Linux-Advocacy Digest #740, Volume #32 Sat, 10 Mar 01 11:13:05 EST
Contents:
Re: New Microsoft Ad :-) (Giuliano Colla)
Re: Why Open Source better be careful - The Microsoft Un-American (Peter Hayes)
Re: Macintosh as an alternative to Windows?? (Peter =?ISO-8859-1?Q?K=F6hlmann?=)
Re: Why Open Source better be careful - The Microsoft Un-American (Peter Hayes)
Re: Microsoft & GM (Norman D. Megill)
Re: definition of "free" for N-millionth time (coding isn't programming) (Austin
Ziegler)
Re: GPL Like patents. (Rob S. Wolfram)
Re: GPL Like patents. (Rob S. Wolfram)
Re: GPL Like patents. ("Les Mikesell")
Re: Computing Power to Peak SOON! (WAS: Moore's Law, continued...) (The Ghost In The
Machine)
Re: Customising Wrap-Up Screen. (WAS: "It is now safe to shut off your computer")
(The Ghost In The Machine)
Re: Windows API (Was Re: Mircosoft Tax) (Bob Hauck)
Re: Macintosh as an alternative to Windows?? (Bob Hauck)
----------------------------------------------------------------------------
From: Giuliano Colla <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.nt.advocacy
Subject: Re: New Microsoft Ad :-)
Date: Sat, 10 Mar 2001 12:12:49 GMT
"Norman D. Megill" wrote:
>
> In article <98c190$hc2$[EMAIL PROTECTED]>,
> Ayende Rahien <[EMAIL PROTECTED]> wrote:
> > "T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > Said nuxx in alt.destroy.microsoft on Tue, 6 Mar 2001 15:03:44 +0800;
> > > >"Edward Rosten" <[EMAIL PROTECTED]> wrote in message
> > > >news:97miug$8dr$[EMAIL PROTECTED]...
> [snip]
> > > >> As an aside, I don't know why NIC card manufacturers haven't put a
> > > >> machanism on the crads to detect an unplugged cable. It shouldn't be
> > too
> > > >> hard since when plugged in, the cable is plugged in to a matched load,
> > so
> > > >> no reflections occur. When it is unplugged, the signals should get
> > > >> reflected, which should not be too hard to detect.
> > > >
> > > >Win2k has this built in - disconnect the cable and see what happens.
> > Works
> > > >well.
> > >
> > > Let me guess; it pops up one of those "is it modal or isn't it?" dialog
> > > boxes, requiring clicking the OK button before you can do anything else
> > > (except shift windows around) every one minute when it detects a link
> > > loss.
> > >
> > > Or maybe it just pops it up once, but thereafter there is no way at all
> > > to tell the difference between the cable not being connected, the
> > > hardware not working, or the driver failing.
> > >
> > > Or else it says "someone may have unplugged the cable" for any link
> > > loss.
> > >
> > > I would be very surprised if it "works well", quite frankly.
> >
> > It pop up a floating message saying "cable is unplugged" (which also happens
> > if the computer/hub on the other side was disconnected or turned off).
> > There is no way you can mistake that for hardware or driver failure.
>
> If the message doesn't tell you *which* side was unplugged, then it is
> not making use of reflections from unmatched loads but just telling you
> that the signal is dead - the message is probably just based on the same
> status bit that lights the little LED (that some cards have) when you
> plug in the cable.
>
> I think older coax Ethernet chips had a built-in TDR (Time Domain
> Reflectometry) function that could locate a cable discontinuity. I
> don't think the more popular RJ45 (10Base-T) chips have this function,
> so it may not be possible even in principle for them to tell you if
> there are reflections from an unmatched load. Someone correct me if I'm
> wrong.
>
> (Speaking of modal dialog boxes, I truly hate it when I'm typing along
> and accidentally "answer" a box that pops up unpredictably so that it
> instantly vanishes before I can read it. Yesterday, Windows 95
> bluescreened on me, and because I was typing when it happened the "Press
> Any Key to Continue" was answered.)
>
One day Microsoft (hopefully) will learn the concepts of
"scheduling", "priority" etc. and they will begin to work
out an OS.
