Linux-Advocacy Digest #44, Volume #32 Wed, 7 Feb 01 22:13:04 EST
Contents:
Re: The Wintrolls (Charlie Ebert)
Re: Wy Linux will/is failing on the desktop (J Sloan)
Re: Would linux hackers like an OpenS windows? (Charlie Ebert)
Re: The Wintrolls (J Sloan)
Re: Microsoft is FUN and Linux is BORING (J Sloan)
Re: KDE Hell (T. Max Devlin)
Re: ERIK FUNKENBUSH CAN'T TELL US ***WHAT*** .NET IS (Charlie Ebert)
Re: NTFS Limitations ("Tom Wilson")
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) ("Tom Wilson")
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (T. Max Devlin)
Re: I don't understand (Osugi)
Re: Wy Linux will/is failing on the desktop ("Weevil")
Re: NTFS Limitations (T. Max Devlin)
Re: Aaron R Kulkis (Bob Hauck)
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (Bob Hauck)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Subject: Re: The Wintrolls
Reply-To: Charlie Ebert:<[EMAIL PROTECTED]>
Date: Thu, 08 Feb 2001 01:46:19 GMT
In article <jrhg6.6909$[EMAIL PROTECTED]>, Erik Funkenbusch wrote:
>It appears that most of Charlie's more exuberant statements are not archived
>on deja, probably due to him putting an X-Archive attribute in his message,
>since he knew they would be used against him.
>
Excuse me? I did what?
>http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=715528378&fmt=text
>
>"From practical EXPERIENCE I can safely say that Linux doesn't crash."
>
>That's not even all, he claims that *NO OTHER OS* has higher uptimes than
>Linux.
>
>http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=705087775&fmt=text
>
>"That and the fact it's totally free for download from the internet
>and it has the most tremendous uptime of any OS on the market."
>
Well, you don't need to search backwards in deja for this.
I will say it again if need be. Linux has the highest uptime of
any OS on the market.
>
>Note that you don't correct my supposed lack of experience. I know exactly
>what i'm talking about.
>
>> And besides that, even if it WERE true that the linux kernel was optimized
>> for 386 chips (which is actually quite a meaningless statement if you know
>> anything about the kernel or kernel architecture in general)
>
>It's not a meaningless statement. There are numerous ways to optimize a
>kernel for a particular processor.
>
It is meaningless because the Pentium III still shares the same
basic command set the 386 did. They just added more commands.
And when you recompile the kernel for these different chips, these
new commands are issued from the kernel.
>1) Use compiler optimizations designed for that processor. These will
>continue to work in most later processors, but you won't get many of the
>speed improvements the processor is capable of.
>
As I just said, if you flip the Pentium III switch on the Linux kernel
and re-compile it, you will be using the Pentium III command set then.
>2) Not using processor specific instructions to take advantage of speed
>increases in later processors.
>
Linux does.
>3) Optimizing for a specific set internal cache type.
>
It does.
>For instance, with the FreeBSD kernel, there are internal options for each
>processor that's supported. By removing options for the 386 and others, you
>increase the efficiency of the kernel.
>
>> because theyre
>> the lowest common demonimator; its a hell of alot better than optimizing
>> the entire operating system for the lowest common demoninator of
>intellect.
>
>Hand waving. You'r not saying anything here.
>
FreeBSD does the same thing only globally with MAKEWORLD.
I'm surprised you didn't nit pick Linux for not being able to
do a MAKE WORLD which it can't do easily.
Now, that would be a comment with substance.
--
Charlie
**DEBIAN** **GNU**
/ / __ __ __ __ __ __ __
/ /__ / / / \/ / / /_/ / \ \/ /
/_____/ /_/ /_/\__/ /_____/ /_/\_\
http://www.debian.org
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Crossposted-To: alt.linux.sux
Subject: Re: Wy Linux will/is failing on the desktop
Date: Thu, 08 Feb 2001 01:49:53 GMT
Erik Funkenbusch wrote:
> I didn't say that. I said 66% couldn't even get it to work, that's not the
> same thing.
