Linux-Advocacy Digest #8, Volume #27 Sat, 10 Jun 00 13:13:07 EDT
Contents:
Re: Dealing with filesystem volumes (Dave Vandervies)
Re: OSWars 2000 at www.stardock.com ([EMAIL PROTECTED])
Re: Open Source Programmers Demonstrate Incompetence ([EMAIL PROTECTED])
Re: Linux faster than Windows? ([EMAIL PROTECTED])
Re: Microsoft W2K lack of goals. (Dave)
Re: Dealing with filesystem volumes (The Ghost In The Machine)
Re: democracy? (Robert J Carter)
Re: Dealing with filesystem volumes (abraxas)
Re: Open Source Programmers Demonstrate Incompetence ("KLH")
Re: Dealing with filesystem volumes (abraxas)
Re: Canada invites Microsoft north (Mayor)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Dave Vandervies)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.advocacy,comp.unix.advocacy
Subject: Re: Dealing with filesystem volumes
Date: 10 Jun 2000 14:40:10 GMT
In article <[EMAIL PROTECTED]>, Cihl <[EMAIL PROTECTED]> wrote:
>"Lawrence D=B9Oliveiro" wrote:
>> =
>
>> Which is the best system for dealing with filesystem volumes?
>> =
>
>> c) Per-volume filespecs (MacOS, though I think it was originally
>> invented on the original UCSD p-system).
>> Pros: System-independent file specifications. A reference to a file on =
>a
>> removable/hot-pluggable volume doesn't depend on the location of its
>> "mount point", drive ordering or any other system-specific
>> configuration: The same reference will work on any other machine
>> mounting the same volume. The reference can also be used to
>> automatically request the mounting of the correct volume on demand.
>
>I had to read this several times to understand. Doesn't sound that
>different from Unix, to me.
One difference is that (if I'm interpreting the description correctly)
it assumes that if a volume is mounted as `My Files' on one system, it
will be mounted as `My Files' on _any_ other system it happens to be
moved to; I can mount a filesystem as `/usr/local' on one computer and
as '/mnt/goofy/usr/local' on another using the Unix model. Being able
to do this has made life a lot easier on more than one occasion.
>
>> Cons: Can't think of any.
>
>Try harder, this time without any bias.
> =
A few that come to mind immediately:
-Namespace conflicts - what happens if you want to move `My Files' to a
system that already has a volume by that name? Any reasonable solution
to this that I can think of immediately removes the only point the OP
put forward in favor of this model, at least for that particular
volume.
-Really awful implementation for people who don't like to point and
click - `Local User Files:Binary and Executable Files' vs.
`/usr/local/bin' - 'nuff said. This can be redone, but why bother when
it's already been done right in the Unix world?
dave
--
Dave Vandervies
[EMAIL PROTECTED]
All my friends and I are crazy. That's the only thing that keeps us sane.
------------------------------
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.be.advocacy,comp.sys.mac.advocacy
From: [EMAIL PROTECTED]
Subject: Re: OSWars 2000 at www.stardock.com
Date: Sat, 10 Jun 2000 09:35:36 -0400
[EMAIL PROTECTED] (rj friedman) said:
>On Sat, 10 Jun 2000 01:37:41 "Brad" <[EMAIL PROTECTED]>
>wrote:
<snip>
>=BB> =BB In fact, it
>=BB> =BBwas the necessity of moving to Page Maker 6 that forced me to sw=
itch from
>=BB> =BBOS/2 in the first place. What do I use Page Maker for? Technica=
l =BB>
>=BBdocumentation.
I've seen Stardock documentation and I've worked in technical writing for=
more
years then Brad has been an adult. I wouldn't hire anyone there as a writ=
er.
>=BB> I've used Page Maker - I pity anyone who used it for
>=BB> technical documentation. It has to be the worst tool for the
>=BB> job I have ever had the misfortune to run across. It's
>=BB> strength is single page layout - you can use it to do
>=BB> manuals, but there are so many other tools out there that
>=BB> are better for the job.
>=BBThis coming from someone using an older version of Framemaker via OS/=
2's
>=BBWinOS2 support.
Ah. An acid-test moment: The actual functional difference between FM 5.1 =
and
FM 5.5/6.0 is?
>"How old" is irrelevant to how the tool gets the job done.
>Anyone with an ounce of brains knows that. I suppose - since
>your livlihood depends on it - you want to convince everyone
>that you have to have the latest and greatest instead of
>"the best tool for the job" so you can sell more software.
>But people who actually use tools to do their work don't
>fall for that baloney.
True.
>=BB> Choosing to leave OS/2 so you can use Page Maker 6 has to be
>=BB> the most pitiable reason I can think of. Frankly, I don't
>=BB> believe you.
