Linux/UNIX bashing from Microsoft?

L. Cranswick L.M.D.Cranswick at dl.ac.uk
Sat Oct 30 19:19:46 BST 1999


> > Has anybody read that one already?
> > 
> > http://www.microsoft.com/NTServer/nts/news/msnw/linuxmyths.asp
> 
> Yes! Fascinating isn't it? ;-)

A question from me about this:

  Why is FreeBSD so moribund(sp?) when it comes to
promoting what is a very good server OS system and
what in some areas looks like a superior client
system over Linux?  FreeBSD does not seem to get
much of a mention from MS or the general public
interested in high quality PC OS's yet it seems to be
the choice of people running massively hit servers.

I don't remember seeing any announcements about
FreeBSD 3.3 coming out on any of the newsgroups
I would expect them to (even if just informal
announcements) such as the unix-webservers newsgroup
etc?  Or is it as a friendly implied, a person must
prove themself to be on a higher psychic plane first
by naturally knowing to go to FreeBSD?  Linux is 
definitely more in your face.  Though especially
with the latest Redhat 6.1 - Linux looks like it
is going into buggy hypeware space - unlike FreeBSD 
3.3.  (have just spend the last few weeks 
on tutorials for safely getting Redhat 6.1 
running in a safe state and getting around 
the install bugs)

----

One recent thing that has definitely flipped me on
the apparent superiority of the people who wrote
the FreeBSD kernel over those doing the Linux kernel
is playing around with a friend on some GPL'd software
of ours by having a debug mode with a deliberate
memory leak (which is not uncommon in user-land 
programs).  Running this as a plain user program,
this crashes/makes unusable: Win95, WinNT 4, 
All types of Linux running X and in command
line mode, Solaris 7, IRIX 6.5.x.  However, FreeBSD
3.3 (and Digital UNIX 4.0d for Dec Alpha) seem to
be able to carry on regardless (and thus give an
operator a chance working out what process is running
amok if they do notice a performance problem)
A good OS kernel should not allow a user-land 
program to make it un-usable.

Are there any ideas why FreeBSD is not susceptable
to this type of thing where as nearly all the other
operating system tried are?

Page with the software and debug info is at:
  http://www.ccp14.ac.uk/people/lachlan/borrocks/#tech
  http://www.ccp14.ac.uk/people/lachlan/borrocks/borrocks95.tar.gz
  http://www.ccp14.ac.uk/people/lachlan/borrocks/borrocks95.zip


Lachlan.

PS: Pointing this out on some advocacy newsgroups got
the main response that anyone who says anything
negative about Linux kernel (even if the same thing
applies to the Microsoft Kernel) is a troll of MS.
(I never claimed to be one of the mature marketting
types!  After all, it looks like advocacy groups
are made for trolls to use :-) )

-- 
Lachlan M. D. Cranswick

Collaborative Computational Project No 14 (CCP14)
    for Single Crystal and Powder Diffraction
Daresbury Laboratory, Warrington, WA4 4AD U.K
Tel: +44-1925-603703  Fax: +44-1925-603124
E-mail: l.cranswick at dl.ac.uk  Ext: 3703  Room C14
                           http://www.ccp14.ac.uk






More information about the Ukfreebsd mailing list