Linux-Advocacy Digest #871, Volume #25 Wed, 29 Mar 00 10:13:04 EST
Contents:
Re: Open Software Reliability (Donal K. Fellows)
Re: Giving up on NT (Bob shows his lack of knowledge yet again) ("LP")
Re: Peter Norton is one smart dude (Mark Hamstra)
Re: Giving up on Tholen ("Joe Malloy")
Re: Penquins Forever! Was (Re: A pox on the penguin?) (Mark S. Bilk)
Re: Windows 2000: nothing worse ("Chad Myers")
Re: Peter Norton is one smart dude (John Sanders)
Re: BEOS 5 the new star in OS's (Tim Kelley)
Re: Why did we even need NT in the first place? (Brian Langenberger)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Donal K. Fellows)
Subject: Re: Open Software Reliability
Date: 29 Mar 2000 11:45:51 GMT
In article <8astfe$[EMAIL PROTECTED]>,
david parsons <[EMAIL PROTECTED]> wrote:
> It comes down, in the end, to the problem that good coding is an art
> that requires a lot of discipline to do properly. Most programmers,
> whether they're doing it for the fun of it, for their chance to get
> into a Linux IPO, or doing it for a living, don't have that
> discipline and will produce horrible unmaintainable code.
All the well-written non-toy code that I've ever seen has had good
documentation that has described the purpose and side-effects of each
function. Not that documentation ever makes code good, oh no...
Donal.
--
Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ [EMAIL PROTECTED]
-- The small advantage of not having California being part of my country would
be overweighed by having California as a heavily-armed rabid weasel on our
borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>
------------------------------
Reply-To: "LP" <[EMAIL PROTECTED]>
From: "LP" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy
Subject: Re: Giving up on NT (Bob shows his lack of knowledge yet again)
Date: Wed, 29 Mar 2000 12:54:27 GMT
<[EMAIL PROTECTED]> wrote in message
news:ZBwD4.13510$[EMAIL PROTECTED]...
> George Marengo writes:
>
> >>>> George Marengo writes [to Bob Germer]:
>
> >>>>> You're absolutely correct -- but what does that have to do with the
> >>>>> Judges opinion being just that, an opinion?
>
> >>>> You called that opinion a fact.
>
> >>> The Judge has opinions, just like I do.
>
> >> Do you also call your opinions "facts"?
>
> > Nope.
>
> So why do you call the opinions of a judge "facts"?
>
> GM] The facts referred to are the legal opinion of a Judge
>
> > Do you?
>
> Irrelevant, given that you're the one who called a legal opinion
> "facts".
The US system of justice does not concern itself with empirical reality, only legal
reality.
The "findings of fact" only refer to the legal findings. For example, one can be
legally guilty of murder, yet in reality, have
not conducted the murder. The "fact" is that you have been found guilty of murder.
This "fact" has nothing to do with whether or
not you actually have done it.
------------------------------
From: Mark Hamstra <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Peter Norton is one smart dude
Date: 29 Mar 2000 08:03:22 -0500
[EMAIL PROTECTED] (Christopher Browne) writes:
> In other words, the fact that there are four groups putting
> substantial efforts into journalling filesystems on Linux [ReiserFS,
> ext3, XFS, IBM JFS] is being "spun" into purported irrelevance.
More, actually. I'm not sure about the journaling status of the
Netware-based filesystems recently announced by Jeff Merkey and
Timpanogas, but the commercial VxFS is definitely a journaling
filesystem, and Veritas has announced their intention to release
a Linux version of VxFS this year. GFS, the Global Filesystem,
is also another (and very interesting) journaling filesystem for
Linux.
--
Mark Hamstra
Bentley Systems, Inc.
------------------------------
From: "Joe Malloy" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy
Subject: Re: Giving up on Tholen
Date: Wed, 29 Mar 2000 08:13:12 -0500
"Little man" Tholen attempts to obfuscate and tholens:
> Jim "little boy" Stuyck writes:
>
> > [EMAIL PROTECTED] wrote:
>
> Still can't get your attribution straight, eh Stuyck (little boy)?