It depends on how one chooses to address the root of their un-productive =
time.
===========================================================
[EMAIL PROTECTED]
===========================================================
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Open Source Programmers Demonstrate Incompetence
Date: Sat, 10 Jun 2000 15:20:47 GMT
In article <[EMAIL PROTECTED]>,
Cihl <[EMAIL PROTECTED]> wrote:
> I take it you wrote this yourself, and nobody in the open-source
> community liked it? :-)
The code or the message? I certainly didn't write the code - nothing I
have written, at least since I've been a professional, has been nearly
that bad. I tend to cringe if I code a conditional branch where it
wasn't needed, but the code I quoted is just all out molasses.
> Sorry to burst your closed-source bubble, but closed-source software
> is no better in this matter, and possibly much worse too, because
> nobody else but the company who made the program ever looks at them!
Not in my experience. I've worked in closed source professionally, and
there are typically source code reviews where other enginers formally
review your source code line by line, and then debate it openly (with
the author). It is almost impossible for fluff like the above to get
by.
> Just curious, where did you get these examples, exactly?
>From the "code snippets" page on SourceForge - just go to that site,
click on code snippets, then go to "C", then look for the ones I quoted.
> I'm sure if anyone else saw this piece of code, the (poor) guy who
> made this would probably be flamed and laughed at and pestered right
> out of SourceForge!
I am glad you agree that the code is really bad. :-)
> One of the biggest strengths of open-source is that this kind of code
> gets noticed and eradicated almost immediately.
But that's the thing. For this code snippet system on SourceForge, you
have the ability to upload your improved versions. These two snippets
date from January and February (IIRC). It's June now, so in the past
four or five months (which is kind of above the upper limit of "almost
immediately"), nobody has uploaded a replacement.
Now you can argue that "code snippets" is not the most important work,
not work which people are interested in fixing, since it may not be
part of a larger work. There's some merit in that, but I am just amazed
people submitted such horrible code.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Linux faster than Windows?
Date: Sat, 10 Jun 2000 15:33:47 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Pete Goodwin) wrote:
> 6 minutes with an Alpha! OUCH! I'd have thought Alpha's would run
much
> faster than that. Perhaps you ought to try Alpha-Linux or Digital
UNIX?
Unix or Linux? Blasphemy! The reason VMS does it slowly is because of
the write through cache; each record is put on the physical disk as
soon as it is written. If the power goes out in the middle of the
write, VMS would have half of the file, but Unix, Linux, and Windows
wouldn't have anything. Indeed, for the CPU benchmarks, on identical
hardware, VMS wipes the floor with Linux. I'd also bet that VMS is much
faster at certain file benchmarks (e.g. delete a random record in this
30 MG file).
> That is interesting, Linux is _slower_ than Windows 98 SE. So much
for the
> outrageous claims of a three fold speed increase!
Indeed. Linux "feels" fast - largely because of the aggressive disk
cache and superfast console driver. I remember when I first used Linux,
in 1993, after using only DOS, I did an "ls" and was absolutely wowed -
it showed up instantly and didn't even appear to scroll by (on an old
386SX). A lot of Linux zealots are deceived into believeing that Linux
is therefore "fast". But "feel" is hardly the benchmark of a system;
most benchmarks I see which do sophisticated performance measurements
do not show up faster on Linux than other systems (including Linux).
A lot of people compare apples to oranges too. It is common knowledge
that Linux compiles code three times faster than Windows. What does
this mean? Is it comparing using the same versions of GCC with the same
flags? Or is it comparing Microsoft's compiler (which does more
sophisticated optimizations) with the optimizing flags on, to GCC with
no flags on? They never state that part ...
Another interesting benchmark would be mail programs. I use Outlook,
which has a sophisticated indexed system. My mail box has more than
5,000 messages, and Outlook can open this mailbox (on a dual Pentium
133, my main Windows workstation), in less than 2 seconds. Unix mail -
which uses nothing but a huge text file, and has to parse through every
byte (!) to open the mailbox, is very slow. On fast Unix systems, such
as Sparc's, it takes several _minutes_ to open a mailbox of 200-300
messages (compared to 2 seconds for a mailbox over 10x bigger on
Windows). Mailboxes over that length are absolutely unwieldy for Unix,
but no problem with a sophisticated mailer (security issues
notwithstanding).
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Dave <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Microsoft W2K lack of goals.
Date: 10 Jun 2000 10:45:06 -0500
In article
<[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> On 2 Jun 2000 21:04:15 -0500, Dave <[EMAIL PROTECTED]> wrote:
> >In article <NeXY4.1007$[EMAIL PROTECTED]>, "Tim"
> ><[EMAIL PROTECTED]> wrote:
> >
> >> Are you $#$#@#$& insane? I really hope your kidding. Please tell me
> >> you
> >> are.
