Linux-Advocacy Digest #409, Volume #32 Thu, 22 Feb 01 16:13:04 EST
Contents:
Re: Whistler/.NET will Help Linux (Steve Mading)
Re: "Linux Usage Linked to AMD Damage" - MSN R&D Division Customer Alert ("Matthew
Gardiner")
Re: Whistler/.NET will Help Linux ("Erik Funkenbusch")
Re: SSH vulnerabilities - still waiting [ was Interesting article ] (The Ghost In
The Machine)
Re: The Windows guy. ("Erik Funkenbusch")
Re: Interesting article (Steve Mading)
Re: How Microsoft Crushes the Hearts of Trolls. (Steve Mading)
Re: How Microsoft Crushes the Hearts of Trolls. (Steve Mading)
Re: How Microsoft Crushes the Hearts of Trolls. (Steve Mading)
Re: How Microsoft Crushes the Hearts of Trolls. (Steve Mading)
Re: How much do you *NEED*? ("Angel Iglesias")
Re: The Windows guy. ("Erik Funkenbusch")
Re: The Windows guy. ("Erik Funkenbusch")
Re: Microsoft backs out of Corel ("Matthew Gardiner")
Re: Whistler/.NET will Help Linux ("Erik Funkenbusch")
----------------------------------------------------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Subject: Re: Whistler/.NET will Help Linux
Date: 22 Feb 2001 20:48:15 GMT
Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
: "Steve Mading" <[EMAIL PROTECTED]> wrote in message
: news:96uudk$e2i$[EMAIL PROTECTED]...
:> The Unix device driver and file system is an object-oriented
:> interface: You have universal operations that work on all
:> device drivers: read/write/open/close.
: You're talking about polymorphism. Polymorphism alone does not make
: something OO. For instance, Visual Basic used polymorphism via COM
: interfaces quite well. In order to be truly OO you need to support the
: entire range of OO principles.
In that definition, very little in computers is OO, since even OO
languages themselves often are missing one or two OO principles.
You're setting the bar so high that almost nothing will clear it.
: You could say the same thing about Windows,
: since it too uses basic functions to operate on both files and devices.
So can I use the commandline "type" command to output a file to a tape
drive? No? Can I select a raw partition as the place to spool off a
.ZIP file I'm making with PKZIP? Didn't think so. Thanks for playing.
Windows does NOT have the same degree of polymorphism as UNIX in this
regard.
: In any event, the capability you're talking about is not an inherant part of
: the file system, but rather of the kernel model. A file handle is a file
: handle, regardless of what file system is used. Remember, Linux can run on
: FAT. Are you going to suggest that FAT is OO?
If you are going to use the literal meaning of "filesystem" like that,
then the phrase "OO filesystem" is inherently meaningless. The filesystem
is just a data format. That can never be OO, regardless of OS. Bitching
that unix has no OO filesystem then becomes a stupid inherently
meaningless utterance, like complaining that you don't like the taste
of blue, or the sound of purple.
I was giving more benefit of the doubt than was warranted, it seems, by
assuming that that *wasn't* what was meant by 'filesystem', and that
the wrong vocabulary term was being used and the meaning was actually
something a bit more high-level than the filesystem.
------------------------------
From: "Matthew Gardiner" <[EMAIL PROTECTED]>
Subject: Re: "Linux Usage Linked to AMD Damage" - MSN R&D Division Customer Alert
Date: Fri, 23 Feb 2001 09:48:41 +1300
Do you know what the really sad part about it is....it could actually be a
press release in the pipe line!
Matthew Gardiner
"BillyG." <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Customer Alert:
>
> As Chief Architect for MSN R&D, I must finally answer, clarify and perhaps
> offer appropriate solutions to the dangers involved in running Linux on
AMD
> processors. Many concerned customers have written statements similar to
> the following: "Lately, I have heard many [... censored ...] things about
> linux in the news. But, is it true that Linux can cause damage to AMD
> Athlon chips?"