Which is an outrageous claim on the face of it.
I used to the "Linux install guy" at a certain University campus
a few years back. I personally installed Linux for probably 100
students during my time there, and gave advice to may more.
I only knew of one system where Linux apparently just
would not install.
That's reality, and it's a far cry from your statistics.
jjs
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Subject: Re: Would linux hackers like an OpenS windows?
Reply-To: Charlie Ebert:<[EMAIL PROTECTED]>
Date: Thu, 08 Feb 2001 01:50:27 GMT
In article <95rdoc$6l9$[EMAIL PROTECTED]>, sfcybear wrote:
>In article <j_Qf6.3647$[EMAIL PROTECTED]>,
> "Erik Funkenbusch" <[EMAIL PROTECTED]> wrote:
>> "Aaron R. Kulkis" <[EMAIL PROTECTED]> wrote in message
>> news:[EMAIL PROTECTED]...
>> > gswork wrote:
>> > >
>> > > Out of interest, having spent time time hacking Linux would coders
>love
>> > > to see the behemoth code that lies underneath Windows?
>> > >
>> > > It would be fascinating would it not? Some of it is probably
>pretty
>> > ^^^^^^^^^^^
>> >
>> > You misspelled "laughable"
>> >
>> > [Ever see Microsoft source code??? Most of it, even college
>sophomores
>> > would be ashamed to sign their name to. No wonder Gates doesn't
>want
>> > anybody to see it.]
>>
>> And how exactly would you know? You've never seen it.
>
>MS software has such poor performance and such lousy uptime capability,
>it can only be because the code is poorly writen.
>
>
>>
>>
>
>
>Sent via Deja.com
>http://www.deja.com/
And poorly designed/thought out....
Folks, you don't need to see the code to determine this.
You just have to say three little words!
I LOVE YOU
--
Charlie
**DEBIAN** **GNU**
/ / __ __ __ __ __ __ __
/ /__ / / / \/ / / /_/ / \ \/ /
/_____/ /_/ /_/\__/ /_____/ /_/\_\
http://www.debian.org
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Subject: Re: The Wintrolls
Date: Thu, 08 Feb 2001 01:58:29 GMT
Erik Funkenbusch wrote:
> I don't have to put words in anyones mouths. Charlie Ebert and others state
> quite matter of factly that Linux NEVER crashes, and have said so numerous
> times.
So, perhaps it never crashed for him - did you consider that?
> Linux has to install to the lowest common denominator CPU, the 386.
On the S/390 mainframe? or an Alpha? How about a PowerPC?
> That
> means the kernel is optimized for that. Some distro's will perhaps install
> a 586 or 686 optimized kernel later in the install process, but it will
> still be a generic one.
Most x86 distros ship with a boot kernel that will run on
a 386 with no mathco, but also include kernels that are
optimized for 486, 586, or 686.
> Mandrade 7.2 is still RPM 3 based.
So, why not stick with RPM 3?
> With far fewer packages available for them.
All rpm packages contain tarballs, so that's nonsense.
> I posted about it quite often a few months ago.
Sorry, I wasn't around then.
> Really? Then explain this:
> http://www.coastnet.com/~pramsey/linux/homenet.html
Looks like 2 isa cards. IIRC I used 2 addresses for my
2 isa cards. But when I bought a new box and used realtek
8139 cards, they shared the interrupt. I just swapped them
for eepro100s, and guess what? They still share interrupts:
#cat /proc/interrupts
CPU0
0: 7719284 XT-PIC timer
1: 92224 XT-PIC keyboard
2: 0 XT-PIC cascade
5: 1973805 XT-PIC EMU10K1
7: 33 XT-PIC usb-ohci
9: 0 XT-PIC acpi
10: 277003 XT-PIC eth0, eth1
12: 1588145 XT-PIC PS/2 Mouse
14: 89872 XT-PIC ide0
15: 3278 XT-PIC ide1
NMI: 0
ERR: 0
As you can see, it all works, no muss, no fuss.
