Linux-Advocacy Digest #124, Volume #32           Sun, 11 Feb 01 16:13:03 EST

Contents:
  Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (The Ghost In The Machine)
  Re: How does this look? (J Sloan)
  Re: Answer this if you can... ("Edward Rosten")
  Re: Another Pete Goodwin "Oopsie"! (Mig)
  Re: Does Code Decay (Karel Jansens)
  Re: Micro-Sinux Distro? (Karel Jansens)
  Re: Good article (mlw)
  Re: Another Linux "Oopsie"! ([EMAIL PROTECTED])
  Re: Linux fails to deliver on the hype (sfcybear)
  Re: Good article (J Sloan)
  Re: Another Linux "Oopsie"! ("Edward Rosten")
  Re: Another Linux "Oopsie"! ([EMAIL PROTECTED])
  Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (The Ghost In The Machine)

----------------------------------------------------------------------------

From: [EMAIL PROTECTED] (The Ghost In The Machine)
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: Sun, 11 Feb 2001 20:33:30 GMT

In alt.destroy.microsoft, Jim Richardson
<[EMAIL PROTECTED]>
 wrote
on Sat, 3 Feb 2001 13:45:59 -0800
<[EMAIL PROTECTED]>:
>On Mon, 29 Jan 2001 01:23:42 GMT, 
> Giuliano Colla, in the persona of <[EMAIL PROTECTED]>,
> brought forth the following words...:
>
>>Jim Richardson wrote:
>>> 
>>> On 28 Jan 2001 07:44:44 -0500,
>>>  Norman D. Megill, in the persona of <[EMAIL PROTECTED]>,
>>>  brought forth the following words...:
>>> 
>>> >In article <dGKc6.19391$[EMAIL PROTECTED]>,
>>> >Chad Myers <[EMAIL PROTECTED]> wrote:
>>[...]
>>> >
>>> >> Besides, Win2K has NTFS5 which doesn't have this problem anyhow.
>>> >
>>> >If so, let's hope they have better luck than these people when they
>>> >try to use Win2K for their enterprise application:
>>> >
>>> >http://www.theregister.co.uk/content/6/16075.html
>>> >
>>> >--Norm
>>> >
>>> 
>>> I did find this story amusing, wonder if M$ will fly some techs and muscle out
>>> to "convince" Delphi to change their minds...
>>> 
>>
>>A quick glance to:
>>
>>http://uptime.netcraft.com/up/graph/?host=www.delphi.com
>>
>>tells that if they did, they didn't succeed, for the moment.
>>
>>It also tells that with Win2k they weren't able to keep their site up
>>for more than half a day.
>
>
>
>Interestingly enough, netcraft still says that delphi is using WinNT/W2K,

So does a simple probe:

    $ telnet www.delphi.com www
    Trying 64.94.242.130...
    Connected to www.delphi.com.
    Escape character is '^]'.
    HEAD / HTTP/1.1
    Host: www.delphi.com
    
    HTTP/1.1 302 Object moved
    Server: Microsoft-IIS/4.0
    Date: Sun, 11 Feb 2001 20:26:51 GMT
    Set-Cookie: ISP=DEI|10028265; path=/; domain=.delphi.com; \
    expires=11-Feb-2004 07:22:08 GMT;
    Location: http://www.delphi.com/dir-delphi
    Content-Length: 153
    Content-Type: text/html
    Cache-control: private

It gets stupider.  I don't know if RFC2068 requires that one put a
trailing slash on the request or not in a Location: field, but it's
clear that the Location: field forgot said slash, and, when accessed,
I get another Object moved message:

    $ telnet www.delphi.com www
    Trying 64.94.242.130...
    Connected to www.delphi.com.
    Escape character is '^]'.
    HEAD /dir-delphi HTTP/1.1
    Host: www.delphi.com
    
    HTTP/1.1 302 Object Moved
    Location: http://www.delphi.com/dir-delphi/
    Server: Microsoft-IIS/4.0
    Content-Type: text/html
    Set-Cookie: ISP=DEI|10028306; path=/; domain=.delphi.com; \
    expires=11-Feb-2004 07:22:08 GMT;
    Content-Length: 149
    
    <head><title>Document Moved</title></head>
    <body><h1>Object Moved</h1>This document may be found <a \
    HREF="http://www.delphi.com/dir-delphi/">here</a>

Dunno if that's misconfig or what, though.  At least they got it
right the second time.