--
Giuliano Colla
Before activating the tongue, make sure that the brain is
connected (anonymous)
------------------------------
From: Peter Hayes <[EMAIL PROTECTED]>
Subject: Re: Why Open Source better be careful - The Microsoft Un-American
Date: Sat, 10 Mar 2001 12:46:54 +0000
Reply-To: [EMAIL PROTECTED]
On Thu, 8 Mar 2001 01:16:55 -0800, "GreyCloud" <[EMAIL PROTECTED]> wrote:
<...>
> I find this topic of great interest! You have a very strong point here! Do
> you realize that only 10% of
> earths' biomass is on the surface and that the rest is deep down under foot?
> There is a fellow that is
> researching a new theory that I believe to be true and also seems to drive
> the political engine of oil.
> This scientists' claim is that oil is produced by the underground bio-mass
> of bacteria. Its a self renewing resource
> and I suspect the oil industry top execs know this
I've read something about this elsewhere, and the ecological implications
are immense. We're already detecting the effects of global warming. The sea
off Scotland's west coast is 1 deg C warmer than 10-15 years ago.
http://news.bbc.co.uk/hi/english/uk/scotland/newsid_1208000/1208372.stm
So an inexaustible supply of crude oil may well soon (within <50-100 years)
tip the global climate into a hothouse to rival Venus'.
But the oil execs are only interested in profit and bigger BMWs.
Peter
------------------------------
From: Peter =?ISO-8859-1?Q?K=F6hlmann?= <[EMAIL PROTECTED]>
Subject: Re: Macintosh as an alternative to Windows??
Date: Sat, 10 Mar 2001 11:16:53 +0100
Erik Funkenbusch wrote:
> Apparently you never programmed Macs in the early days. It was enough to
> pull your hair out. People say the Windows API is ugly, but the MacOS API
> was much worse. The MacOS was originally written in Pascal, and much of
> the
No, I did not. Your point? I wrote about the new OS, not the old one.
> C/C++ functionality had to emulate Pascal. In addition, since the Mac
> didn't have virtual memory, they had to have some way to move memory
> around with the app going bezerk, so they created something called
> "Handles" (only slightly similar to descriptors) in which they were
> pointers to pointers. Any memory you allocated had to be derefenced twice
> to access, since the MacOS kept track of the handles and could move the
> memory anywhere it wanted to.
I know all this stuff. Your point?
>
> In addition, most major revisions of the OS up until about MacOS 7 caused
> significant backwards compatiblity issues. Apple broke API's at their
> whim
> (not by accident either, they did it on purpose). It's a shame that they
> couldn't carry over their ferocious defense of the single menu bar into
> standards for programming the OS.
>
I repeat, I wrote about the new OS, not the old one.
> Don't you think 2001 and is a little late to finally be coming out with
> Pre-emptive multitasking, Memory protection, and full virtual memory?
>
> Cripes, Apple worked on their Next Generation OS (Copeland) for close to
> 10 years before finally giving up and admitting defeat that their
> programmers
> just couldn't "fix" the MacOS. They needed to start over from scratch.
>
Well, Apple in that case had the grace to give up. MS did not, they
stuffed Win9X unto the world.
Peter
--
Windows is just the instable version of Linux for users who are too
dumb to handle the real thing.
Microsoft's Product Strategy: "It compiles, let's ship it!"
------------------------------
From: Peter Hayes <[EMAIL PROTECTED]>
Subject: Re: Why Open Source better be careful - The Microsoft Un-American
Date: Sat, 10 Mar 2001 12:51:03 +0000
Reply-To: [EMAIL PROTECTED]
On Tue, 06 Mar 2001 07:23:52 -0500, Aaron Kulkis <[EMAIL PROTECTED]> wrote:
> Bloody Viking wrote:
<...>
> > Oil production WILL max out - and soon. The car companies know it. THAT is why
> > only now do you see petrol-electric cars on sale.
>
> The hybrid vehicles are just as dependant upon petroleum as before.
They don't have to be. Electrolysis converts water to hydrogen and oxygen
relatively efficiently. Run your Otto cycle engine on H and O. No nasty
by-products. No need for energy sapping catalytic converters. Or use the H
and O in fuel cells, but maybe they aren't powerful enough yet.
Supplement the H and O engine or fuel cells by charging the batteries with
braking effect and with more efficient solar panels. Eakes out more miles
per charge.
I read somewhere there wasn't enough copper in the world to wind the coils
for the motors for all the electric vehicles that will be needed. That was
some time ago, maybe it's not the case now.