> >> I mean come on, your talking out of your bunghole.
> >> I guess this is just a trolls bait, and I took it... but anyway if you
> >> really beleive that
> >> you are a complete ignoramous.
> >> Come on, I dare you to say it, Windows 2000 is faster than Windows 95.
> >> Do
> >> you
> >> have half a brain in your head?
> >
> >I've got Win2000 Server and Win98se on the same machine here. PII 400,
> >192 megs ram. Win2000 Server is *much* more responsive as a desktop OS
>
> It needs 192 megs of ram? THAT IS PATHETIC!
Did I say it *needed that? No, it's just what I have. I think the
minimum for Server is 128 megs, 64 megs for Win2000 Pro.
It just runs faster (as *any* real OS will) with more memory.
Even if it did "need" 192 megs - so what? 192 megs is cheaper today
than 2 megs was in 1988. In those days ram was $500 a meg!!!!! In 10
more years we'll all be running dual 4 ghz processor machines with 64
gigs of ram and 16 terabytes of disk space. And it will *still* cost
under $2000.
People like you will be pissing and moaning about Windows 2010 (or
whatever) needing a gig of ram and waxing philosophic about the "good
old days of lean and mean Win2000 that only needed 128 megs"!!!!
Dave
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.advocacy,comp.unix.advocacy
Subject: Re: Dealing with filesystem volumes
Date: Sat, 10 Jun 2000 15:47:47 GMT
In comp.os.linux.advocacy, Lawrence DčOliveiro
<[EMAIL PROTECTED]>
wrote on Sun, 11 Jun 2000 00:58:55 +1200
<[EMAIL PROTECTED]>:
>Which is the best system for dealing with filesystem volumes?
>
>a) Drive letters (all versions of Windows OT and NT, including Windows
>2000).
>Pros: You got to be kidding.
Uh....short prefixes? :-) But you're right, there's not much
in the way of pros, or praise, here.
>Cons: They reaassign themselves at the slightest excuse; add a new
>drive, and all bets are off as to which of your existing drive letter
>assignments will stay the same.
They're (relatively) hard to parse in programs. Instead of simply
skipping through the tulips -- erm, I mean, slashes -- one has to
actually do a little bit of work, checking the first letter.
>Verdict: Stupid 1970s way of doing things that should be ashamed to be
>still showing itself in the 21st century.
>
>b) Mount points (all UNIXes and Linsux).
>Pros: Pretends to make all your volumes look like a single filesystem.
>Cons: Only *pretends* to make all your volumes look like a single
>filesystem (all kinds of within-file-system-only things don't work, like
>hard links). Notoriously error-prone: Copy files to a mount point
>directory when the volume isn't actually mounted, then mount it,
>and--where did those files go? Not only are they on the wrong volume,
>but you can't even access them until you dismount the second volume
>again!
I believe that particular issue has been fixed, in some form. I
don't remember the details, though.
Note that hard links have to be used carefully, and I for one
would rather use soft ones.
>Verdict: Incompletely thought-out idea. How come the Linux folks are so
>focused on being so faithful to UNIX, when they could be *fixing* some
>of those long-standing, well-known UNIX problems?
>
>c) Per-volume filespecs (MacOS, though I think it was originally
>invented on the original UCSD p-system).
>Pros: System-independent file specifications. A reference to a file on a
>removable/hot-pluggable volume doesn't depend on the location of its
>"mount point", drive ordering or any other system-specific
>configuration: The same reference will work on any other machine
>mounting the same volume. The reference can also be used to
>automatically request the mounting of the correct volume on demand.
>Cons: Can't think of any.
Actually, one con is an indefinite wait, as the user is required to
respond to the requester if the volume name isn't known to the computer
(and it assumes a floppy needs insertion). It also has problems if
the volume name isn't what the program expects (the same is true of
Amiga and NT), and the volume name cannot be changed without
renaming the volume on the volume itself.
>
>Verdict: MacOS-style file specifications definitely seem to be the way
>to go in the next computing millennium. Why are other systems still
>using such primitive ways of doing things?
I'd have to agree, for the most part.
d) Per-volume filespecs and logical names with interactive volume mount wait
(AmigaDOS, ca. 1983 or 1984)
Pros: A system can locate its base install anywhere -- could be floppy,
could be logical name; no patching required for config files.
The program will not fail.
Cons: The program can go into indefinite wait, as the user is required
to properly respond to the "Please insert" requester before it can
continue. (Note that this is overridable, but then the open will
fail.)
