Linux-Advocacy Digest #733, Volume #31 Thu, 25 Jan 01 19:13:04 EST
Contents:
Re: Comparison: Installing W2K and Linux 2.4 (Daniel Tryba)
Re: New Microsoft Ad :-) ("ono")
Re: New Microsoft Ad :-) (T. Max Devlin)
Vendor Neutral Linux Certifications, or, LPI vs. SAIR (Bruce)
Re: Comparison: Installing W2K and Linux 2.4 (Steve Mading)
Re: Comparison: Installing W2K and Linux 2.4 (Steve Mading)
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (Johan Kullstam)
Re: Comparison: Installing W2K and Linux 2.4 (Shane Phelps)
----------------------------------------------------------------------------
From: Daniel Tryba <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Comparison: Installing W2K and Linux 2.4
Date: 25 Jan 2001 23:47:22 GMT
In comp.os.linux.advocacy Conrad Rutherford <[EMAIL PROTECTED]> wrote:
>> >> > familiar GUI up and running with all your hardware ready to rock.
>> >> ^^^^^^^^^^^^ Only if you are used to Windows.
>> > Who hasn't?
>>
>> MacOS users for example.
> Yea, but it looks a lot like Windows and a whole lot not like any linux
> gui. so your point is moot.
fvwm2-95 looks like windows. Gnome and KDE try hard to look like
windows, they make it easy enough to launch applications by double
cliking on the desktop or selecting them from the "start" menu.
>> > Besides, your Matrox will come up in svga mode - that's frame buffer
> mode -
>> > so guess what, according to your definition of supported, W2K handled
> the
>> > G400 too.
>>
>> Yeah, 640x480 or 800x600 with 256 colors, Whooopie. The builtin
>> framebuffer drivers support anything that your monitor can handle in
>> high or truecolor. They even have some acceleration builtin, which can't
>> say for the SVGA drivers.
> Woopee! So, do you intend to run your card in this framebuffer mode for
> ever? No, I'm certain you'll go and get the latest builds and compile them
> of course.
I rarely start X, only to use Netscape or watch some DivX (I use my
Linux PC as a vcr). Most of the time I'm just working with a 160x64
characters "textmode". Theres abolutley no need for running X most of
the time.
> But, it's the same thing. I'd say i've never heard of ANY card
> you couldnt' finish setup with.
How about Hercules, EGA and CGA? Yeah, it's old stuff, not manufactured
anymore unsupported, but if you are running a server you don't need
anything more fancy.
> Then you'd do the normal thing, go find the
> drivers for the items that didn't get fully installed. Tell me it's not the
> same with Linux - of course it is.
Ofcourse you need drivers with Linux for your hardware, the little
difference is that with Linux I get all my drivers with the kernel
source, no need to search for them at the manufacturers site or any
where else. But maybe that's my problem since I have old an unsupported
hardware.....
> The difference is, EVERYTHING has windows
> drivers, not everything has linux drivers (and those that do are no where
> near as mature and often not even supported).
Older isn't quite that well supported anymore apperantly. And I find
that's a shame, but we can't blame Windows since it's the manufacturers
jobs to write the drivers...... But if we can't blame windows for lack
of support by hardware producers, how is Linux to blame then?
>> Didn't look at the CD, downloaded them from their website, and there you
>> will find both Windows and Linux drivers. So what case are you resting?
>> (http://www.matrox.com/mga/support/drivers/latest/home.cfm)
> Well, you had to go to the site after installation to get them. Chances are
> the drivers are on the CD already so you could probably install them during
> the original setup. But, see my previous reply, of course you'll go to the
> site to get drivers - IF they are available. Without doubt, there are more
> windows drivers than for any other OS. So it's YOUR comment I've rebuffed.
See above, _all_ drivers for my hardware are in the kernel. Only thing
not supported without the driver from the matrox site is the TV-out in
XFree86, but with 2 monitors hooked up I can't use the TV....
>> Who cares what it costs, it used to work with pre Windows 2k MS
>> operating systems, why is there no support for it in 2k??? The
>> replacement 21040 is even older and that one still has support!
>> BTW the 21040 cost me about 7$ 3 years ago, so the cost of the product
>> doesn't say anything about how it works, the 3com 595 was more expensive
>> (it's an 100Mbps NIC).
> You know, I have no idea. Ask 3COM. Unless you somehow feel it's Microsoft's
> responsibility to write drivers for other vendor's old out dated and not
> supported or even manufactured hardware?? weird...
How come that Linux still supports it, it uses the same driver for the
905 series, so there can't be that big of a difference.