>
> Yes, this certainly is the case. That's why care and vigilance must be
> applied to the maintenance of AMD chips. Go to
> http://rd.msn.com/linux-shortcomings/amd/default.asp and purchase the
> latest version of WinCoolant.exe and install. WinCoolant.exe will help
you
> monitor your CPU's coolant level. If low, WinCoolant will notify you via
> MSN email. The authorized way to check for leaks is to place a piece of
> MSN-Certified Cardboard (#MSN-9735674cb) under your computer and wait
> overnight. Then, look for reddish wet spots. To replace lost coolant,
> carefully mix a 30-40-30% solution of purified water - ethylenstugel* -
> butilpharol phenuxum*. Use only XP-HygroGauge Ver2.4 (#MSN-7725943hg) to
> verify concentration ratio. Refill the heat-exchanger through the nozzle**
> located on back of motherboard. Check the level again with XPDipstick
> Ver4.6 or WinCoolant and refill if needed.
>
> -Billy G
> Chief Architect, MSN R&D
>
>
> *Proprietary Coolant Solution distributed solely by MSN R&D - a Division
of
> Microsoft Corp, Redmond WA (Only available to MSN customers on 5-year
> service agreement).
>
> **Coolant Nozzle on CPU may be locked. See your nearest MSN dealer for
> purchase of nozzle-key. (Update: Nozzle-keys are only available when
> bundled with XPW2K1-SE; server edition. Financing available.)
>
> As you can see, this MSN R&D patented heat exchanger system is not only a
> CPU Cooling System upgrade; it's a Lifestyle Upgrade.
>
> +++This press-release in no way acknowledges the existence of leaky
> MSN-CPU-Cooler gaskets. In case of MSN-Lab-Verified leaky gaskets, we
will
> work quickly to make available the necessary Service-Package within six
> months. Pricing on the Service-Package will be determined at time of
> release.
>
>
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: Whistler/.NET will Help Linux
Date: Thu, 22 Feb 2001 14:59:26 -0600
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Erik Funkenbusch in comp.os.linux.advocacy on Wed, 21 Feb 2001
> >"Ed Allen" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> In article <Ntqk6.392$[EMAIL PROTECTED]>,
> >> Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
> >> >"Ed Allen" <[EMAIL PROTECTED]> wrote in message
> >> >news:[EMAIL PROTECTED]...
> >> >> In article <Xqmk6.374$[EMAIL PROTECTED]>,
> >> >> Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
> >> >> >
> >> >> >I happen to agree that people in their garages can overthrow MS.
Most
> >> >Linux
> >> >> >people seem to think they can do just that. Congress and the DOJ
> >doesn't
> >> >> >seem to think that is true.
> >> >> >
> >> >> What you don't seem to understand is that Microsoft is lying
about
> >> >> how likely it is that a few people in a garage will survive long
> >> >> enough to affect them.
> >> >
> >> >[ followed by...]
> >> >
> >> >> Open Source will eventually topple Microsoft whether the program
is
> >> >> Linux or something else. These are impassioned people who are
> >> >> willing to contribute what little they can because it gets
returned
> >> >> many times over.
> >> >
> >> >Do you enjoy contradicting yourself?
> >> >
> >> I did not contradict myself. You seem quick to jump to conclusions
> >> and ignore what else is before you.
> >>
> >> Open Source is not limited to a single garage it is supported by a
> >> community of developers far larger than any commercial software
firm
> >> could afford spread across the entire world in most cases.
> >
> >And nowhere in any of my messages did I make reference to this fictitious
> >"single garage" that your theory is based on.
> >
> >In fact, I stated specifically "kids" (plural) and "garages" (plural).
>
> So if you make your rhetoric vague enough, your argument becomes
> irrefutable? You are the one who misunderstood, Ed's comments, Erik.
Ed was responding to *MY* comment, and insinuating that I had claimed that
it was a single garage. I didn't.
> >> The Open Source community is now millions strong, time for some
> >> reorganization.
> >
> >And completely irrelevant to the thread.
>
> But not the comment you remarked upon, oddly enough. Whether that does
> or does not make it relevant to "the thread" is something of an
> epistemological argument, don't you think?
The comment was in regards to my comment. I certainly have the right to
refute it as being irrelevant to something I said.
