMAIL PROTECTED]> wrote:
>>> >> >>
>>> >> >>>some of these standards. And because of the Macintosh's excellent
>>> >> >>>support for networking, Apple product users are well-connected.
>>> >> >>
>>> >> >>This I don't quite understand. Not from a 1990's AppleTalk
>>> >> >>perspective, but from a September 2000 perspective, how are Apple
>>> >> >>product users "well-connected" compared to the rest of computerdom
>>> >> >>(meaning, NT and ME)?
>>> >> >
>>> >> >I'm having trouble with the phrase 'rest of computerdom' which I
>>> >> >assumed
>>> >> >would mean what people actually use, not NT and ME.
>>> >>
>>> >> For better or worse, that is what most people actually use.
>>> >
>>> >Most people use NT and ME?
>>> >
>>> >You're out of your mind (what little apparently remains).
>>>
>>> Or a derivative of one of those OSs, yes, Joe, that _is_ what most
>>> people use.
>>
>>No, they don't. "Derivative" means later work derived from the earlier
>>one.
>
>Agreed. So I'll just say "Windows".
>
>
Dos and derivatives would be accurate. Windows covers such a wide
range of mutually incompatible OSs that it's perhaps useful as a
marketing man's collective noun, but not really from an engineering
viewpoint.
--
Mark - remove any ham to reply.
(Killed (sigserv (This sig is reserved by another user)))
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: How low can they go...?
Date: Mon, 25 Sep 2000 18:22:49 GMT
In comp.os.linux.advocacy, [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
wrote
on Mon, 25 Sep 2000 16:09:57 -0000
<[EMAIL PROTECTED]>:
>On Sat, 23 Sep 2000 17:48:15 GMT, The Ghost In The Machine
><[EMAIL PROTECTED]> wrote:
>>In comp.os.linux.advocacy, [EMAIL PROTECTED]
>><[EMAIL PROTECTED]>
>> wrote
>>on Fri, 22 Sep 2000 23:17:30 -0000
>><[EMAIL PROTECTED]>:
[snip for brevity]
>>> What's this "we" stuff? ppppffffttt!
>>
>>Well, if you've never bought a computer with preinstalled MS software ever
>>in your lifetime, then I suppose it would be just "we minus jedi". :-)
>>(It's possible!)
>
> The last prebuilt computer I bought was an Atari 520STe.
Ah ha! Well, my apologies then; I was not aware of this. :-)
[snip]
--
[EMAIL PROTECTED] -- wondering if Amigas were ever sold by motherboard :-)
------------------------------
From: [EMAIL PROTECTED] ()
Subject: Re: The Linux Experience
Date: Mon, 25 Sep 2000 18:23:19 -0000
On 25 Sep 2000 17:57:51 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>On Mon, 25 Sep 2000 17:48:10 -0000, [EMAIL PROTECTED] wrote:
>
>> However, "doc" is a rather obvious mnemonic. Also, the
>> sorts of documentation in /usr/doc do tend to conform
>> to common characteristics as opposed to manpages.
>
>Yeah, they do have common characteristics -- they don't actually
>include user documentation.
>
> ruffbruff% find /usr/doc|head
> /usr/doc
> /usr/doc/adjtimex-1.3
> /usr/doc/adjtimex-1.3/README
> /usr/doc/ldconfig-1.9.5
> /usr/doc/ldconfig-1.9.5/README
> /usr/doc/vmware
> /usr/doc/vmware/CHANGES
> /usr/doc/vmware/README
> /usr/doc/vmware/EULA
> /usr/doc/vmware/DHCP-COPYRIGHT
>
>> Merely dumping everything together is "willy-nilly"
>> as opposed to subjecting something resembling
>> structure to it.
>
>That would be fine if /usr/doc had some "structure". Maybe they need
>to make a directory called
>"/usr/doc/READMES_and_COPYING_and_other_files_that_dont_belong_anywhere_else"
There's more to /usr/doc than just the package miscellaneous
documentation actually. Not to mention, there is infact a
term for "stuff that doesn't really fit in a reasonably
named category". It's called miscellaneous.
You don't necessarily want to clutter up you "one stop
shopping" point of entry with everything and the kitchen
sink either. Even the *hlpfiles themselves should have
some sort of organization.
The likely alternatives that you would provide don't do even
this. So unless you would like to drag big iron into the
disucussion, Unix documentation in general and Linux in
specific aren't as terrible as they're made out to be.
--
It takes a special kind of courage to face what we all have to face.
I'm going to give my psychoanalyst one more year, then I'm going to Lourdes.