>> Yep, that's what the framebufferdriver does. But windows doesn't support
>> it in the base distrobution. So why do you want to compare Windows with
>> add-on drivers to Linux without downloading extra stuff? Sounds unfair
>> to me.
> No, see my previous replies - you confirmed what I meant in your comments
> too.
Nope, I have a fully functional system _without_ any aditional
drivers/software.
>> I do, I don't throw anything away if it's still usefull. Just like the
>> NIC which workes perfect with win9x, NT4 and Linux.
> you know, eventually I sell my car with 100,000 miles on it even if it is
> running. I care about performance. AND as I wrote above, why blame MS if
> 3COM doesn't write drivers for a card it no longer makes or supports??
Why blame Linux then for lack of driver support if the mnufacturer
doesn't write them or refuses to give some specs so other will.
> Yes, the asus had a hardware malfunction - how that is a MS problem I cannot
> fathom, i felt it totally unreleated to this thread. I too had threee
> P2B-ds's and had to send them all in for the fix by asus for free. Yea, that
> can happen. Can happen to any OS.
Sure flaky hardware can kill any OS. _But_ take a look at:
http://http://www.asus.com.tw/Products/Techref/Acpi/solution.html
and you will see a couple of dates in the document:
<quote>
Please check the table below to see if your board requires rework and
the detail rework information. All ASUS multi-processor motherboards
will be reworked before shipping out as
from Oct. 01, 1999.<br><br>
</quote>
and an other one in the html source at the bottom:
<quote>
<font size="-1"><a HREF="/Company/copyright.html">Copyright</a> (C)
ASUSTeK COMPUTER INC. All Rights Reserved.</font>
</center></address>
<!-- <font size="-2"><I>Last-Madified: 07/20/98</I></font> -->
</quote>
So the problem existed well before the launch of Windows 2k in january
2000. But still the board is listed in the textfile on the first
bootfloppy as an ACPI safe board and that's what suprises me, apparently
they didn't do any tests, and didn't get any feedbacks from the beta
testers. And with windows you can't uninstall acpi after installation
(at least I can't find it at the MS website), so
maybe they should have altered the folowing line in TXTSETUP.SIF:
ASUSP2B-DS02/03/99=ASUSP2B-DS02/03/99
to something bit more conservative. Or let the user make the choice
wheter to install or not them selves (with a default to yes since most
systems are not affected).
> Switching from ACPI to non-acpi is a
> backwards step, when is the last time you downgraded your system when
> changing it - think in those terms and you'll see why it's not a high
> priority. Who'd want to run a non-acpi system these days? Oh wait, you still
> hang on to $3 throw-away bin NICs.
What do you need acpi for? I have no idea (and I obviously can't test
it). If I need the machine it's turned on, if I don't need it, it's off.
I can use the bios and schedulers like cron or at to make sure that the
machine is on and doing stuff if I'm not around. So what good is acpi?
BTW I turned powermanagement off in the bios, but in all their wisdom
Microsoft decided I needed it anyway.
--
Daniel Tryba
------------------------------
From: "ono" <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.nt.advocacy
Subject: Re: New Microsoft Ad :-)
Date: Fri, 26 Jan 2001 00:43:20 +0100
> The tests are sending random crap through OS communication channels in
> order to see whether the OS will screw up or not. When an application
> runs amok, who knows what it's going to do? It would be nice to know
> that no matter WHAT an app did, the OS would keep ticking. That's what
> these tests show.
Did you actually read the article? I don't think you did because they just
state about applications that fail. In case you lost the link, here it is;
http://www.cs.wisc.edu/~bart/fuzz/fuzz-nt.html.
Btw: give me one method (or 'crap' i can feed the os with) that will crash
W2K. I've yet to find one.
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.nt.advocacy
Subject: Re: New Microsoft Ad :-)
Reply-To: [EMAIL PROTECTED]
Date: Thu, 25 Jan 2001 23:54:32 GMT
Said . in alt.destroy.microsoft on Fri, 26 Jan 2001 12:45:06 +1300;
>> C'mon, start thinking. These test are like when you put water into the tank
>> of a car and measure how long it takes for the engine to die.
>> Why sould I protect my application in a release build from random data when
>> the data is always generated on the same machine from the same programs?
>> You unix/linux people must be really desperate to prove fault in ms software
>> to take such crap at face-value.
>
>The tests are sending random crap through OS communication channels in
>order to see whether the OS will screw up or not. When an application
>runs amok, who knows what it's going to do? It would be nice to know
>that no matter WHAT an app did, the OS would keep ticking. That's what
>these tests show.