Peter
------------------------------
Crossposted-To: alt.conspiracy.microsoft,alt.destroy.microsoft,alt.microsoft.sucks
Subject: Re: Microsoft & GM
From: [EMAIL PROTECTED] (Norman D. Megill)
Date: Sat, 10 Mar 2001 13:11:08 GMT
In article <98cs5f$bnr$[EMAIL PROTECTED]>,
Terence Kam <[EMAIL PROTECTED]> wrote:
> 13. You'd have to press the "Start" button to turn the engine off.
There is a Stop button. But as typical for Windows it is binary-only
and you have to purchase it as third-party software.
http://www.softwrights.com/downstop.htm
I guess the binary-only mystique of Windows gets people to PAY for
the most trivial things. Probably the code that handles the 15-day
free trial timeout and nag screens is bigger than the app itself.
--Norm
------------------------------
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy,misc.int-property
From: Austin Ziegler <[EMAIL PROTECTED]>
Subject: Re: definition of "free" for N-millionth time (coding isn't programming)
Date: Sat, 10 Mar 2001 09:48:49 -0500
On 10 Mar 2001, Steve Mading wrote:
> In comp.os.linux.advocacy JD <[EMAIL PROTECTED]> wrote:
>: Every time people bring up things like 'modifying TCP/IP' or
>: changing 'Kerberos', it is the SPECIFICATION that is vulnerable.
> There's no difference, conceptually. The specs are source code for
> programming the programmers.
In reality, there's a significant difference. The specs identify the
minimum capabilities -- and places where variation is
allowed/disallowed, etc. Unless there's no variation allowed, then the
spec isn't a "program" but a roadmap. One doesn't need to change and
redistribute the spec to implement it, just obtain it.
-f
--
austin ziegler * Ni bhionn an rath ach mar a mbionn an smacht
Toronto.ON.ca * (There is no Luck without Discipline)
=================* I speak for myself alone
------------------------------
From: [EMAIL PROTECTED] (Rob S. Wolfram)
Subject: Re: GPL Like patents.
Date: 10 Mar 2001 13:30:11 GMT
Reply-To: [EMAIL PROTECTED]
Les Mikesell <[EMAIL PROTECTED]> wrote:
>I don't understand why the communication method between two components
>should determine whether they become 'derived' from each other or
>not.
I'd say that the communication method determines if the two pieces of
code form one single program during runtime.
> Would using a java method directly be different than using
>the same via RMI?
Sorry, can't comment. I have no idea how RMI works.
> Is there a difference between using a .dll in
>process or running it in another process?
How do the two processes communicate? Via RPC-like constructs? Then I'd
say the two processes are not the same program.
> What about dynamic
>linking via dlopen() where the code in question decides at runtime
>to link in another library - often ones that didn't even exist when
>the calling code was written? How can something be 'derived' from
>other code written in the future?
This just moves the point of focus from the program to the library. In
such case the library has to be GPL compatible for you to link to it.
Cheers,
Rob
--
Rob S. Wolfram <[EMAIL PROTECTED]> OpenPGP key 0xD61A655D
Anyway the :// part is an 'emoticon' representing a man with a
strip of sticky tape across his mouth.
-- R. Douglas
------------------------------
From: [EMAIL PROTECTED] (Rob S. Wolfram)
Subject: Re: GPL Like patents.
Date: 10 Mar 2001 14:11:24 GMT
Reply-To: [EMAIL PROTECTED]
Roberto Alsina <[EMAIL PROTECTED]> wrote:
>Rob S. Wolfram wrote:
>> IANAL, but upon reading section 0 I get the clear impression that both
>> static and dynamic linking do taint.
>
>Section 0 is not a legally binding part of the license. It's a preamble.
Nope. Section 0 is the first section of the "terms and conditions" of
the GPL and mainly sets the definitions for the words to be used in
later sections. Yes, it is legally binding. The preamble precedes the
terms and conditions and is clearly marked as such.
>The issue is not derivative work, because it's not the GPL that defines
>that. It's "greater work" or "larger work", which the GPL does. There is no
>doubt that a dynamically linked program is not a derivative work of the
>library in all cases, only in some cases.
The phrases "greater work" or "larger work" are not present in the text
of the GPL, not even in the preamble or the "how to apply" section.
A copy and paste from section 0:
] The "Program", below,
] refers to any such program or work, and a "work based on the Program"
] means either the Program or any derivative work under copyright law:
] that is to say, a work containing the Program or a portion of it,
] either verbatim or with modifications and/or translated into another
] language.
Indeed, the GPL does not define "derivative work". Copyright law does.