-- Woody Allen
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Id Software developer prefers OS X to Linux, NT
Date: Mon, 25 Sep 2000 18:25:45 GMT
In comp.os.linux.advocacy, PistolGrip
<[EMAIL PROTECTED]>
wrote
on Sun, 24 Sep 2000 19:48:33 -0500
<cixz5.278$[EMAIL PROTECTED]>:
>"C Lund" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> In article <4yoz5.235$[EMAIL PROTECTED]>, "PistolGrip"
>> <[EMAIL PROTECTED]> wrote:
>>
>> > Well... can't believe I'm jumping in the silly argument.
>>
>> Note that I jumped out of it some time ago. I got fed up with that
>> merry-go-round.
>>
>> > But, what version
>> > of NT/2000 are we talking about?
>>
>> I think he was talking about NT in general.
>>
>> > I'll point out a *few* things W2k has that Win9x doesn't.
>>
>> (long and informative list snipped)
>>
>> Again, it would seem W2K has the same old interface, but lots of new stuff
>> "under the hood". Yes?
>
>Yes, although the interface is stream-lined considerably. It still looks a
>lot like Win9x but 'feels' better to me.
Well, for starters it's a derivative of NT, not Win/DOS... :-)
It should be quite a bit more reliable, if the drivers don't crash.
(Disclaimer: I don't own a copy of Win2k.)
>
>Dave
>
>
--
[EMAIL PROTECTED] -- and when were NT and Win9x supposed to merge
together, again? :-)
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: Never tell me again that Windows is easy to install!!! It's a lie!
Date: Mon, 25 Sep 2000 18:32:11 GMT
In comp.os.linux.advocacy, Chris Sherlock
<[EMAIL PROTECTED]>
wrote
on Mon, 25 Sep 2000 23:15:57 +1000
<[EMAIL PROTECTED]>:
>Why would rank newbies want to *use* extfs or xiafs?
Agreed, they would not.
More experienced users might, but only on an experimental basis.
extfs was phased out long ago (it was replaced by ext2fs, and ext3fs
is in the pipeline somewhere). xiafs was once touted as a good
low-memory solution, but apparently that's no longer an issue;
the only problem would be for those that used it in the past.
>
>Chris
>
>The Ghost In The Machine wrote:
>>
>> In comp.os.linux.advocacy, [EMAIL PROTECTED]
>> <[EMAIL PROTECTED]>
>> wrote
>> on Mon, 18 Sep 2000 12:47:47 -0700
>> <8q5s76$vl8$[EMAIL PROTECTED]>:
>> >
>> >Nigel Feltham <[EMAIL PROTECTED]> wrote in message
>> >news:8q5mqf$env3j$[EMAIL PROTECTED]...
>> >>
>> >> How often do linux distro's remove drivers for old hardware in new
>> >> versions - my copy of
>> >> mandrake 7.1 still seems to carry drivers for every device that was
>> >> supported when I had
>> >> my first linux distro (slackware 2.0) back in october 1994 as well as most
>> >> devices invented
>> >> since them except the horrible windblows only type hardware (mainly
>> >>modems).
>> >
>> >I recall the removal of support of one filesystem from Linux, but that is
>> >about it.
>>
>> I count two: extfs and xiafs. However, I think one can probably try
>> to back-patch the current kernel with older source (no warranties
>> here, folks!), as every kernel since about v1.0pl15 is still available.
>> (For some reason, though, ftp.kernel.org doesn't store the complete
>> sources for 1.0.0 through 1.0.9, just the patches. *scratches head*)
>>
>> Whether they're compatible with various aspects of the support system
>> (bread() [*] et al) is not clear to me, since I haven't looked in there
>> lately. But in theory one could go in there and fuss with it; I've
>> wondered myself about what routines would be needed to build a
>> standalone (i.e., user-level code, as opposed to kernel-level) tool
>> that tests/exercises the file system, and/or a Very Stupid File System to
>> go with it. (This VSFS would be extremely braindead, making no attempts
>> at all to optimize allocations of blocks, and would primarily be a tool
>> to study how file systems are accessed in Linux, and how to build
>> one from scratch. One of my many "wish list" projects, I guess... :-) )
>>
>> Of course, expecting a rank newbie to know about this sort of stuff
>> would be ridiculous -- but rank newbies don't remain such for long. :-)
>>
>> [*] Block read, not a baking product. :-)
>>
>> --
>> [EMAIL PROTECTED] -- insert random misquote here
--
[EMAIL PROTECTED] -- insert random misquote here
------------------------------
From: [EMAIL PROTECTED] (Bruce Scott TOK)
Subject: Re: "Overclocking" Is A Bad Idea
Date: 25 Sep 2000 20:26:01 +0200
In article <[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
>On 25 Sep 2000 20:11:35 +0200, Bruce Scott TOK <[EMAIL PROTECTED]> wrote:
>>Don't look now but there are more and more Pentium chips in those Linux
>>clusters people are building. The ones I know are running plasma
>>physics codes to do such things as MHD stability, turbulence, or
>>detailed particle motion in complicated magnetic surface geometries.