There is no error in Jim's attribution line, "little man" Tholen. Those
first four letters stand for "Kook of the Month" and that's what you were
duly elected, for March, 1998, thus attesting to your exploits on
uselessnet, "little man."
> >> Jim "little boy" Stuyck writes:
>
> >>> [EMAIL PROTECTED] wrote:
>
> >> Still can't get your attribution straight, eh Stuyck (little boy)?
Again, no error on Jim's part, "little man" Tholen. You're definitely a
kook, that's for sure!
> > Where did I go wrong?
>
> The first four letters, Stuyck (little boy).
Yet *you* responded anyway, as you have for years now, "little man" Tholen.
That proves Jim's attribution is correct, "little man" Tholen. QED, "little
man" Tholen.
--
"USB, idiot, stands for Universal Serial Bus. There is no power on the
output socket of any USB port I have ever seen" - Bob Germer
------------------------------
From: [EMAIL PROTECTED] (Mark S. Bilk)
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Penquins Forever! Was (Re: A pox on the penguin?)
Date: 29 Mar 2000 13:49:32 GMT
Robert Heininger <[EMAIL PROTECTED]> wrote:
>ax <[EMAIL PROTECTED]> wrote:
>:Robert Heininger <[EMAIL PROTECTED]> wrote in message
>:>ax <[EMAIL PROTECTED]> wrote:
>:>:Matthias Warkus <[EMAIL PROTECTED]> wrote in message
>:>:>Robert Heininger <[EMAIL PROTECTED]> wrote:
>:>:>>Matthias Warkus <[EMAIL PROTECTED]> wrote:
>:>:>>:ax <[EMAIL PROTECTED]> wrote:
>:>:>>:>But in Linux, all penguins are lazy sitting with round belly.
>:>:>>:>They cannot walk or fly. They cannot even stand up
>:>:>>:>with fat belly. Linux penguins must have been eating
>:>:>>:>too much "free" stuff.
>:>:A really smart signature!
>:>: - a big "L" shape at the bottom.
>:>: - a word "Linux" on top of the "L" shape.
>:>: - a "<" shape at the end of the "Linux".
>:>:It means, "Linux" is being pushed ("<") to the corner ("L"),
>:>:which is the choice of the GNU Generation.
>:Your signature means a lot. Here is another one:
>: - a big "L" shape representing a chair
>: - a big word "LINUX" sitting on a chair
>: - a big "<" shape representing a lock
>:It means: lock "Linux" to a chair, which is the "Choice
>:of the GNU Generation".
>:Now, everything in Linux are "sitting" including all
>:heavy belly Linux penguins and "Linux" itself. Keep sitting!
The above quotes are all from "ax" -- [EMAIL PROTECTED] He also
wrote: "'consumer view' may not matter at all to Linux
Penguins. Linux Penguins are all blind deaf-mutes."
If you look him up in DejaNews, you'll find that all the
articles from this anonymous person are about Linux, and
almost all of them are very negative toward it. In one
large thread that he started, which has the word "Microsoften"
in the Subject, he repeatedly complained that all four Linux
distributions that he tried would not support his hardware,
including an HP laser printer. Since such printers are in
fact well supported, he was asked repeatedly to name the
exact printer that he has, so he could be helped with it, but
he refused to do so, and kept complaining.
"ax" has also claimed several times that Open Source is
destroying the software industry, and that it's leading to
a multitude of closed systems, both of which assertions are
clearly untrue. His opposition to Open Source, expressed as
"free stuff", also appears in his remark quoted above. He
additionally wrote that he couldn't get any help with Linux,
while the evidence in his posts is that he refused to cooper-
ate with the people who were trying to help him.
All of this behavior supports the conclusion that "ax" is yet
another anonymous anti-Linux propagandist. The fact that he
refuses help with his alleged failed installations, and goes
out of his way to fabricate nasty characterizations about
Linux, like the four quoted above about penguins and your
signature, make it unlikely that he is simply confused or
innocently mistaken.