The GPL merely re-iterates it.
>RMS says using a component through CORBA is not tainting. I can dig a
>reference if you really want it. That's how GNOME expected to use the
>Netscape code when it was not dually licensed under the GPL.
>
>It does make the GPL kinda useless, since almost anything can be turned
>into a omponent.
That's good to hear.
However I think the time is very due to stop referring to how RMS
interprets the license, but rather how a judge interprets it. Also I
think this should happen in both the U.S. and in Europe, because the
copyright laws between these are a bit, ... er ... different ;-) (DMCA
anyone? With an ounce of UCITA perhaps? ;-))
>I have. I regret ever releasing a line of code under the GPL. I know I am
>not the only one ;-)
I'm sure you're not. But if I ever release a piece of code I most likely
will do so under the GPL. I kinda agree with Stallman's thinking. I do
respect though, that not everyone thinks that way and they should be
free not to.
Cheers,
Rob
--
Rob S. Wolfram <[EMAIL PROTECTED]> OpenPGP key 0xD61A655D
The number of UNIX installations has grown to 10, with more
expected.
-- The Unix Programmer's Manual, 2nd Edition, June 1972
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Subject: Re: GPL Like patents.
Date: Sat, 10 Mar 2001 15:09:54 GMT
"Roberto Alsina" <[EMAIL PROTECTED]> wrote in message
news:98aqj0$15jac$[EMAIL PROTECTED]...
> Les Mikesell wrote:
>
> >
> > "Roberto Alsina" <[EMAIL PROTECTED]> wrote in message
> > news:98al9u$16uuu$[EMAIL PROTECTED]...
> >
> >> > Upon describing a "derivative work"
> >> > there is no timeframe specified when the "derivative work" should
> >> > contain (a portion of) the Program, so you should consider the worst
> >> > possible case, i.e. in any moment, even only during runtime.
> >>
> >> The issue is not derivative work, because it's not the GPL that defines
> >> that. It's "greater work" or "larger work", which the GPL does. There
is
> > no
> >> doubt that a dynamically linked program is not a derivative work of the
> >> library in all cases, only in some cases.
> >
> > What RMS or even the GPL itself says in regard to what is a derivative
> > is pretty much irrelevant. This must be defined by copyright law
itself,
> > since the GPL puts no restrictions on use,
>
> Oh, but there's a catch there. The GPL SAYS it puts no restrictions to
use.
> Ask RMS about "user-does-the-linking", and he will come up with something
> that looks suspiciously similar to a restriction in use:
Yes, any reasonable interpretation would see it that way, but no one
has accused RMS of being reasonable. His opinion is based on the
view that a program that requires a library that is under the GPL is
a derived work even though it does not contain the covered material.
> I asked RMS if a program developed to use the intergace of GDBM (GPLd
> library) and compiled against a binary compatible do-nothing library would
> be legal.
>
> Consider that in this case I, the developer, don't even need to own a copy
> of the GDBM lib. Since I have never licensed it under the GPL, I can NOT
> violate the GPL :-)
>
> However, he told me that when the user receives the binary of my program
> and installs it on his system, and then the system links it to the GDBM
> library, there is a violation of the GPL.
>
> I was honestly sick of the argument so I didn't push it further, but since
> the developer never licensed GDBM, the GPL violator must be the user.
Since
> the user is not making copies, distributing, etc, he must be violating it
> by use, which the GPL specifically says he can't do.
Note that even RMS agrees that if there are alternatives to the GPL library
that could be used, then it is clearly a user's decision at link or run time
and not an attribute of the other source to use GPL'd routines. If you
follow the RIPEM case through you will see how this works. The
RIPEM code wanted to use the GPL'd gmp math library, but not
distribute it together in order to have less restrictions on how their
own code could be used. However, because of RMS's threats over
this exact issue, they were forced to write 'fgmp' as compatible
substitute library. Fgmp is not as efficient as gmp and may never
have actually been used by anyone, but in RMS's mind the existence
or lack therof of this unrelated work determines whether or not
code using the interface to gmp is a derived work under copyright
law.
> This is one example of the things RMS claims are illegal and that I can't
> see any remotely reasnoable way that can be so. I belive RMS to be
> intelligent. I believe he is lying to himself.
He has explained why he thinks using an interface creates a derived
work in the circumstance where only one library provides that
interface. I have yet to hear of anyone else who agrees with that
view...