Lets be honest: it is *required* that no matter WHAT an app does, the OS
keeps ticking. That is the whole point. And even Macintosh, with
*cooperative multitasking*[1], does that better than Windows!
[1] Cooperative multitasking! Can you believe it! Boy were they
morons![2]
[2] This message brought to you as a public service by the engineers of
comp.os.linux.advocacy.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: [EMAIL PROTECTED] (Bruce)
Subject: Vendor Neutral Linux Certifications, or, LPI vs. SAIR
Date: Thu, 25 Jan 2001 23:59:14 GMT
I'm interested in learning Linux from a professional perspective. I
have taken an intro course and have been using Linux for over a year
now. I've installed many distributions and have learned a lot. But,
there are gaps. I want to get certified with either LPI's offering or
SAIR's offering to both (a) give myself a structured study goal and
(b) obtain a well rounded and respected vendor-neutral Linux
certification. I'm going to do this self-study, using various
distributions and several PC's including a Sparc workstation.
My question is, which is better or has the better future? SAIR seems
more academic and thorough, but focuses on multiple distributions
(which you can argue is both good and bad). LPI seems to be more well
known at this time, and is supposedly distro neutral, but it doesn't
seem to cover as much ground and detail as SAIR's cert. LPI seems to
get more press, but there seems to be this tone that SAIR isn't going
away and will compete quite heavily with LPI.
Does anyone have an experience with which they think is better or is
about to "Take off" ?
Bruce
Vendor Neutral Linux Certifications, or, LPI vs. SAIR
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Comparison: Installing W2K and Linux 2.4
Date: 25 Jan 2001 23:56:15 GMT
In comp.os.linux.advocacy Conrad Rutherford <[EMAIL PROTECTED]> wrote:
: Oh really? Then perhaps you'll tell me why this article was taken from a
: PRO-LINUX website that was linked to by slashdot... (which is down right
: now, again!)
: http://www.thedukeofurl.org/reviews/misc/kernel2224/5.shtml
The article is fine. The error is in your pretending that a
kernel compile and install is anything like a default distribtution
install, as if they could be compared. Compare distro install
to distro install if you want to be honest (which of course, you
don't).
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Comparison: Installing W2K and Linux 2.4
Date: 25 Jan 2001 23:52:17 GMT
In comp.os.linux.advocacy Conrad Rutherford <[EMAIL PROTECTED]> wrote:
: Gee, my scenario is how to install W2K versus how to install the 2.4
: kernel.
Apples and oranges. And you know it. W2K is a kernal and associated
support files, and (*importantly*) an install program. Compare it to
installing a Linux *distribution*, not a kernel recompile. (Which
has no parellel in Windows becasue you *can't* unless you work at
Microsoft.)
I've installed Slackware, Redhat, and Suse. Slackware and RH were
FASTER than installing Windows NT or Windows 98 was (I haven't tried
W2K yet, and won't until it becomes absolutely necessary). SuSE
was slower becasue they follow a similar install path to the way
Microsoft installs do things - reboot, reboot, reboot.
------------------------------
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?)
From: Johan Kullstam <[EMAIL PROTECTED]>
Date: Fri, 26 Jan 2001 00:02:39 GMT
"Chad Myers" <[EMAIL PROTECTED]> writes:
> "Johan Kullstam" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > "Ayende Rahien" <Please@don't.spam> writes:
> >
> > > "T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> > > news:[EMAIL PROTECTED]...
> > > > Said Ayende Rahien in alt.destroy.microsoft on Fri, 19 Jan 2001 06:58:01
> > > > >"." <[EMAIL PROTECTED]> wrote in message
> > > > >news:[EMAIL PROTECTED]...
> > > > >> > > Linux is not at all at fault in this scenario. You have issues
> > > with the
> > > > >> > > limitations of one filesystem. Exactly like the limitations of FAT
> > > or
> > > > >> > > NTFS (I know NTFS can handle larger files than ext2, but that
> > > doesn't
> > > > >> > > mean it doesn't have its limits).
> > > > >> >
> > > > >> > The only real limitation of NTFS I'm aware of is slow new-file
> > > creation when
> > > > >> > dealing with orders of tens of millions of files.
> > > > >>
> > > > >> There are limitations on file sizes and numbers, as there must be...
> > > > >> luckily, the max filesize with NTFS is huge, but it wont be long before
> > > > >> people are hitting that limit too (if they haven't already).
> > > > >
> > > > >16 Exabytes ???
> > > > >16 billion Giga byte.
> > > > >
> > > > >I'm not sure exactly *what* you can put into a file to get into that
> > > size.