What all of this means regarding you is that since "ax" seems
to be trying to impede the Linux/Open Source movement, you
should absolutely *not* follow his advice about anything,
including your signature.
>:>:>A five-line Figlet signature! Jesus Christ!
Now this, on the other hand, is from mawa, a very knowledge-
able and expert Linux user, whose credentials are beyond
reproach. However, I think he is being rather narrow and
picky about what constitutes a suitable signature. For some
reason he seems to have an aversion to "figlets" (large fonts
made by combining ASCII characters). Maybe this originates
from a concern about saving Net bandwidth, back when the
capacity of the Internet was only 1% of what it is now.
Anyway, the pro-Microsoft spammers waste thousands of times
as much bandwidth in c.o.l.a, because they post a hundred or
more articles per day here of Microsoft propaganda, most of
which have nothing (or nothing truthful) to do with Linux,
and are therefore totally inappropriate.
>:>YIKES! I get the point. Thanks all!
Look, the anti-Linux "ax" has posted 3 or 4 nasty, lying
comments against your signature. That probably means that
he thinks it's effective, and therefore he doesn't want
people to see it. So, since you obviously want to boost
Linux, you *should* use it. As to mawa, I think he's
expressing a personal preference, and does not represent
anyone but himself on this matter. I'm a longtime Usenet
poster myself, and I think your signature is excellent --
well designed and with a good message.
If you feel like adding a penguin to it, here's one:
-o) Pascal Bleser
/\\ <[EMAIL PROTECTED]>
_\_v C++/UNIX Development,
ATOS Payment Systems (Aachen)
From: Pascal Bleser <[EMAIL PROTECTED]>
He can look to the right, too:
Robert Heininger __
# / / __ _ _ _ _ __ __ #
(o- # / /__ / / / \// //_// \ \/ / #
//\ # /____/ /_/ /_/\/ /___/ /_/\_\ #
v_/_ # The Choice of the GNU Generation #
>It means:
>
>1) Spreading the word about the Linux revolution.
>2) That I'm proud about choosing to be a member of the GNU generation.
>3) It's not commercial spam.
>4) Your mileage may vary.
>5) Optional: If folks don't like it, I suggest that they put me
> in their kill filters, and/or ignore my postings.
Right on!
------------------------------
From: "Chad Myers" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Windows 2000: nothing worse
Date: Wed, 29 Mar 2000 14:22:22 GMT
"Jim Richardson" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> >But it's not that big of a deal, really, because IIS runs user connections
> >with different user contexts (usually the anonymous account IUSR_SERVERNAME,
> >but if they log on using NTLM or SSL/Basic, then they assume that user's
> >security context, and thusly, that user's permissions and rights.
> >
>
> hm, I don't see the diff from apache et-al really. (but that may or may
> not have been the point, I don't recall :)
The point originally was made that IIS runs everyone as the "inetpub" user
and has no concept of security or user context, I believe. Which, of course,
is completely false, as I just illustrated.
<honest sincerity>
JOOC, (off topic somewhat), how does apache handle anonymous users and
authenticated users?
</honest sincerity>
> >> >What happens if you lose the root password?
> >> >
> >>
> >> reboot into single user mode and fix it. Reboot required, no reinstall
> >> required.
> >
> >And this is a good thing?
> >
> it is for the casual home user, if it's a big time data center, then
> they better start with physical security and work up from there. If
> they don't have phys-sec, they do not have a secure system. OS doesn't
> even enter into it until after physical security is taken care of.
But with NT (SP3 and later), if your physical security was somehow
compromised, it would be rather difficult and time consuming to get
access to the machine. With Win2K, if the files are EFS, it's near
impossible to ever get the files out.
It's frightening to know that someone could just walk up with a boot disk
in *nix and set the root and have their way with the system...
I suppose you could PGP the important stuff, though...