>but the graph is a little behind the times, and queso and other tools put
>www.delphi.com as IBM AIX 4. So netcraft is a little out of date here.
> As you point out, the uptimes for delphi on W2k were abysmal.

I can't say I know, myself.  It helped my employer when we switched
from NT4 to Win2k, that much I can tell you.  By how much, I can't,
since I don't know.

[.sigsnip]

-- 
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191       6d:23h:59m actually running Linux.
                    We are all naked underneath our clothes.

------------------------------

From: J Sloan <[EMAIL PROTECTED]>
Subject: Re: How does this look?
Date: Sun, 11 Feb 2001 20:34:31 GMT

Pedro Duato wrote:

> > Redhat Linux 7.0
>
>    I would not use this one ... now there are 200MB patches out.

I see about 80 MB - in any event the update is painless
and, the 2.4 kernel drops in.

I'm getting ready to try out Red Hat 7.1 and Suse 7.1

First distros to come with 2.4 kernel

jjs



------------------------------

From: "Edward Rosten" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x
Subject: Re: Answer this if you can...
Date: Sun, 11 Feb 2001 20:34:42 +0000

According to the troll FAQ, several trolls banding together can turn an
average troll in to a great troll. That was a crap troll, so I wouldn't
bother if I were you.

Find a better one.

-Ed

-- 
Did you know that the reason that windows steam up in cold|Edward Rosten
weather is because of all the fish in the atmosphere?     |u98ejr
        - The Hackenthorpe Book of lies                   |@
                                                          |eng.ox.ac.uk

------------------------------

From: Mig <[EMAIL PROTECTED]>
Subject: Re: Another Pete Goodwin "Oopsie"!
Date: Sun, 11 Feb 2001 21:29:37 +0100

Pete Goodwin wrote:

> It seems to me Linux has a whole slew of different ways of doing things.
> I've already heard how these are "trivial" and "inconsequential", however,
> when I setup a printer, I'd have thought it was setup that way everywhere.
> What I find in the case of Linux is that this is not true.

BS.. you setup a que and configure a printer for it. 
That cat be that hard to understand for a pro like you or is it? 

-- 
Cheers

------------------------------

From: Karel Jansens <[EMAIL PROTECTED]>
Subject: Re: Does Code Decay
Date: Sun, 11 Feb 2001 13:21:31 +0100

Matthias Warkus wrote:
> 
> It was the Fri, 09 Feb 2001 16:40:58 -0500...
> ...and Aaron Kulkis <[EMAIL PROTECTED]> wrote:
> [plutonium fission bomb]
> > You don't even need that...  We're talking about something the size of
> > a softball.  You can merely make a sphere with a cylinder hollowed out
> > of it...and a matching cylinder which drops (by gravity) into the
> > hollowed-out space.  Then make a time-delay device that gives you a
> > two-hour delay.
> 
> ROTFLMAO.
> 
> You think that letting a piece of plutonium drop into a subcritical
> mass *by gravity* will give you an explosive chain reaction? Pretty
> naive.
> 
There will be _some_ fission, but the result is what is technically
<G> known as a "nuclear fire-cracker": a bang which spreads your
precious plutonium over the neighbourhood. It's nasty, agreed, but you
could have done sooo much more with that plutonium.

Not to mention the "interesting" side-effects of casting plutonium
(Hint: Don't try this in your backyard shed), which tend to reduce the
life-expectancy of your average, upwardly mobile terrorist to
something like _days_. 
-- 
Regards,

Karel Jansens
==============================
"Go go gadget linux." Zoomm!
==============================



------------------------------

From: Karel Jansens <[EMAIL PROTECTED]>
Subject: Re: Micro-Sinux Distro?
Date: Sun, 11 Feb 2001 13:53:52 +0100

mlw wrote:
> 
> 
> What scares me more than an M$ Linux, is a set of "Microsoft Extensions" for
> Linux which allow a port of M$ office. Microsoft would surely promote the
> extensions as a way of porting Windows applications to Linux. If Microsoft can
> get a toe hold into the Linux market with this strategy, they could work a
> situation where their illegal monopoly power in Windows, makes them an instant
> giant in the commercial Linux environments.
> 

Wouldn't that be a piece of cake for Microsoft, if they really wanted
to? Just bundle a taylored and trimmed copy of something like Win4Lin
with the Office box and sell the stuff. I'm sure they can take home a
licencing deal with whoever it is who makes Win4Lin.