> "Most folks" don't even know hdparm exists. It took me 3 months to figure
> out that DMA wasn't on, and then many hours to figure out the settings for
> it.
But you're not supposed to be "most folks", so don't
use that excuse here.
jjs
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Subject: Re: Microsoft is FUN and Linux is BORING
Date: Thu, 08 Feb 2001 02:02:21 GMT
Erik Funkenbusch wrote:
> "J Sloan" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > However, microsoft doesn't dare use a single box like
> > ftp.cdrom.com or ftp.freesoftware.com, each of which
> > use a single BSD box.
>
> Both of which are just FTP sites and don't process dynamic.
On the contrary, in addition to being the busiest ftp sites
on the planet, they are also serving web pages, shell
sessions, and all the normal Unix services.
> ftp.microsoft.com runs on one server as well.
I'm sceptical of that - if it's a single address it's
no doubt a "local director" or some such. It goes
aganst microsoft's own "best practices" to have
a single box, so there's no way. In any event, it
doesn't come close to the traffic the single BSD
cpu handles.
jjs
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.linux.sux
Subject: Re: KDE Hell
Reply-To: [EMAIL PROTECTED]
Date: Thu, 08 Feb 2001 02:05:05 GMT
Said Steve Mading in comp.os.linux.advocacy on 7 Feb 2001 07:36:56 GMT;
[...]
>Has it occurred to any of the people in this argument that maybe,
>just maybe, it's possible that *both* the Davidians and the ATF
>were guilty, and that neither side was in "the right"? Perhaps
>the madman *did* torch his own followers out of some demented
>sense of religious martyrdom, but the ATF shouldn't have put him
>under the siege in the first place. There's nothing illegal about
>being a moron who thinks he's some kind of god.
No, but it is illegal to resist arrest, as counter-intuitive and scary
as that may seem to the libertarianites. Perhaps the ATF *could* have
avoided a siege. Other than the fact that the religious madman decided
to murder his own followers and their children to avoid arrest on a
relatively petty charge, though, there's scarcely any reason to presume
they "shouldn't" have. And the fact that Koresh turned out to be a
murderous madman, not just a religious flake who happened to own guns,
doesn't really support that point, either.
To be honest, I find the Ruby Ridge incident many times more disturbing.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: ERIK FUNKENBUSH CAN'T TELL US ***WHAT*** .NET IS
Reply-To: Charlie Ebert:<[EMAIL PROTECTED]>
Date: Thu, 08 Feb 2001 02:08:23 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
>On Wed, 07 Feb 2001 21:32:09 GMT, Johan Kullstam
><[EMAIL PROTECTED]> wrote:
>
>>"Aaron R. Kulkis" <[EMAIL PROTECTED]> writes:
>>
>>> I'm merely asking you to tell us to define what .NET is.
>>
>>ok i'll bite, please define what .NET is.
>>
>>--
>>J o h a n K u l l s t a m
>>[[EMAIL PROTECTED]]
>>Don't Fear the Penguin!
>
>It's a VM and a class library.
>
>Kind of like Java/Swing except that: 1) .NET (supposedly) supports
>more than one language on top of the VM, and 2) .NET allows
>"unmanaged" access to memory, whereas the JVM does not.
>
>See, it's not that hard!
>
Don't forget unmanaged access to the system registry!
They have to have a way to weasle out your installed base of software!
--
Charlie
**DEBIAN** **GNU**
/ / __ __ __ __ __ __ __
/ /__ / / / \/ / / /_/ / \ \/ /
/_____/ /_/ /_/\__/ /_____/ /_/\_\
http://www.debian.org
------------------------------
From: "Tom Wilson" <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations
Date: Thu, 08 Feb 2001 02:09:19 GMT
<[EMAIL PROTECTED]> wrote in message
news:95s91o$2n6$[EMAIL PROTECTED]...