> > > >
> > > > Precisely what they said about the 2 Gigabyte limit. ;-)
> > > >
> > > > And they were really sure *they* were right, too. ;-)
> > >
> > > Difference is in the size.
> > > And the 2GB limit in what exactly? FAT has it (actually, it's a partition
> > > limit, but that is beside the point) but it's justifiable, FAT was designed
> > > in the 70s.
> > > Linux on 32bit has(d) it, it's not justifiable, because need for such files
> > > exist for a long time,
> >
> > i agree. linux should move to 64 bit size_t for files regardless of
> > processor. linux-2.4 will do large files, but C is a cranky beast
> > sometimes and updating software can be cumbersome.
> >
> > > I can assure you that there was no need for 2GB files
> > > in the 70s, when FAT was designed.
> >
> > unix style filesystems with the inodes &c were also designed in the
> > 70s. however, it's not the age of the filesystem design. it's the
> > also competence of the design and the goal of the design. FAT was
> > made for floppies and tiny systems. unix filesystems were made for
> > hard drives and larger systems. it's still MS's fault for keeping
> > such a bad design as FAT and trying to keep it going where it doesn't
> > belong, but age is not the issue.
>
> It's interesting then, now that FAT has moved on, whereas ext2fs
> has not.
ext2fs is only a few years old.
> (NOTE: I realize FAT sucks, I'm not trying to claim it's
> better than ext2fs, just more updated).
unix filesystems were expanded from plain inodes, to indirect inodes
and later to double and tripple indirect inodes to keep up with disk
drive size increases. unix' filesystem design has been patched and
kludged since the 70s just like cp/m or mircrosoft FAT.
btw i am using reiserfs for about a year. so far i like it.
--
J o h a n K u l l s t a m
[[EMAIL PROTECTED]]
Don't Fear the Penguin!
------------------------------
From: Shane Phelps <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Comparison: Installing W2K and Linux 2.4
Date: Fri, 26 Jan 2001 11:08:46 +1100
Conrad Rutherford wrote:
>
> "Paul Colquhoun" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > On 25 Jan 2001 13:35:25 -0600, Conrad Rutherford <[EMAIL PROTECTED]>
> wrote:
> > |
> > |"Philip Van Hoof" <[EMAIL PROTECTED]> wrote in message
> > |news:[EMAIL PROTECTED]...
> > |> Conrad Rutherford wrote:
> > |> >
> > |> > OK, lesse...
> > |> >
> > |> > W2K:
> > |<snip>
> > |
> > |Ahhh, I clipped all your crap because you simply skipped my point and the
> > |thread. You didn't address my text - instead created a strawman and some
> > |more made up scenarios of your choosing...
> > |
> > |Can't take the heat...
> >
> >
> > Sorry, *you* are accusig *him* of using "made up scenarios of your
> choosing"
> >
> > That must be the best example of the pot calling the kettle black that
> > I've seen in a long time.
>
> Gee, my scenario is how to install W2K versus how to install the 2.4
> kernel - is that a big stretch of the imagination for something someone
> might do? I didn't specify hardware or any conditions except, how to get it
> running BASICALLY.
>
> wow - this must have really hit a nerve... P)
The 2.4 kernel was officially released a couple of weeks ago. W2K was
officially released in Feb 2000, but the RC was frozen around December
1999 from memory. Official release of a Linux kernel is more akin to MS
freezing a RC and releasing it to manufacturing for the CDs to be burned.
The chief difference is that Linux kernels can be downloaded. You
don't have to wait for the shrink-wrap release.
It's a rather different distribution model
The 2.4 kernel is available in RPM format from a number of sites.
So is W2K SP1
You can compile the 2.4 kernel if you want to, or download the RPM
if you prefer, or wait for Redhat or Suse or Mandrake or TurboLinux
or Storm or... to release it in their next distro. Or stick with an
earlier kernel.
Similarly, you can download W2K SP1 from ms.com (tricky at present)
or tucows or .... if you prefer, or you can downaload each individual
component of SP1 (I assume) and install separately or wait for a cover
CD on a PC magazine or just plain not bother.
Please explain what the big difference in installation there is.
You are comparing 2 different things. A better comparison may be
2.4 vs Whistler, which is apparently now in early Beta.
BTW, in case you missed it above - an "official" Free Software or
Open Source release is akin to release to manufacturing (or burning
if you prefer) of a commercial shrink-wrap product.
I have noticed that a number of commercial software vendors allow
downloading of pre-release software (possibly time-bombed) before
the shrink-wrap version is out.
------------------------------
** 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
******************************