What am I missing?


-- 
Regards,

Karel Jansens
==============================
"Go go gadget linux." Zoomm!
==============================



------------------------------

From: mlw <[EMAIL PROTECTED]>
Subject: Re: Good article
Date: Sun, 11 Feb 2001 15:41:56 -0500

Pedro Duato wrote:
>    NT have threading, it seems to be quicker than fork() even on Unix
> OS's, and it will gain popularity as Java (I hate it) become more and more
> important programming language.
>    I have heard IIS is multithreaded, which does avoid it to fork(), but I
> can't
> certify this personally.

Many UNIX like operating systems, including Linux have real kernel scheduled
threads, and yes creating a thread is far more efficient than calling fork(),
but Apache does not call fork on every connection, it only calls fork() to
increase the number of processes answering queries. Apache processes hang
around and answer more than one HTTP request.

Thread vs fork() has very little to do with a program's ability to be
responsive. How a it is designed is critical.

-- 
http://www.mohawksoft.com

------------------------------

From: [EMAIL PROTECTED]
Subject: Re: Another Linux "Oopsie"!
Reply-To: [EMAIL PROTECTED]
Date: 12 Feb 2001 07:05:23 +1100

Pete Goodwin <[EMAIL PROTECTED]> writes:

>Under Linux I selected the Epson 640 Color stylus driver when I installed 
>Linux Mandrake. What do I find? The Gimp doesn't use this, it defaults to 
>postscript instead.

Yes. That's how it works. The driver will then pick up the Postscript and
convert it to something your printer understands.

>The Gimp picks postscript by default. I had to override that

In other words, you sabotaged the whole system. The printer driver *expects*
to get Postscript. What you sent it was probably interpreted as ASCII text,
translated from ASCII to Postscript by the generic (input) printer filter, 
and then translated to your printer's command language by the driver (i.e.
the output filter).

>and get it to select the Epson printer instead.

And you then sent the Epson's command language to a queue that expects
Postscript. And it didn't work. Big surprise!

>Image that! I install a driver at installation time, and I install it
>_again_ per an application. Why kind of nonsense is this?!?

Nonsense you shouldn't be doing. Simple as that.

Yes, the GIMP has its own drivers for some printers. If you know how to,
you can circumvent the OS-wide driver, and send stuff directly from the
GIMP to your printer. It might even look better if you do that, because
the app-specific driver might be able to use info that gets lost when
first translating into Postscript.
However, if you *don't* know how to do this, you shouldn't be doing it.

>THERE IS ONLY ONE PRINT QUEUE, IT'S CALLED "lpr".

>The Gimp defaults to Postscript output (for some reason best known to The 
>Gimp).

Then maybe you really should live by your own words --- The Gimp knows best
why it defaults to the things it defaults to, and maybe you should at least
consider the possibility that defaulting to those settings is not a mistake,
but rather how things work. As is the case for this "problem" of yours.

Bernie
-- 
No man's life, liberty or property are safe while the
    legislature is in session
unidentified NY State Surrogate Court Judge

------------------------------

From: sfcybear <[EMAIL PROTECTED]>
Subject: Re: Linux fails to deliver on the hype
Date: Sun, 11 Feb 2001 20:47:28 GMT

In article <[EMAIL PROTECTED]>,
  imekon@$$$REMOVE$$$.freeuk.com (Pete Goodwin) wrote:
> sfcybear <[EMAIL PROTECTED]> wrote in <963q7n$872$[EMAIL PROTECTED]>:
>
> >And you wonder why no on believes what you say.
>
> I thought that was a given with some people.

Not with me, you earned your standing as unbievable with the trollish
way you post.


>
> Anyway, I'm human, I make mistakes. I apologise.
>
> --
> Pete Goodwin
> ---
> On that unstable much loved system known as Windows 98 SE.
>
>


Sent via Deja.com
http://www.deja.com/

------------------------------

From: J Sloan <[EMAIL PROTECTED]>
Subject: Re: Good article
Date: Sun, 11 Feb 2001 20:55:30 GMT

Pedro Duato wrote:

> Why ? Software is just software, and XOR EAX,EAX is the same
> be it open source or be it $10000.

That would seem to favor open source.