-Chad
------------------------------
From: John Sanders <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Peter Norton is one smart dude
Date: Wed, 29 Mar 2000 08:22:19 -0600
Drestin Black wrote:
>
> "R.E.Ballard ( Rex Ballard )" <[EMAIL PROTECTED]> wrote in message
> news:8bpglc$til$[EMAIL PROTECTED]...
> > In article <[EMAIL PROTECTED]>,
> > Gary Hallock <[EMAIL PROTECTED]> wrote:
> > > "Stephen S. Edwards II" wrote:
> > >
> > > > AFAIK, the journaling filesystems for Linux
> > > > are still under development.
> > > > Are reiserfs, Xfs, or Jfs in widespread use yet?
> >
> > Actually, journaling filesystems are not as much of a priority
> > for Linux as they were for NT. It's likely that Windows 2000
> > will eventually drop them.
>
> where the fuck do you come up with this shit? eventually drop them... yea,
> and I heard Windows 2001 dropped support for video cards too...
>
> >
> <snip boring FS crap>
>
> > Windows Fat 16 and Fat 32 stores files in large clusters ranging
> > from 4k to 32k per cluster. Windows NTFS supports 512 byte blocks
> > but still orients storage toward clusters. As a result, it's
> > undesirable to have lots of tiny little files. This means that
> > the ability to recover a file corrupted by a race condition related
> > crash is much more important. Linux and UNIX allocate blocks in
> > units of either 512 bytes or 1024 bytes (depending on the original
> > formatting).
>
> No different than the 512 byte blocks NTFS uses - where do you come up with
> this "orients" towards clusters shit?
>
> Furthermore, ext2 directories are designed so that
> > hundreds of files can be stored in a single directory and each file
> > can be quickly located. This makes it more practical to put hundreds
> > of small articles (news postings, e-mails, receipts, orders,...) into
> > individual files under a single directory.
>
> Wow - just like NTFS
>
> >
> > Additionally, tools like grep, awk, sed, and perl make it easy to
> > create summaries of these tiny little documents in very short time.
> > The use of pipelines for update systems makes storage of logs via
> > the "tee" command very easy. In fact, in modern versions, both the
> > storage and the pipelines are usually optimized to minimize latency.
>
> yawn - been there, done that, fortunately tools under NT are better,
> thanks...
>
> >
> > When you really need a large database, the databases have their own
> > journalling system.
>
> gee, like SQL Server...
>
> >
> > > > Is there another one besides these that works well?
> >
> > Journalling is a bit like compression. At one time companies
> > like Stack and later Microsoft thought it would be a really neat
> > thing to compress everything on the hard drive. They started
> > compressing each track, and eventually the entire drive. The
> > problem was that much of the content was already compressed (GIF,
> > JPEG, and ZIP files for example). Ironically, the zip file is
> > actually the decendent of two UNIX commands called ar and compress.
> > Eventually, PKWare and the authors of ARC reached a settlement,
> > and PKWare improved the compression as well as the indexing.
>
> And that has what to do with the topic?
>
> >
> > In Windows, you created problems when you tried to doublespace
> > the entire drive. You couldn't upgrade, and if you messed up
> > the system too badly, you had to start all over again. A corrupted
> > byte in a single sector could corrupt the entire filesystem.
>
> where DO you come up with this CRAP?!
>
> > Most people today preferr to compress the files they want to archive,
> > and keep the current files uncompressed. We also use compressed
> > formats such as GIF (compressed BMP), JPEG (compressed photographs),
> > PDF (compressed simplified postscript), and of course the zip file.
>
> yawn... point?
>
> >
> > In UNIX I could compress any file, send it across the internet,
> > and uncompress it on the receiving end. It's a paridigm shift.
>
> wow - no shit? you mean, just like I do when I use ZIP? (or ARC back when)
>
> > In the Windows paradigm, compression must be written into a monolithic
> > applications which is compiled into the production version. Once
> > the function is added, it cannot be altered, replaced, or removed.
>
> HA!
>
> > In the UNIX paradigm, compression is a simple command that takes
> > whatever comes in the standard input, compresses it, and pours it into
> > the standard output. This isn't terribly efficient for things like
> > sorting, but it's very good for things like encryption, compression,
> > journalling, back-up, recovery, filtering, and summarizations.