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
alt.destroy.microsoft,alt.microsoft.sucks,comp.os.ms-windows.nt.advocacy
Subject: Re: Computing Power to Peak SOON! (WAS: Moore's Law, continued...)
Date: Sat, 10 Mar 2001 15:29:23 GMT
In comp.os.linux.advocacy, Bloody Viking
<[EMAIL PROTECTED]>
wrote
on 10 Mar 2001 05:06:56 GMT
<98ccpg$7ht$[EMAIL PROTECTED]>:
>
>The Ghost In The Machine ([EMAIL PROTECTED]) wrote:
>
>: Dumb question maybe, but ... how do they prevent the water from
>: picking up ions, like, say, from rusting pipes? (I suspect they
>: don't use iron piping, though -- but copper could do the same thing, and
>: maybe PVCs would put CO2 (actually carbonic acid) ions in there as well.)
>
>: Do they filter it on the trip back?
>
>Actually, that's a good question. It will get in a case of a high-priority
>system like a broadcast transmitter, like a powerplant. A powerplant will
>take the condensate water and send it into a deminerialiser to further
>clean it up after the boilers.
>
>For a good high power watercool system like a large broadcast transmitter on
>the anode, you will want distilled demineralised water in the system. For an
>overclock PC, you will want quality water but can't get it, much like how you
>can't for a car cooling system but rely on coolant chemistry.
Actually, there's distilled water available; it's for steam irons and such
(same reason). It might even be demineralized.
It's available in gallon jugs.
>
>If I was to attempt overclocking on a watercool CPU, I would likely go to
>alcohol cool and ensure it was demineralised. Alcohol cool has the advantage
>of cold temps. But cold temps have the problem of defrosting the computer
>after a batch run. No good. Alcohol would have the advantage of a cleaner
>footprint come a leak/spill. The spill problem is mostly why I would use
>Everclear booze in a liquid cool system. My computer wouldn't be "aquatic".
>Instead it would be "alcoholic"! I'm about as aquatic as a chimpanzee, so why
>should I own a computer more "aquatic"?
Just as a rather dumb observation: I'm not at all sure how one would work
around this, but there's a liquor tax on pure alcohol under some
conditions (denaturants are placed in such things as isopropyl alcohol
to render it unfit for drinking; these denaturants show up as turbidity
when the alcohol is poured into water (or vice versa, I forget which);
I don't know if they'd affect its use in a cooler such as you describe).
As another dumb question: I take it the alcohol doesn't change physical
state while warming up (cooling down the chip), unlike, say, ammonia
or other such refrigerants. (The thought of a computer having a noisy
compressor doesn't really thrill me, admittedly... :-) ) Or does it?
Alcohol evaporates rather rapidly...
>
>--
>FOOD FOR THOUGHT: 100 calories are used up in the course of a mile run.
>The USDA guidelines for dietary fibre is equal to one ounce of sawdust.
>The liver makes the vast majority of the cholesterol in your bloodstream.
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 33d:22h:56m actually running Linux.
This space for rent.
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: alt.destroy.microsoft
Subject: Re: Customising Wrap-Up Screen. (WAS: "It is now safe to shut off your
computer")
Date: Sat, 10 Mar 2001 15:35:01 GMT
In comp.os.linux.advocacy, Ayende Rahien
<[EMAIL PROTECTED]>
wrote
on Sat, 10 Mar 2001 11:42:26 +0200
<98ctvp$ar4$[EMAIL PROTECTED]>:
>
>"Bloody Viking" <[EMAIL PROTECTED]> wrote in message
>news:98cbo4$75t$[EMAIL PROTECTED]...
>>
>> I just got done customising the Windows 95 wrap-up screen, the "it's now
>safe
>> to turn off your computer" screen. It now says:
>>
>> It's now safe to type "mode co80 and light off UNIX, the OS Bill Gates
>hates!
>> GNU's Not UNIX!!!
Dunno if "mode co80" works anymore. :-)
>>
>> Thanks! That was after a few hours of quality coding time working on a pet
>> snail billpay proggie in C on Linux. For what it's worth, Linux IS UNIX in
>my
>> book, it's a GNU freeware UNIX.
>>
>> Ah, the pleasure of having the OS of Big Iron on a PC. UNIX is the OS of
>Big
>> Iron computing, and while we may enjoy it on our boxes, it will always be
>THE
>> OS of Big Iron. How could anyone pass up the chance to play with an OS
>like
>> Linux, a PC freeware UNIX? Maybe some of us are hackers (in the good sense
>of
>> the word) after all. (:
>
>Why won't you get Solaris? A true Unix, free (unless you got a monster for a
>workstation) and work on x86.