> Like it or not, Windows NT/2000 is a real OS, and if open source
> code does not perform as well as under FreeBSD, GNU/Linux or other
> like-unix OS's is just because open source has been designed with
> that unixism in mind, and in a certain way, it does prove it is some kind
> of OS dependent, or at least, optimized for if you don't want to say
> dependent.

True, apache was designed for Unix, but porting it to all
other signiificant and not so significant OSes was a real
indication of the robust, portable design.

jjs



------------------------------

From: "Edward Rosten" <[EMAIL PROTECTED]>
Subject: Re: Another Linux "Oopsie"!
Date: Sun, 11 Feb 2001 21:01:05 +0000

In article <SUBh6.8626$[EMAIL PROTECTED]>, "Pete Goodwin"
<[EMAIL PROTECTED]> wrote:

> Edward Rosten wrote:
> 
>> Unless the hardware is very new, then RH6.2 should do. Even if the HW
>> is new, it's no problem to upgrade GS, the Kernel and XFree.
> 
> I checked at their website and they don't have the following: Linux 2.4,
>  XFree86 4.0.x and ReiserFS. SuSE 7.1 (out Monday) apparently does.

Use Suse then.

-Ed
 


-- 
Did you know that the reason that windows steam up in cold|Edward Rosten
weather is because of all the fish in the atmosphere?     |u98ejr
        - The Hackenthorpe Book of lies                   |@
                                                          |eng.ox.ac.uk

------------------------------

From: [EMAIL PROTECTED]
Subject: Re: Another Linux "Oopsie"!
Date: 11 Feb 2001 20:54:05 GMT

Pete Goodwin <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] wrote:

>> Mart van de Wege <[EMAIL PROTECTED]> wrote:
>> > I plead guilty. I have confirmed in another thread that in my
>> > personal opinion (confirmed by experience) Mandrake 7.2 was
>> > rushed to the market.
>> 
>> One...

> http://x65.deja.com/getdoc.xp?AN=726298021&CONTEXT=981813346.757202969&hitnum=2
> http://x65.deja.com/getdoc.xp?AN=721452552&CONTEXT=981813346.757202969&hitnum=161
> http://x65.deja.com/getdoc.xp?AN=721459108&CONTEXT=981813346.757202969&hitnum=162

> Two.

> There you are, another one (the above are three statements by the same 
> author). You sure you want more? You sure you still want to call me a liar?

MORE!




=====.

------------------------------

From: [EMAIL PROTECTED] (The Ghost In The Machine)
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: Sun, 11 Feb 2001 21:05:37 GMT

In alt.destroy.microsoft, Ayende Rahien
<[EMAIL PROTECTED]>
 wrote
on Sun, 4 Feb 2001 08:26:45 +0200
<95itd7$fda$[EMAIL PROTECTED]>:
>
>"The Ghost In The Machine" <[EMAIL PROTECTED]> wrote in
>message news:[EMAIL PROTECTED]...
>> In comp.os.linux.advocacy, Aaron R. Kulkis
>> <[EMAIL PROTECTED]>
>>  wrote
>> on Thu, 01 Feb 2001 21:44:43 -0500
>> <[EMAIL PROTECTED]>:
>> >Champ Clark III wrote:
>> >>
>> >> In article <95bh0f$t75$[EMAIL PROTECTED]>, Ayende Rahien wrote:
>> >> >
>> >> >"J Sloan" <[EMAIL PROTECTED]> wrote in message
>> >> >news:[EMAIL PROTECTED]...
>> >> >> Ayende Rahien wrote:
>> >> >>
>> >> >> > "J Sloan" <[EMAIL PROTECTED]> wrote in message
>> >> >> > news:[EMAIL PROTECTED]...
>> >> >> >
>> >> >> > > Is it true that windows 2000 finally got filesystem quotas
>> >> >> > > somewhat similar to what Linux has had for years?
>> >> >> >
>> >> >> > Yes.
>> >> >> > Is it true that Linux finally got the SMP support that NT
>> >> >> > had for years?
>> >> >>
>> >> >> Linux has had smp support since version 1.1.31.
>> >> >
>> >> >And it was *bad*.
>> >> >
>> >> >> That was ~1995.
>> >> >
>> >> >NT had it since 3.1 (from the start, that is).
>> >>
>> >>         You still dodged something here.  That was quota's that
>> >> got this little thing going.  I'll bet that quota'ing is in more
>> >> use then say,  hrmm..  Mulit-CPU's?  Hell,  Novell has quota's!
>> >> VMS has had quotas for years! I can't even think of a *nix that
>> >> doesn't have a quota'ing system for it......  What took so long?
>> >
>> >Microsoft--relearning mid-20th centuryt technology...in  the 21st.
>>
>> Not only that, but making it "usable" for the masses by slapping
>> on silly icons, pulldown menus that slowly vanish beautifully
>> and/or animate from the pointer as the user selects a scascade item,
>> windows with scrollbars, gadgets, rollover labels that change
>> color as the mouse rolls over them (wow, psycho, man), and built-in
>> richly-formatted help text files that tell one the bit that he
>> knew already ("yes, I KNOW that's a toggle button with a label,
>> you moronic program!").
>
>You would *love* Mac OS X, then.
>When you minimize something, it shrink and vanish like a genie. And there
>are a lot of other eye-candy there to make your head hurt.