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.security.ssh
Subject: Re: SSH vulnerabilities - still waiting [ was Interesting article ]
Date: Thu, 22 Feb 2001 20:55:41 GMT
In comp.os.linux.advocacy, Shane Phelps
<[EMAIL PROTECTED]>
wrote
on Wed, 21 Feb 2001 21:09:20 +1100
<[EMAIL PROTECTED]>:
>
>
>Chad Myers wrote:
>>
>> "Theo de Raadt" <[EMAIL PROTECTED]> wrote in message
>> news:[EMAIL PROTECTED]...
>> > T. Max Devlin <[EMAIL PROTECTED]> writes:
>> >
>> > > Said Charlie Ebert in comp.os.linux.advocacy on Mon, 19 Feb 2001
>> > > [...]
>> > > >What is *THIS* fucking bullcrap about Chad?
>> > > >
>> > > >Why?
>> > >
>> > > Because he is a troll who, for reasons mysterious to the rest of us,
>> > > finds validation in being replied to, regardless of what the reply is or
>> > > what things he has to say, be they facts, inflammatory opinions,
>> > > misrepresentations, or outright lies, in order to get it.
>> >
>> > No no no.
>> >
>> > He's not a troll. He's a net-kook.
>>
>> More insults and more avoidance of facts from Theo de Raadt.
>>
>> Thank god I don't use any of your software. I value quality and security.
>>
>> -Chad
>
>
>Please enlighten us Chad.
>
>What do you regard as secure??
>
>... and why, pray tell?
Secure is when the green glowing indicator light on the complex
but easy-to-use GUI front control panel shows that all is well.
Secure is when the hasp of the lock is closed and the lock is
glowing with a yellow aura, indicating that everything's OK.
Secure is when the anti-sarcasm detector isn't going BEEP BEEP BEEP ...
oh, waitaminnit, that's my sarcasm *generator* that's doing that.... :-) :-)
--
[EMAIL PROTECTED] -- "hello Chief? Maxwell Smart here...."
EAC code #191 17d:12h:26m actually running Linux.
Darn. Just when this message was getting good, too.
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: The Windows guy.
Date: Thu, 22 Feb 2001 15:07:35 -0600
"Donovan Rebbechi" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Wed, 21 Feb 2001 18:47:47 -0600, Erik Funkenbusch wrote:
> ><[EMAIL PROTECTED]> wrote in message
>
> >Should it? Are you guessing? Have you actually read this in the book?
> >Simply provide a page reference.
>
> I take it we're talking about "Operating System Concepts", Silberschatz/
> Galvin.
Yes.
> Try P659, last paragraph:
>
> ----
> "The simplest form of communication between processes is by
> *pipes*,
>
> ...
>
> A pipe is essentially a queue of bytes between two processes".
>
> ----
>
> From this it seems that they are making it pretty clear that a pipe
> is a form of IPC. It is not simply reading output from one command
> and using that as input for another.
There are several kinds of pipes actually. For instance, Win32 has Named
pipes as well as shell pipes. Shell pipes, as being referred to here, are a
byproduct of the shell diverting the output of one program into the input of
another.
But still, the definition does not state anywhere about the necessity of
multitasking to be there for it to be a "real" pipe. The definition also
holds for the temporary file used in MS-DOS.
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Interesting article
Date: 22 Feb 2001 20:51:36 GMT
In comp.os.linux.advocacy Les Mikesell <[EMAIL PROTECTED]> wrote:
: "Steve Mading" <[EMAIL PROTECTED]> wrote in message
: news:96v133$im8$[EMAIL PROTECTED]...
:>
:> It demonstrates exactly the opposite.
: How do you get that? If it has to be done over it means it was done wrong
: the first time. Do you want to encourage users to make their own
: mistakes with the security of their files?
Some things have to be done over and over even if they are done right
the first time, because circumstances change. In my example, you will
note, I described a department with new people coming into it that
weren't there when the groupd was first made.
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: How Microsoft Crushes the Hearts of Trolls.