>
> You think that is very good? HA! You didn't mention the built-in file system
> compression NTFS has ... of course.
>
> >
> > Journalling is a similar function of the paradigm shift. Since the
> > monolithic Windows applications generally don't support journalling,
> > archiving, and revision control, you must make these features a
> > function of the operating system. Ideally, you'd make journalling
> > an option at the directory level, so that you could store stable
> > information into unjournaled locations and store volitile information
> > (checks, receipts, registrations,...) into the journalling directory.
> >
> > > > NOTE: I am not baiting... I'm really asking out of curiosity.
> >
> > It's nice to see some civilized discussions in this group.
>
> how would you know?
>
> >
> > > Suse now ships with reiserfs.
> >
> > Typically, in a Linux system, you might put one of your partitions
> > under reiserfs because that's were you wanted to put checks. But
> > you wouldn't just casually journal everything you did.
> >
>
> fortunately with Windows we just decide on a file by file or folder by
> folder or drive by drive basis if we want compression or not and with a
> simple click of the mouse we can compress or uncompress anything we'd like
> and the File system takes care of everything including automatic
> decompression when the file is accessed. If we need compression on a file
> for transmission we turn to the ZIP or RAR or ACE formats and their
> appropriate applications. All of which support full integration into the GUI
> so that making and breaking a compressed file is as easy as a quick click.
> (and unlike unix we handle ALL formats of compression from a single app,
> ACE, ZIP, LHA, MS-CAB, RAR, ARC, ARJ, GZIP, TAR, ZOO whatever...
I was editing a file in vi the other night. I decided to take a
break. I don't save it, 'cause I'm going to be right back. Stretched
out on the couch and fell asleep. Storm comes through. Wife literally
pulls the plug on my computer. Next day, I boot up. Oh! I've got
mail. It's vi telling me that I can recover my file from the night
before. Sure enough, there it all is, right where I left off. And I
didn't click on anything. Of course, this file was never compressed
anywhere, and that really makes me sad, but I do have my file back. Who
needs a journaling file system? :-)
Let's really drive this one into the ground with some additional
threads: "Why Journaling for the Desktop?" or "Journaling: Will the
average user click on it?" or "Giving up on Journaling" or "Journaling:
What's the REAL TOC?"
--
John W. Sanders
===============
"there" in or at a place.
"their" of or relating to them.
"they're" contraction of 'they are'.
------------------------------
From: Tim Kelley <[EMAIL PROTECTED]>
Subject: Re: BEOS 5 the new star in OS's
Date: Wed, 29 Mar 2000 08:28:13 -0600
piddy wrote:
>
> BEOS 5 is ready for downloading, but the site is too busy.
>
> Fortunately you can download BEOS at other places. Check the Beos
> newsgroup.
> Here's a fast site:
> http://download.cnet.com/downloads/0-10108-100-1594977.html?tag=st.dl.10000_
> 103_1.lst.td
>
> Btw, it's well worth taking a look at. It's fast, looks great, is as
> easy as the Mac. Just click around and you can figure things out.
> Don't count on it working with Win-modems and the off brand sound
> cards though.
>
> It defaulted to 640x480 on my computer, but in less than 5 minutes I
> found a way to adjust it to 800x600 and changed the refresh rate from
> 56 to 72 to cut down on flicker. Try that with Linux! My wheel mouse
> worked and scrolled most windows. I'm seriously thinking of getting a
> different modem and sound card and using this for web browsing, file
> downloading, and fun stuff.
>
> If it had apps, I'd say it had an excellent chance to make it big.
>
> piddy -- Linux now sucks more!
Tell ya what, dippy, or piddy, or whatever, BeOS ain't gonna make
it unless linux does. Unless the defacto OS is something
non-proprietary, there won't be any room but for one.
Be may be easier to get started in than linux, but it's
application availability problem is (by far) worse. Especially
the types of programs that should be running on it (audio, video,
etc.).