> In comp.os.linux.advocacy Charlie Ebert <[EMAIL PROTECTED]> wrote:
>
> >>You should really read up on it. All that stuff that Sun promised about
> >>Java, but never came through on? That's .NET, plus more.
> >>i
>
> Indeed. Its everything that sun promised, minus the cross platform
> capabilities.
>
> Its a stupid idea, bad implementation, and utterly useless.
>
> Just like everything else microsoft has done for the past 10 years.
Component Object Model - FUD Edition.
I'm awash with giddy anticipation...
------------------------------
From: "Tom Wilson" <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations (Was: RE: Red hat becoming illegal?)
Date: Thu, 08 Feb 2001 02:16:29 GMT
"Giuliano Colla" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Chad Myers wrote:
> >
> > "Giuliano Colla" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > Erik Funkenbusch wrote:
> > > >
> > > > "Tom Wilson" <[EMAIL PROTECTED]> wrote in message
> > > > news:4yRf6.575$[EMAIL PROTECTED]...
> > > > > > (HINT: The subscription based services are only a tiny part of
it, and
> > > > > > something that very few .NET programs will take advantage of.
If this
> > > > is
> > > > > > the only argument you can come up with, you're going to be quite
> > > > > surprised).
> > > > >
> > > > > The very fact that feature is being proposed is enough to conjure
up past
> > > > > memories of subscription based software from the early eighties.
It is a
> > > > > blatant rip-off and causes your TCO to skyrocket. Actually, i'm
surprised
> > > > > its' taken this long, with the Internet being what it is now, for
someone
> > > > > to seriously pursue such a course again. The consumer sector said
no,
> > > > > resoundingly, to DIVX and i'm hoping that the commercial sector
takes the
> > > > > same tact with this profit mongering.
> > > >
> > > > The subscription based thing is there, much like Windows Terminal
Services
> > > > is there. You can use it, it works, but it's really only for a very
small
> > > > subset of the population.
> > > >
> > > > The subscription services are for companies that regularly upgrade.
The TCO
> > > > is reduced because everyone is automatically updated at the same
time,
> > > > without the need (or very little need) of maintenance and license
> > > > management.
> > > >
> > > > This is really only for large companies. Small companies will
continue to
> > > > buy discrete versions of the products (non-subscription). MS simply
can't
> > > > get away with removing the non-subscription option. People simply
won't
> > > > upgrade if they're not interested in doing so.
> > > >
> > > > > I've heard some of the jucier technical details of .NET and, as a
> > > > > developer, I see the potential. I also see the scenario I just
ranted on.
> > > > > We've made the decision not to develop for it and we won't. If it
takes
> > > > > off, and I don't see it doing so... One of the alternative OS's
will just
> > > > > have to be modified to counter it. Be it Linux or BSD.
> > > >
> > > > Well, since C# and the CLR are now ECMA standards, this is a
possibilty.
> > > > The real benefit of .NET will be the Java-like cross platform
capability
> > > > (think CE, 32 bit windows, 64 bit windows, MacOS X all from the same
EXE,
> > > > each optimized for their own platforms by the .NET runtime compiler
(which
> > > > is much more like SmallTalk than Java))
> > > >
> > > > And, if people write .NET for Linux, as they'll be able to do from
the
> > > > standards, you can run on Linux as well.
> > >
> > > What for?
> >
> > If you have to ask, then you really have no clue about the .NET.
> >
> > You should really read up on it. All that stuff that Sun promised about
> > Java, but never came through on? That's .NET, plus more.
> >
>
> Well, MS jumped upon Java, made is as crappy as they only can do, and
> what's the result?
> Any reason why .NET should be better? From a company which is forced to
> outsource to a Linux company its DNS's because they're not able to set
> them up them properly? Please, try to be serious!