Date: 22 Feb 2001 20:54:25 GMT
In comp.os.linux.advocacy Bob Hauck <[EMAIL PROTECTED]> wrote:
: On 20 Feb 2001 23:49:33 GMT, Steve Mading <[EMAIL PROTECTED]>
: wrote:
:> I want to be able to SEE the layout of the code.
: You want Python. <http://www.python.org/>
No, I don't want python, because I don't like enforced indenting
rules. The best way to express things is not always with
rigid indenting rules.
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: How Microsoft Crushes the Hearts of Trolls.
Date: 22 Feb 2001 20:56:08 GMT
In comp.os.linux.advocacy Johan Kullstam <[EMAIL PROTECTED]> wrote:
: you can do that. [] and {} are available for user use. you can
: fairly easily hack the lisp reader to accept any of () [] or {} as
: parentheses.
It would have helped greatly if the person who taught it to me had
actually mentioned this. That would have gone a long way toward
improving my opinion of the syntax.
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: How Microsoft Crushes the Hearts of Trolls.
Date: 22 Feb 2001 20:57:42 GMT
In comp.os.linux.advocacy Gary Hallock <[EMAIL PROTECTED]> wrote:
: In article <96s7ui$a4u$[EMAIL PROTECTED]>, "Steve Mading"
: <[EMAIL PROTECTED]> wrote:
:> What is the simplest algorithm for getting the numerical value of a
:> digit character in ASCII:
:> x = ord(ch) - ord('0')
:> Since the ord() is really a do-nothing operation that exists purely to
:> tell the compiler "yes, I really meant to do this, It's not a mistake.",
:> the above is really nothing more than a numerical subtraction operation.
:> (and even looks that way in C: ch - '0' )
:>
: Actually, this works for EBCDIC also since the digits are contiguous -
: 0xf0 - 0xf9. Still, I wouldn't recommend this method since it does make
: assumptions on the representation of characters.
You would prefer an if/else ladder checking each and every character
one at a time? Ick.
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: How Microsoft Crushes the Hearts of Trolls.
Date: 22 Feb 2001 20:58:54 GMT
In comp.os.linux.advocacy Matthias Warkus <[EMAIL PROTECTED]> wrote:
: It was the 20 Feb 2001 23:52:24 GMT...
: ...and Steve Mading <[EMAIL PROTECTED]> wrote:
:> I don't understand your defense of EBCDIC on mulitnational
:> grounds. It's even worse than ASCII in that regard.
: I did not defend EBCDIC.
: You did not read my posting.
Context. It was in a thread comparing ASCII to EBCDIC.
------------------------------
From: "Angel Iglesias" <[EMAIL PROTECTED]>
Subject: Re: How much do you *NEED*?
Date: Thu, 22 Feb 2001 09:26:13 +0100
> So with Microsoft software you pay _twice_.
It depends on how complex are you requirements.
> Unstable drivers will freeze anything. What kind of argument is that? At
> least with linux, you can choose to leave out unstable drivers
Just answering to an argument like this with another equal.
> How do I
> recompile Windows?
What kind of question is this ? Why should I want to ? Do you think
that by compiling your kernel unstable drivers switch onto stable ones ?
> Only if you insist on tinkering. A properly configured linux system for
> business use hardly needs any updating at all. Heck, it doesn't even need
> reboots.
It DOES need reboots (I assure you), obviously much more less that
Windows;
anyway, a properly configured Windows is the argument a Winvocate would tell
you ...
> It _does_ mean something if a badly behaving application can take the
> entire computer down.
Outlook virii under NT/2000 don't use to do that if properly configured
system.
> Viri, troyans and worms are a risc to any computer; at least with linux,
> they only affect the careless/dumb/gullible user and not the rest of the
> system.
The same for a well set up Windows.
> And we can't stop laughing with all the dumb idiots who _still_ pretend
> that their Windows is stable hehehe :-)
I do not pretend Windows is stable. I pretend is not as unstable as
it is usually said in here ... and keep on laughing about a lot of things
...
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: The Windows guy.
Date: Thu, 22 Feb 2001 15:10:53 -0600
"Edward Rosten" <[EMAIL PROTECTED]> wrote in message
news:972i60$jfu$[EMAIL PROTECTED]...