>>Typical runs are on 16 or 32 chips. More commonly people use alphas but
>>that is changing with the speed increases of the pentiums. One
>>disadvantage of the alphas is that their caches are too small and you
>>take inordinate penalty for misses (which are inevitable in most
>>codes). Simple PC chips and ethernet parallelism (using LAM MPI) are an
>>excellent way to get a reasonable bang for a small buck.
>>
>>Sparcs are faster, but not _that_ much faster, and they cost more. Same
>>for alphas.
>
> Sparcs are really not that much more expensive these days. Sure
> you can get some pieces and parts for PC's cheaper than just about
> anything else. However, for anything above the point of being your
> own VAR, PC's aren't as cheap as they are represented to be.
For full boxes I agree, but for clusters you may be better off without
full boxes.
If you have to do everything yourself, and you are a physicist first and
computer jock second, building things from motherboards and components
is far easier with the PC stuff. If you are rich enough to afford real
support from Sun, then maybe that is better. It all depends on the
budget and who ends up doing all the work.
--
cu,
Bruce
drift wave turbulence: http://www.rzg.mpg.de/~bds/
------------------------------
From: "Simon Palko" <[EMAIL PROTECTED]>
Crossposted-To:
comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Yeah! Bring down da' man!
Date: Mon, 25 Sep 2000 14:34:25 -0400
<[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Mon, 25 Sep 2000 12:42:02 GMT, Chad Myers <[EMAIL PROTECTED]>
wrote:
> >
> >"K" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]...
> >>
> >> "Boris" <[EMAIL PROTECTED]> wrote in message
> >> news:avRe5.37215$[EMAIL PROTECTED]...
> >> .....
> >> .....
> >> > Because SOAP is W3 standard, it's not os/vendor specific. However,
new
> >> run-time for VB,
> >> > C++, etc. and other elements of development framework (MS Developer's
> >> Studio .Net edition)
> >> > will be supplied for Windows only by MS.
> >> >
> >> > Boris
> >>
> >> What Microsoft is attempting to do here is extend their control to all
or
> >> much of the internet just like they control millions of users of PC
with
> >> Windows. It might not be vendor but Microsoft will make the
standards and
> >> change them at it's whim.
> >
> >Black helicopters....
> >
> >Give me a break...
>
> Kerberos.
So many people are so misinformed on this issue, that I'm not even going to
correct you except to say that you're wrong. If you want more, let me know,
and I'll lay it out for you. But MS didn't do anything to break their
implementation of Kerberos, or to make it not function with other system.
--
-Simon Palko
"More fun than a barrel of monkeys... with dynamite strapped to their
backs!"
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: Never tell me again that Windows is easy to install!!! It's a lie!
Date: Mon, 25 Sep 2000 18:46:50 GMT
In comp.os.linux.advocacy, [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
wrote
on Mon, 25 Sep 2000 16:15:40 -0000
<[EMAIL PROTECTED]>:
>On Sat, 23 Sep 2000 19:04:41 GMT, The Ghost In The Machine
><[EMAIL PROTECTED]> wrote:
>>In comp.os.linux.advocacy, [EMAIL PROTECTED]
>><[EMAIL PROTECTED]>
>> wrote
>>on Mon, 18 Sep 2000 22:50:28 -0000
>><[EMAIL PROTECTED]>:
>>>On Mon, 18 Sep 2000 16:34:36 +0100, Stuart Fox
>>><[EMAIL PROTECTED]> wrote:
>>>>
>>>>"Tim Kelley" <[EMAIL PROTECTED]> wrote in message
>>>>news:[EMAIL PROTECTED]...
>>>>> Nigel Feltham wrote:
>>>>> >
>>>>> > >This kills me. People bitch about Windows 9x still having legacy DOS
>>>>> > >support, and when they begin to take it out, suddenly those same people
>>>>> > >bitch because it's gone.
>>>>> > >
>>>>> >
>>>>> > They don't bitch about legacy support in itself being gone, they
>>>>complain
>>>>> > about
>>>>> > the support for their hardware being removed -
>>>>>
>>>>> Exactly. So my video card is obsolete because MS says so? Fuck
>>>>> that. That is the sort of arrogance that is driving a lot of
>>>>> people to try linux.
>>>>
>>>>MS has precisely fuck all to do with whether your video card is supported.
>>>>Your hardware manufacturer decides whether to write the driver, not MS. MS
>>>>may bundle the driver, but they certainly don't write them...