Its the kind of behavior you'd expect from a company that slaps new names on
existing technologies, cripples them either out of commercial considerations
or shear ineptitude, makes grandiose statements about their potential
impact, releases them, then spends the rest of the product's lifespan either
denying the faults or fixing those that can't be ignored. If a product is
too badly broken, it is simply renamed and the cycle starts anew.
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations (Was: RE: Red hat becoming illegal?)
Reply-To: [EMAIL PROTECTED]
Date: Thu, 08 Feb 2001 02:17:40 GMT
Said Charlie Ebert in alt.destroy.microsoft on Thu, 08 Feb 2001 00:36:22
GMT;
>In article <xu8g6.6437$[EMAIL PROTECTED]>, Erik Funkenbusch wrote:
>>"Charlie Ebert" <[EMAIL PROTECTED]> wrote in message
>>news:[EMAIL PROTECTED]...
>>> YEAH! Everything that bad old SUN company promised to do but
>>> never came thru on.
>>
>>Yes, such as submitting Java to ISO, and then ECMA for standardization.
>>
>>> Perhaps the $20 million Microsoft just agree'd to pay them for
>>> *STEALING JAVA* in the first place might have had something
>>> to do with that.
>>
>>You haven't a clue. The $20 million was to buy out their existing contract
>>with Sun. Sure, they won the battle, but they lost the war. Microsoft
>>can't use the Java compatible logo, but Java doesn't mean anything anymore.
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
>
>Wholy shit. This guy really is an absolute loonatic.
Did you think we were kidding? ;-D
You heard him right, Charlie: since Microsoft lost, Java "doesn't mean
anything anymore". This kind of reverse logic is provided by the
traditional sock puppet technique of insisting that any technology which
Microsoft hasn't been able to dominate must be presented as "old" and on
the way out while monopoly crapware leads the way to the future.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: Osugi <[EMAIL PROTECTED]>
Subject: Re: I don't understand
Date: Thu, 08 Feb 2001 02:17:56 GMT
In article <[EMAIL PROTECTED]>,
Matthew Gardiner <[EMAIL PROTECTED]> wrote:
> Hypothetically, it is meant to show the superior multi-threadedness of
> each OS. Mp3 decode/encode require a bit of cpu grunt, hence, on a
> properly threaded OS, this should be a piece of piddle. Personally, I
> don't use that as a bench mark, because it is so unreliable and can be
> influenced by outside factors, such as processor speed and memory. I
> prefer the real world experience. For example, in my case..I run Applix
> Office, Mozilla 0.7 and Mp3 at the same time without and notice-able
> performance loss (mind you I have a PeeCee w/ Freebsd 4.1, 256MB RAM,
> 550 PIII Coppermine).
>
> Matthew Gardiner
Since we are all swapping stories,
I once burned a cd while listening to mp3s and running setiathome.
Fortunately for the cd, I was burning as root. But even so, the mp3s
sounded fine and the cd came out ok. I doubt that that would have been
the case under Windows98 or WinME.
(at the time, it was a k6-2 450 with 128 meg running kde)
To answer the original question, I personally have no need or desire to
play more than mp3 at a time.
--
Osugi Sakae
I will not be filed, numbered, briefed or debriefed.
I am not a number, I am a free man. -The Prisoner
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Weevil" <[EMAIL PROTECTED]>
Subject: Re: Wy Linux will/is failing on the desktop
Date: Thu, 08 Feb 2001 02:33:39 GMT
ono <[EMAIL PROTECTED]> wrote in message
news:95scnj$3gs$[EMAIL PROTECTED]...
>
> > > So why run Linux?
> >
> > Why are you running Windows? I'm running it because I've yet to see
> > Netscape or KNode take out my desktop. Sure I've found other problems,
but
> > in general usage, none so far.
> I've yet too see IE take out my desktop!