> >> >> > The definition of a pipe is to divert the output of one program
> >> >> > into the input of another. Multitasking is not a part of the
> >> >> > definition.
> >> >>
> >> >> Yes. Without multitasking, it can't do that properly.
> >> >
> >> > Perhaps you can point out a single credible definition of "pipe" that
> >> > defines it with multitasking?
> >>
> >> i don't need to cite anyone since I can prove my case:
> >
> > In other words, you're making up your own definition.
>
> No, I'm proving the requirments based on functionality
No, you're making up your own definition.
> >> program_that_never_stops[*] | head -10
> >
> > Try this under Unix (or any other multitasking OS) using the same
> > program.
> >
> > program_that_never_stops | tail -n 10
> >
> > tail needs an end of file before it can know where to start. So
> > likewise, it doesn't work.
>
> No, that is an erroneous argumet. I have given as example of something
> that does work with piping but will only work under a multitasking OS.
No, you stated that in order for pipes to be "real" they have to work in all
cases. I just provided a case where it doesn't, just like you tried to
provide a case where it didn't to prove your point.
> >> Since I have just proven that piping does not always work under single
> >> tasking, I have proven that it needs to have multitasking to work
> >> properly. I do not need to cite someone elses definition because I have
> >> proven that it needs multitasking.
>
> > Piping doesn't always work under multitasking either. Your arguments is
> > moot and inaccurate.
>
> You've given an example where pipes fail because what you're trying to to
> is non computable (so of course a computer can't do it). Citing a non
> computable problem as a reason for pipes not working under multitasking
> is not a valid argument because the computation fails for other reasons.
It's a perfectly valid argument. Pipes do not require multitasking.
> I have cited a computable problem which works under multitasking
> but can not work under single tasking. I still stick by this as proof
> that pipes need multitasking to function properly.
And I cited one that doesn't work. Stop changing your defintions.
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: The Windows guy.
Date: Thu, 22 Feb 2001 15:12:03 -0600
<[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Wed, 21 Feb 2001 18:47:47 -0600, Erik Funkenbusch <[EMAIL PROTECTED]>
wrote:
> ><[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> On Wed, 21 Feb 2001 18:31:02 -0600, Erik Funkenbusch <[EMAIL PROTECTED]>
> >wrote:
> >> ><[EMAIL PROTECTED]> wrote in message
> >> >news:[EMAIL PROTECTED]...
> >> >> On Wed, 21 Feb 2001 14:05:54 -0600, Erik Funkenbusch
<[EMAIL PROTECTED]>
> >> >wrote:
> >> >> >"Edward Rosten" <[EMAIL PROTECTED]> wrote in message
> >> >> >news:970tqj$i83$[EMAIL PROTECTED]...
> >> >> >> > "proper pipes"?
> >> >> >> >
> >> >> >> > The definition of a pipe is to divert the output of one program
> >into
> >> >the
> >> >> >> > input of another. Multitasking is not a part of the
definition.
> >> >> >>
> >> >> >> Yes. Without multitasking, it can't do that properly.
> >> >> >
> >> >> >Perhaps you can point out a single credible definition of "pipe"
that
> >> >> >defines it with multitasking?
> >> >> >
> >> >> >By credible, I mean a university, book, or other experts
definition.
> >> >>
> >> >> Actually, any academic text on operating systems will
> >> >> extol the value of exploiting parallelism. This is what
> >> >> a "real pipe" can do that mere copies don't. They (and
> >> >> multitasking in general) allow for tasks to proceed as
> >> >> a collection of smaller tasks that may exercise different
> >> >> parts of the system simultaneously.
> >> >
> >> >Then it should be quite simple for you to provide a credible
reference.
> >> >Right?
> >>
> >> Any edition of the "dinosaur" book should do.
> >
> >Should it? Are you guessing? Have you actually read this in the book?
> >Simply provide a page reference.
>
> Chapter 1, 5th edition.