>>>
>>> It is infact, Microsoft's OS and device driver interface.
>>>
>>> Unless you can come up with some compelling reason why MS
>>> should break with the past and make all of those older Win9x
>>> drivers unusable, you're just making feeble excuses for M$.
>>
>>It would not be unreasonable to go with a new driver model that
>>doesn't support VxD's, which are similar to the TSR's (and have some
>>of the same problems) of the bad old DOS days.
>
> That tells one NOTHING of what is actually so wrong about VxD's.
True; if one's not familiar with "load me first" issues with TSRs,
one won't understand that the same issue plagues users of VxDs.
(I base this on comments from Andrew Schullman in his book,
_Unauthorized Windows 95_. I don't know this personally, but there
are a lot of powerful hooking routines in Win95 that can be taken
advantage of by VxDs, apparently -- and they've been in there for
awhile. Of course, two VxDs can try to hook the same routine,
with varying results...and all this is reminiscent of hooking interrupts
with TSR's during CONFIG.SYS and AUTOEXEC.BAT.)
>
> Also, it would be Microsoft's fault for adopting a bad paradigm
> in the first place. It's all ultimately their OS. If they couldn't
> plan well for the future in the past then that is also their fault.
I'm not sure Microsoft gets all the blame, although I don't know who
designed the BIOS. My understanding, for instance, is that the
BIOS still has the CGA pixel routines (INT 10H) to draw -- well, pixels.
Also to write characters on the screen.
But it was so insufferably slow, developers started to write directly
to the vidram (which was either at 0B000:0 or 0B800:0, depending
on card), and got locked into all subsequent designs. It would have
been nice to be able to fetch the RAM base in paragraph:offset form
(say, in DX:BX), but such was never offered to my knowledge.
But you're right, they get most of the blame for doing some strange things
in their Win32 design (holdovers from their not-so-great Win16 design);
couldn't they have at least used FAR pointers rather than shorts
and words? Or did they worry so about upward compatbility that the
structures had to be the same size on all systems? (Is this a partial
explanation for their little-endian dependency, the ability to
save memory structures on disk (as resources)? I wonder..)
>
> They've always been a backwards looking company and now it's biting
> users in the butt.
Yeah, but it makes them profitable. :-) (I'm not sure if users
always want to upgrade their favorite packages just because the
OS gets upgraded; that's a point in their favor if managed well.
Of course, I'm not sure MS always managed it that well, either.)
>
>>
>>Of course, this breaks everything dependent on VxD's. Not good.
>>One would hope that Microsoft made a token attempt to contact the
>>manufacturers with a warning about this issue, as opposed to
>>merely plastering a warning on a webpage somewhere, deep in their
>>websystem.
>>
>>Also, Microsoft has been making noises about merging the Win/DOS and
>>the WinNT lines for years now. I wish they'd just do it already,
>
> They also made noises about OS/2 becoming their OS standard.
Yeppers. Way back when -- Before The Monopoly. :-)
>
>>but legacy issues keep cropping up. :-) This may be one of them,
>>although how one screws up a perfectly good generic 640x480x16 VGA
>>driver is beyond me.
>
>
>--
>
> "Truth is stranger than fiction, because fiction has to make sense."
>
> Atlanta:
> An entire city surrounded by an airport.
It could be worse. Isn't there a city in Spain where the main runway
is bisected by a city thoroughfare/highway? :-)
--
[EMAIL PROTECTED] -- wondering when 747s will be able to STOL, like
Harrier jump jets
------------------------------
From: [EMAIL PROTECTED] (mark)
Subject: Re: So did they ever find out what makes windows98 freeze up all the time?
Date: Mon, 25 Sep 2000 19:41:20 +0100
In article <[EMAIL PROTECTED]>, David M. Butler wrote:
>Pete Goodwin wrote:
>
>> Some of the graphics tools leave a lot to be desired. They all do
>> different things, with different styles and menus. I seem to need to use
>> multiple applications to do what I do on Windows with two.
>
> Quite true... I definately have more luck with graphics apps in Windows.
>I tried a demo for some big commercial graphics package in Linux, but it
>never quite worked. It's just a hobby for me, else I would have remembered
>to include it along with my lack-of-games comment.
>
Yeah, I wouldn't really get involved with big graphics stuff for work
either. Our work systems are telecoms related, which isn't strong on
graphics. It's strong on data management, manipulation, mining etc.,
so we have lots of unix.
The gimp looked pretty impressive when I looked at it, but like yourself,
I'm not expert enough to determine whether it's appropriate for the job.
Maybe someone else can comment?
--
Mark - remove any ham to reply.
(Killed (sigserv (This sig is reserved by another user)))
------------------------------
** 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
******************************