--
Tim Kelley
[EMAIL PROTECTED]
------------------------------
From: Brian Langenberger <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Why did we even need NT in the first place?
Date: 29 Mar 2000 14:41:51 GMT
In comp.os.linux.advocacy Chad Myers <[EMAIL PROTECTED]> wrote:
<snip>
:> : b.) doesn't require a group of people to maintain
:>
:> ...and I maintain my box without any help.
: How many users on your network? For a large network, Linux usually requires
: constant care if you plan on doing anything on it other than DNS or WWW.
On a large network, the server boxes typically require the least amount
of care - Linux included. More than likely the administrators
(plural, by necessity) will be working to ensure the vast amount of
client machines are working properly. Even if the OS is zero maintenance,
that's still a lot of hardware to check up on.
<snip>
:> I have more than 2 gigs of files on my system now.
:> Oh, but if you mean *individual* files larger than 2 gigs, Linux
:> supports that now also - just get some real hardware like a nice
:> 64bit Alpha box. If you want to use substandard hardware, you
:> have to expect some limitations.
: 2GB in one file. And no, I don't want to drop thousands more for my server
: when I can do it just fine in a nice Compaq x86 server with NT. Why go
: up to 64-bit and pay all that to have functionality I can already get for
: a fraction of the price?
Much the same way a 386 has all the functionality of an Athlon.
Compaq themselves places their AlphaServer as "the proven performance
leaders, powered by speed-burning Alpha processors."
http://www.compaq.com/products/servers/index.html
If you need 2gig+ files and the performance to handle them,
why go cheap? Get a nice Alpha or Sparc system and watch the
limitation disappear.
:> : -oh yeah, and it scales well. It actually lives up to the term SMP, unlike
:> : Linux.
:>
:> Solaris actually lives up to the term SMP. Think 128 processors.
: It can recognize 128 processors, but it doesn't use them. It's scalability
: beyond 16 or so processors drops off rapidly. Whereas with a Win2K cluster
: for a fraction of the price, I can get factors larger performance.
: And! NT/2K actually can use more than one NIC with more than one processor
: and not have it bottleneck like Linux!
Doesn't use them? The actual number of processors is
64 (my mistake), but Solaris certainly does use each and
every one with a linear scale in performance. Your original
assertion was that NT was some master of SMP, but in actuality
it requires clustering to get the same benefits that a Solaris
box enjoys in a single box.
If you want to change that assertion to mean that NT lives up
to the term SMP on cheap clustered hardware, then you have to
contend with Beowulf clustered Linux boxes which offer the
same sort of benefits on the same sort of hardware.
:> Linux does well enough considering the hardware, but I don't think
:> x86 can hold a candle to Sparc in the SMP department.
: View the TPC-C benchmarks. Note the #1 and #2 slots.
TPC-C is a database benchmark, not an OS benchmark.
According to the TPC-C specification itself (page 6):
"The relative performance of systems derived from this benchmark does
not necessarily hold for other workloads or environments.
Extrapolations to any other environment are not recommended."
available at:
http://www.tpc.org/cspec.html
In short, TPC-C is not a general purpose benchmark.
:> Does Windows run well on modern hardware (like 64-bit+)?
: Win2K was running on Alpha (still could, just not supported),
: and it is running on Merced.
I'm glad to see it catching up.
: Just because Linux runs on Alpha, doesn't mean it's 64-bit, nor does it mean
: it runs well on Alpha. Likewise with Sparc. It'll run, but there are
: much better
: OSen for those platforms.
If Linux is compiled on a 64 bit platform, with 64 bit data types
with a compiler that generates 64 bit code, why wouldn't it use the
hardware accordingly?
As for Sparc, on single processor machines Linux actually outperforms
Solaris. For multiprocessor Sparc boxes, Solaris is the better choice -
which is why it is such an SMP powerhouse. And for both Alpha and
Sparc a variety of UNIX is the only way to go - simply because Windows
isn't offered for either that I've seen.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.advocacy) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