What's wrong with Linux? Solaris is good, admittedly, but there's
nothing wrong with Linux, and it's probably more readily available.
There is one technical point -- and Linus (or someone else) might have
fixed it by now: Linux tends to allocate memory it doesn't have and then
do a kill -9 on a process when it faults a page and there aren't any
more available (RAM is used up and swap is exhausted). Solaris doesn't
have this problem, but the flip side is that one has to declare all
swap space: if the workstation has 128M and a program wants 512M of
memory, there'd better be 384M swap available (at least!) or the program's
request will fail, regardless of whether the program actually needs all
that memory (it might be using some sort of sparse addressing scheme,
for example). Admittedly, one could go either way with this.
(IIRC all this, anyway.)
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 33d:23h:03m actually running Linux.
I was asleep at the switch the rest of the time.
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Crossposted-To: alt.linux.sux,alt.destroy.microsoft
Subject: Re: Windows API (Was Re: Mircosoft Tax)
Reply-To: bobh = haucks dot org
Date: Sat, 10 Mar 2001 15:43:42 GMT
On Sat, 10 Mar 2001 03:43:56 +0200, Ayende Rahien <[EMAIL PROTECTED]> wrote:
>
>"Bob Hauck" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> On Thu, 08 Mar 2001 06:57:04 -0000, Ray Chason
>> <[EMAIL PROTECTED]> wrote:
>> > You really need that widget set. From here springs the great weakness
>> > of X as a user interface. We have GTK+ and Qt, and they don't look
>> > the same and they don't interoperate seamlessly.
>>
>> Oh, it is much, much, worse than that. There is also athena, athena-3d,
>> motif (v1 and v2), openlook, tk, and probably more that I've forgot.
>> How will we ever manage?
>
> How about standatising on something?
That is happening, with KDE and Gnome being the ones that seem to be
winning. OTOH, if a program was written with an older widget set it is
unlikely that the maintainers will change it just for the sake of being
"standard".
I find it interesting that Windows is going the other way, becoming less
standard by allowing users and developers to "skin" programs.
>Seems to work pretty well on communication area.
Phones used to all be black, now they aren't. Doesn't seem to have
impaired functionality. All of the widget sets I mentioned talk to X,
just as all phones connect to the phone network.
--
-| Bob Hauck
-| To Whom You Are Speaking
-| http://www.haucks.org/
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: Macintosh as an alternative to Windows??
Reply-To: bobh = haucks dot org
Date: Sat, 10 Mar 2001 15:43:54 GMT
On Sat, 10 Mar 2001 02:14:41 -0600, Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
>"Peter Köhlmann" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> As much as I hated Apple for the look-and-feel - stuff they did years
>> ago, they DID have decent hardware and a quite capable OS. Better
>> than anything MS could bring forward, that is.
>Apparently you never programmed Macs in the early days.
No, but I programmed Windows in real mode. In the early days.
> In addition, since the Mac didn't have virtual memory, they had to
> have some way to move memory around with the app going bezerk, so they
> created something called "Handles" (only slightly similar to
> descriptors) in which they were pointers to pointers.
Golly, that sounds really familiar. In fact, it is identical to Windows
in real mode.
Then, when Win 3.1 came out and real mode was declared dead, we still
had a bunch of limitations related to the real mode architecture. How
about near vs far heaps, limited "system resources", only one instance
of large- model programs, and reliance on DOS memory for system data
structures? Those lived on even in Win95 to some degree.
Seems to me that if what you say is true, Windows copied much of it's
badness from the Mac. MS can't even improve that which it steals.
> In addition, most major revisions of the OS up until about MacOS 7
> caused significant backwards compatiblity issues. Apple broke API's
> at their whim
For crying out loud man, they changed cpu families in there! MS has had
the benefit of running on the same cpu family the whole time and still
has had many of the same issues.
>Don't you think 2001 and is a little late to finally be coming out with
>Pre-emptive multitasking, Memory protection, and full virtual memory?
Finally he has a point. Yes, they should have done OS-X years ago.
And, in fact, they tried. Several times.
--
-| Bob Hauck
-| To Whom You Are Speaking
-| http://www.haucks.org/
------------------------------
** 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
******************************