I'm aquiver with anticipation.  Not.

Mind you, I have seen iMac demoing itself.  Slick, but glitzy/kitschy.
Are they doing a Website or promoting a TV commercial, complete with
fast pans and rollovers?  I like the notion of rollovers to a point
(it shows the window system/OS is still alive) but sheesh.

>
>> Except that the RTF files are not standard HTML, TeX, or PDF,
>> the vanishing pulldown menus and rollovers are useless gewgawery
>> (although a well-designed pulldown can help in documenting),
>> and the silly icons are just silly -- what *does* that floppy
>> mean in Word, for example?  How does one deduce that it means
>> "save to disk"?  How stupid is that??
>
>Actually, you are supposed to learn it *once*, and from them on, you
>recognize it, because it's used almost everywhere.

Yeah, I know.  But floppies are just so '80's. :-)  Not sure what
other icon I would use, myself -- an arrow from a small box with a
grooved front (reminiscent of an air grille for hard drives) could
be used for load, and an arrow to a small box for save, but that
would be a bigger icon to draw, perhaps.

Or I might punt on the issue and just have "Load" and "Save".  Icons
aren't required for everything, after all!  (Especially if someone's
supposed to be able to read English anyway.)

>
>> I'm still holding out for soft links.  Has Microsoft scheduled them yet?
>
>Describe what you mean when you say soft links (my unix terminology is in
>another language, and not very good to begin with).
>I think that they did.

A soft link (also called a symbolic link) is an entry in the filesystem
that points to another entry, which can either be a directory, a file,
a device, or even another soft link.

For example, on my Debian system:

$ ls -l /usr/lib/X11/XF86Config 
lrwxrwxrwx    1 root     root           19 Jan 22 14:25 /usr/lib/X11/XF86Config
-> /etc/X11/XF86Config

This is not a hard link; /usr can be on a different mounted volume and it
would still work.

Soft links can specify absolute or relative pathnames, and are transparent
to open(), exec(), and stat() ( lstat() exists for that very reason).
Not sure what rename() will do with them.

Soft links can fail; the object to which they point may not exist.

Soft links differ from hard links, in that a hard link is another entry
to the same inode -- an inode is an internal identifer in Unix and Linux
to the actual file or directory object.  Because inodes are volume-specific
(e.g., the standard inode for the directory root of an ext2 volume, and
for many other Unix filesystems, is the number '2' (0 and 1 probably being
used for other purposes, presumably superblock and badblocks) -- check out
the following result, for example (-i displays the inode number):

$ ls -id / /h1 /h2 /h4 /h5 /h6
      2 /        2 /h1        2 /h2        2 /h4        2 /h5        2 /h6

:-)

) hard links cannot go across volumes.

Hard links also create additional links (st_nlink in the stat()
structure); a file is not actually deleted until its link count
goes to zero.  Directories have to be treated specially since they
link to themselves (the '.' points to the same inode) and their
children link to them.  Old Unix systems did not allow for
deletion of directories in a user-level program; one had to use
'rmdir' or 'rm -r', which was set-uid.

Soft links differ from .LNK files in Explorer in that .LNK files are
actual files, with some data in them that Explorer reads and interprets
as a soft link.  .LNK files, therefore, are Explorer-specific -- although
other programs might be able to parse them using DLL calls, I don't know.
But it's not as transparent, by any means.

HTH.

-- 
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191       6d:00h:14m actually running Linux.
                    >>> Make Signatures Fast! <<<

------------------------------


** 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
******************************

Reply via email to