>
Everyone reading this who has ever used IE knows you're lying. IE takes out
desktops on every single computer it runs on. Pretty much everybody reading
this knows that. Who do you think you're talking to, anyway? Windows fans?
jwb
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations
Reply-To: [EMAIL PROTECTED]
Date: Thu, 08 Feb 2001 02:43:16 GMT
Said Steve Mading in alt.destroy.microsoft on 7 Feb 2001 22:07:19 GMT;
>In comp.os.linux.advocacy T. Max Devlin <[EMAIL PROTECTED]> wrote:
[...]
>:>I was not considering the 'base' hardware like the CPU and motherboard.
>:>I was thinking of cards plugged into the expansion bus, produced by
>:>companies much smaller than Intel. Things like sound cards, video
>:>cards, and so on. Granted, I didn't make this clear.
>
>: No, it was clear, and I was well aware that you were considering
>: peripherals. Does that mean you don't see the relevance of my example,
>: or was I not clear enough myself?
>
>The example isn't relevant because Intel is not very representative
>of the problem. The problem comes into play only when you consider
>the large number of very small companies each making their own
>seperate, incompable peripherals.
How on earth is a small company going to survive making incompatible
peripherals? No, the example was quite relevant, though you may be
right that it is not indicative of the current market in peripherals.
What makes it relevant is the same argument you make could have been
made about motherboards two years ago. This proves, if I might be so
bold, that the argument is not just fundamentally incorrect, but
contrary to reality. There is nothing about open source low-level code
which makes it inherently lag behind proprietary code. I'll agree that
it is easier for a vendor to put together something proprietary. But
you seem to be convinced, without sufficient cause, that this is an
inherent impulse of hardware vendors.
The fact is, hardware vendors sell more hardware when they have open
source software to ensure that their market opportunity is as large as
possible, whether it be OSes for CPUs or drivers for peripherals. No
competitive hardware developer has any reason whatsoever to avoid open
source. Anti-competitive hardware developers, of course, are a
different story, but one can be sure that their hardware will both cost
more and be worth less than the alternative. There's no other reason,
after all, to be anti-competitive, particularly considering
monopolization is illegal.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: Aaron R Kulkis
Reply-To: bobh -> haucks org
Date: Thu, 08 Feb 2001 02:43:41 GMT
On Wed, 07 Feb 2001 21:49:34 +1100, Shane Phelps <[EMAIL PROTECTED]> wrote:
>
>"Aaron R. Kulkis" wrote:
>>
>> Bob Hauck wrote:
>> >
>> > On 6 Feb 2001 11:37:14 GMT, Bloody Viking <[EMAIL PROTECTED]> wrote:
>> >
>> > >Watch for longhairs who use heavy-duty calculators as a Palm Pilot...
>> >
>> > What about the ones who use a Palm Pilot as a heavy-duty calculator?
>>
>> Summary execution.
>>
>> "Thou shalt have no other calculator before HP", and all that.
>Well, *I* thought my TI58 was pretty hot stuff
I favor HP-style calcs. I therefore use NeoCal on my Palm. It is RPN
and has a good programmer mode. Not programmable though, so I have
Mathpad too. And Palm Telnet, which I have actually used to configure a
router. Do that with a TI-58!
>If God had meant us to use RPN he would have given us Forth instead of C
Didn't he say "go Forth and multiply"?
--
-| Bob Hauck
-| To Whom You Are Speaking
-| http://www.haucks.org/
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations (Was: RE: Red hat becoming illegal?)
Reply-To: bobh -> haucks org
Date: Thu, 08 Feb 2001 02:43:45 GMT
On Wed, 7 Feb 2001 03:40:34 -0600, Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
>"Bob Hauck" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> On Tue, 6 Feb 2001 06:15:11 -0600, Erik Funkenbusch <[EMAIL PROTECTED]>
>wrote:
>>
>> >Well, since C# and the CLR are now ECMA standards, this is a possibilty.