I said a page reference. You might as well be saying "It's on the internet
somewhere", go look it up yourself, i'm too lazy to prove my own statements"
------------------------------
From: "Matthew Gardiner" <[EMAIL PROTECTED]>
Subject: Re: Microsoft backs out of Corel
Date: Fri, 23 Feb 2001 10:01:54 +1300
I'm not really worried about Corel or Microsoft. Had Corel focused on its
core fundimentals instead of going into unknown territory, in the case of
the netwinder (I would hate to know how much was wasted in that little pipe
dream), Corel wouldn't be in the deep shit they are now. Also, the lack of
willingness to compete with Microsoft head on is another issue. In 1997 the
NZ Army had just upgraded there computers from running Wordstar 2000, DBIV,
123, and Harvard Graphics to Pentium's running Windows NT, hence they needed
a new office suite. When the NZ Army were looking for software companies
that were interested in providing an Office Suite for the army, the one who
won was Microsoft, they offered a terrific deal, the deal allowed all
computers in the army, AND all army personal who had computers at home, to
be able to load Office. Corel didn't even offer a deal (from what I have
heard from sources) even close to what Microsoft offered. Hence, the
reason why Microsoft in some respects are successful, they chase customers,
unlike Corel, who just sit around hoping someone will, out of the good ness
of their own heart, choose Corel over their competitors.
Matthew Gardiner
"Tim Hanson" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> http://linuxtoday.com/news_story.php3?ltsn=2001-02-22-009-20-NW-CL-MS
>
> By Brian Proffitt, LinuxToday
>
> "In what appears to be a move to dodge yet another Department of Justice
> investigation, Microsoft Corporation may be selling its shares in Corel
> Corporation--at half the price it bought those shares.
>
> "According to the S-3 registration statement filed by Corel Corporation
> Feb. 21, 24 million Series A shares will be converted to common shares
> and then registered for sale by Microsoft. This step is necessary
> because the Series A shares are not convertible by Microsoft directly.
>
> "What the registration does not say is how many of these converted
> shares Microsoft will actually put on the selling block. But, if
> Microsoft sells all of these shares at the proposed price of $2.5625 per
> share, then Microsoft would only make $61.5 million--a figure that
> represents half of the $135 million price Microsoft paid for the shares
> in October, 2000.
>
> "Though no reason was given for the conversion, Microsoft's known cash
> reserves, which are fairly large, tends to fuel speculation by industry
> observers that Microsoft is not using this sale to generate cash, but is
> seeking to avoid yet another Department of Justice anti-trust
> investigation.
>
> "The 24 million Series A shares represent the entirety of Microsoft's
> October 2 purchase of the non-voting shares in Corel Corp.
>
> No one at Microsoft or Corel could be reached for comment.
>
> --
> Dimensions will always be expressed in the least usable term.
> Velocity, for example, will be expressed in furlongs per fortnight.
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: Whistler/.NET will Help Linux
Date: Thu, 22 Feb 2001 15:16:01 -0600
"Steve Mading" <[EMAIL PROTECTED]> wrote in message
news:973tuf$gks$[EMAIL PROTECTED]...
> : In any event, the capability you're talking about is not an inherant
part of
> : the file system, but rather of the kernel model. A file handle is a
file
> : handle, regardless of what file system is used. Remember, Linux can run
on
> : FAT. Are you going to suggest that FAT is OO?
>
> If you are going to use the literal meaning of "filesystem" like that,
> then the phrase "OO filesystem" is inherently meaningless. The filesystem
> is just a data format. That can never be OO, regardless of OS. Bitching
> that unix has no OO filesystem then becomes a stupid inherently
> meaningless utterance, like complaining that you don't like the taste
> of blue, or the sound of purple.
>
> I was giving more benefit of the doubt than was warranted, it seems, by
> assuming that that *wasn't* what was meant by 'filesystem', and that
> the wrong vocabulary term was being used and the meaning was actually
> something a bit more high-level than the filesystem.
What MS was proposing as an OO File System was totally different. It was,
in fact, file system dependant, and was, in fact, not referring to what you
are. I'll find the information on what exactly they were referring to and
post it.
------------------------------
** 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
******************************