The OS appears to be able to handle a volume name that is the same as
a logical name, but it's not immediately clear which one takes precedence.
Verdict: Not bad, actually, except for the wait.
e) Per-node filespecs and mountable shares with semi-proprietary protocol
(NT/SMB)
Pros: The data volume need not be local. Mounts are automatic.
Cons: Not horribly flexible. If a node changes name, the path has to
change too; shares are nameable, which is a plus, but go with the node.
Proprietary protocol, which means vendors are locked out.
Verdict: It works, but not well.
--
[EMAIL PROTECTED] -- insert random misquote here
------------------------------
From: [EMAIL PROTECTED] (Robert J Carter)
Subject: Re: democracy?
Crossposted-To: comp.os.linux.misc
Date: Sat, 10 Jun 2000 16:22:21 GMT
In article <[EMAIL PROTECTED]>,
Smitty <[EMAIL PROTECTED]> writes:
> Mark Wilden wrote:
>
>> Salvador Peralta wrote:
>> >
>> > let's remember that the United States is not now, nor has it ever been a
>> > democracy.
>>
>> Yes it is. It's a representative democracy. The people do rule, through
>> their elected officials (in theory, at least).
>
> You are misinformed on that point, Mark. Please refer to the U.S.
> Constitution and the legal definitions of republic and democracy.
> Smitty
>
>
I think it it YOU who are being misinformed. Saying a republic is
cannot be a democracy is like saying an orange can't be a fruit. They
are not mutually exclusive.
--
Robert J Carter at Oghma dot on dot ca
Use My initials to reach me via e-mail
------------------------------
From: [EMAIL PROTECTED] (abraxas)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.advocacy,comp.unix.advocacy
Subject: Re: Dealing with filesystem volumes
Date: 10 Jun 2000 16:40:45 GMT
In comp.os.linux.advocacy Cihl <[EMAIL PROTECTED]> wrote:
> And another thing, for the umpteenth time:
> I AM NOT GOING TO BUY A NEW COMPUTER, AND CERTAINLY NOT A FSCKING
> MAC!!!
As odd as it sounds, ive got an fscking mac sitting right next to my desk
at this very moment. It shouldnt take too long though, is only a 9gig drive
and its full of mp3s. Yay linuxPPC.
=====yttrx
------------------------------
From: "KLH" <[EMAIL PROTECTED]>
Subject: Re: Open Source Programmers Demonstrate Incompetence
Date: Sat, 10 Jun 2000 09:43:06 -0700
[snip]
There is bad code out there. But the theory is that someone will notice the
bad code and substitute it with good code.
Perhaps this message would be more useful on one of the Sourceforge mailing
lists?
------------------------------
From: [EMAIL PROTECTED] (abraxas)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.advocacy,comp.unix.advocacy
Subject: Re: Dealing with filesystem volumes
Date: 10 Jun 2000 16:52:28 GMT
In comp.os.linux.advocacy Sam Morris <[EMAIL PROTECTED]> wrote:
> Per-volume systems do no such thing. If I add another disk to my Mac, each
> formatted partition appears on the desktop and doesn't screw up everything
> else that came before.
After you format the partition of course.
The same exact thing happens when I put a scsi drive in my linux box. I
put the drive in, turn it on, it gets recognized and im asked if id like
to format it.
> Example: Installing a new Hard Disk.
> On my Mac: Connect the cables up inside and turn it on (both IDE/SCSI).
> Initialise the disk when asked to do so by the MacOS. Do whatever I want
> with the new disk.
If its scsi, you still have to set ID correctly. You dont have to if its
IDE, because most IDE macs have only one drive bay.
> On my PC: Look in manual to discover jumper settings for Master/Slave (IDE
> only; I haven't tried to add a SCSI card to my PC).
Actually, that stuff is usually on a little sticker on the drive itself
somewehre.
> Connect up cables. Boot
> up. Discover that my CD is now E:, not D: and every damn thing that expects
> it's files to be on a CD now needs to be told differently.
AAaahhhoo...you mean WINDOWS. I thought you said 'PC'. :)
=====yttrx
------------------------------
Subject: Re: Canada invites Microsoft north
From: Mayor <[EMAIL PROTECTED]>
Crossposted-To:
comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.os.ms-windows.nt.advocacy
Date: Sat, 10 Jun 2000 09:50:51 -0700
In article <tinman-1006001048200001@dsl-64-34-84-
49.telocity.com>, [EMAIL PROTECTED] (tinman) wrote:
>Oh, no, you're not baiting me into this one, I'm not falling
into another
>tholenesque spiral this week, I've got gardening to do....
>
Your gardening is irrelevant.
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
------------------------------
** 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
******************************