>> >The real benefit of .NET will be the Java-like cross platform capability
>> >(think CE, 32 bit windows, 64 bit windows, MacOS X all from the same EXE,
>> >each optimized for their own platforms by the .NET runtime compiler
>> >(which is much more like SmallTalk than Java))
>>
>> Ok, so why won't this suffer from the same "write once test many times"
>> problems that Java (and every other cross-platform language) has?
> I didn't say it wouldn't.
No, not exactly, but you did talk about a promised land where the same
binary would run on "Windows and Mac OS-X", with no mention of having to
test on each platform, which you constantly bring up WRT Java. Slip
your mind did it?
> The failure of WORA was not Java's failure, but rather Sun's refusal
> to relinquish control enough to allow standardization and to implement
> what needed to be implemented and their constant promising of things
> they couldn't deliver in a timely fasion.
None of which addresses the question of whether letting MS put in
proprietary extensions would help or hurt the goal of WORA. Are you
trying to say that Sun should have allowed every licensee to add in
their own special stuff? How is that reducing the testing problem?
> It took sun years to get a decent JIT (HotSpot) and years to develop a
> standard native interface, and years to develop component models like Java
> Beans...
Things like JNI are what _causes_ write once test everywhere! Are you
trying to say that having *more direct* access to platform API's would
reduce the problem? Can't you see that if you follow this line of
thinking to its logical conclusion, you must conclude that MS really
intends .NET to be an MS-only thing?
> Also, Java suffered from the fact that Sun only supported the Java
> language. Yes, there are other languages that target the JVM today,
> but they are not supported by Sun and discouraged.
I don't mind Java as a language. It seems reasonably well-designed to
me (not that I'm a language designer). But even if it was lousy, this
does not explain how MS is going to handle the objections I outlined.
I did not ask why Java has failed (I don't think it has yet). I asked
how MS was going to address *your* objections to Java, which *also*
apply to .NET equally well.
>> >And, if people write .NET for Linux, as they'll be able to do from the
>> >standards, you can run on Linux as well.
>>
>> Except for the things that don't. Since MS, the promulgator of the
>> standard, and you can be real sure that ECMA won't be calling the shots,
>> will not itself be providing .NET for Linux (or probably anything except
>> Windows), why won't we see the same problems we saw with Java on Linux
>> prior to Sun and IBM seeing the light?
>
>MS has already contracted with Corel to port .NET to Linux.
Which is _exactly_ how they got Frontpage and DCOM on Unix. And look at
what great products those turned out to be. Howcome a company with the
resources of MS can't do things in-house anyway? Why contract out such
a strategic project? Maybe because they want the port to be "less" than
what Windows has, they want it to be at least a little incompatible,
while still getting to claim in their feature lists that it is "cross
platform"? That was my point, which you did not address at all.
>> Your rosy view of .NET assumes that they will be able to overcome all of
>> the same problems Java faces, but because they are Microsoft it'll all
>> be a cakewalk.
> Java's problem was Sun. They killed it with their management.
Like I said, it'll aparently all be better because it is MS. That is
not good enough. What specifically is MS doing differently? How are
they countering the objections I outlined, beyond saying "we're
different" I mean. The objections, once more, are:
1. Binary portability precludes use of platform specific features. Yet
MS is claiming that a deficiency of Java is the lack of platform
specific features. How is .NET going to support platform specific
features and yet be platform neutral?
2. By controlling the specification of .NET, MS puts itself in a
position to squeeze out competing platforms that it doesn't like.
Again, this is exactly what MS claims about Java, that by controlling
the standard Sun is using it as a weapon against MS. That MS has got
ECMA to rubber-stamp their spec means little, and I think you know it.
What's really different here?
3. Given #1 and #2 above, why should anybody who isn't already an MS
slave support .NET? All the FUD that MS has put out about Java applies
equally to .NET even if you leave out all the licensing nonsense.
--
-| 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
******************************
