[LWN Logo]
[LWN.net]

Sections:
 Main page
 Linux in the news
 Security
 Kernel
 Distributions
 Ports
 Development
 Commerce
 Announcements
 Back page
All in one big page

See also: last week's Back page page.

Linux links of the week


Everything Linux has been redesigned and is worth a look. Lots of good documention, pointers, and other stuff. It is, however, heavily frame-based.

Over the last few months, announcements from large, commercial database companies have gotten a lot of attention in the Linux community. It's time to remember that there is also a thriving project that is producing an industrial-strength object-relational database as free software. PostgreSQL has produced some impressive accomplishments, and they're not stopping yet. Release 6.4 is currently in beta, and should be out shortly. This is a good project, and certainly deserving of a look.


October 22, 1998

   

 

Letters to the editor


Letters to the editor should be sent to editor@lwn.net. Preference will be given to letters which are short, to the point, and well written. If you want your email address "anti-spammed" in some way please be sure to let us know. We do not have a policy against anonymous letters, but we will be reluctant to include them.
 
   
Date: Fri, 16 Oct 1998 14:14:49 +0200
From: Andi Kleen <ak@muc.de>
To: lwn@lwn.net
Subject: Comments on "impressive list of XEmacs advantages"

Hallo,

You posted a link to Rose Romildo Malaqui's list in the LWN Distributions 
page on why redhat should
include XEmacs.  But the funny thing is that most of the cited advantages
are not relevant to Linux at all and others are outdated.

Here are some:

* Binaries are available for many common operating systems. 
	- RedHat couldn't care less.

* Some internationalization support (including full MULE support, if
  compiled with it.) 
	- GNU Emacs 20.x has the same, partly even more MULE support
	(although RedHat ships it with these features disabled)

* ToolTalk support.
	- Only on Solaris, not on Linux.

* Better Motif compliance.
	- ... When compiled with Motif, which RedHat doesn't and can't.

Also in my experience (I got burned a few times) XEmacs is more buggy 
than GNU Emacs, rms might be rather conservative on accepting features
but he does a good job on release stable code (although I must admit
that 20.3 could have been better).

Also GNU Emacs redisplay engine is much faster then XEmacs', both on
ttys (it does sophisticated screen drawing optimization which got dropped
from XEmacs) and on X11 (the variable width fonts and inline image support
slows it down a lot). 

Generally I think RedHat should not get into the "Debian trap" of putting
so many packages into their distribution that a uniform quality control of
complete releases becomes impossible. 

-Andi

   
Date: 19 Oct 98 05:27:11 PDT
From: Gopalakrishnan P <gopalji@netscape.net>
To: editor@lwn.net
Subject: Linux and Oracle

Hi,

Thanks for the excellent job, that you guys are doing at lwn.
Your recent editorial has prompted me to write this.

Of late Linux is getting more attention from all the corners,
and big companies like Oracle and IBM has jumped in to the
bandwagon with their products and support. But we shouldn't
mistake this for their love to Linux or open source software.
One common thing that all these companies share, is an anti
microsoft feeling at more or less different degrees. And the
increasing popularity of Linux has given them a nice opportunity
to use this platform for giving a lesson to microsoft.

While we need all the applications from Oracle and other major
companies to run on Linux, we should also see that Linux doesn't
become a tool in their hand to fullfill their ambitious goals alone.
There should be ample return to the Linux community, and to
the vast majority of users in terms of free software and
support for open source projects.

One more point. Now we have an industry strength open source
operating system. With GNOME showing excellent progress, and
KDE getting more and more fine tuned (apart from the Qt 
licensing problem) in the near future we will have a decent
GUI also in the open source way. This should satisfy the
needs of a small or medium business to run all their applications
based on Linux. But there is something missing. That is an
industry strength Data Base Management System. True that
from Oracle to Sybase everyone is supporting Linux today and
their products are excellent. But we do need to have a similar
product in the open source. In my opinion the one which can
become a candidate for this slot is Postgresql. It is already
has an object relational model. And of late, the implementation,
documentation etc. have improved a lot. Technically what is
lacking there is a scripting language (something similar to
Oracle's PL/SQL). Apart from that it needs a lot of support
from the open source and Linux community. Something like
GNOME or Apache project is enjoying today. With this Postgresql
can grow in to an industry strength DBMS and fill the
void felt in this area. Hope you guys at lwn can do
something about this!

Thanks for all the information that you bring up =

every week.

Wishing all of you at lwn the very best.

--Gopalakrishnan
   
Date: Tue, 20 Oct 1998 01:27:00 +1000 (EST)
From: Conrad Sanderson <conrad@hive.me.gu.edu.au>
To: editor@lwn.net
Subject: Linux and Intel: dealing with a different devil


So Intel has invested in Red Hat, and "we are all happy that such a
high profile company is investing in Linux".  Not so quick brother.
I've never liked Intel - be it because of the mediocrity of their 
processors (right from the start), their alliance with Microsoft,
or their own monopolistic and bullying tendencies.
And now I have found this:

http://www.igc.org/faceintel/

Every Linux user, particularly those guys at Red Hat, should have a look
at the above site, just so we know who we are dealing with.


Conrad Sanderson - Microelectronic Signal Processing Laboratory
Griffith University, Queensland, Australia
http://spl.me.gu.edu.au   [ under construction ]

   
Date: 16 Oct 1998 02:14:23 -0000
From: Eric Smith <eric@brouhaha.com>
To: editor@lwn.net
Subject: Microsoft claims of not having a monopoly.

Microsoft claims not to have a monopoly, and cites Linux as an example,
further claiming that it was developed by one person.  You refuted the
one person claim in your 15-Oct-1998 issue, but you didn't carry it far
enough.

Although I certainly don't want to downplay Linus Torvald's role in
leading the development of the Linux kernel, not even the kernel was
developed by a single individual.  And for an apples-to-apples
comparison, it is not realistic to compare the Linux kernel (by itself)
to Microsoft Windows, since Windows (of any flavor) is much more than a
kernel.  The contents of a typical Linux distribution have taken more
resources to develop than *ANY* one corporation can muster, even
Microsoft.

Citing Linux as evidence that "Market entry costs are very low and
profit opportunities vast in software platform technology" is thus
completely absurd.  There has not been (and will never be) an accounting
for the cost of development of Linux, but it has been astronomical.  The
fact that it was largely a volunteer effort does not in any way support
the notion that it was "low cost".  If anything, the Linux experience
demonstrates that even with thousands of engineers developing an
operating system, it still may not be possible to effectively compete
with Microsoft.

Of course, many of us in the Linux community would like to believe that
Linux will eventually offer effective competition for Microsoft, and we
are optimistic about it, but it will likely still take at least several
years for this to happen.

Sincerely,
Eric Smith

   
From: dblake@phy.ucsf.EDU (David T. Blake)
Date: Thu, 15 Oct 1998 08:05:30 -0700 (PDT)
Subject: Oracle/MS/Linux
To: editor@lwn.net

Editor,
	you seem to suggest a futuristic scenario in which 
Linux could become fragmented as part of a turf war in which
a major company decides to push its own Linux developemnt, and
fragments the developer team. 
	This, IMHO, is not a real possibility. Linux grew out
of freedom, and almost every developer values that. The amount of
labor it gets in exchange for this freedom is something that no
company, not even Microsoft, could compete with. 
	Besides, if Linus has established one thing, it is that
he has good intentions and is committed to doing what is in the 
best interests of Linux. And for that, he gets loyalty from
the other developers. I don't think anyone could buy the kernel
development away from Linus. 
	If anything could happen, it would be an outmarketing of 
something else compared to Linux. But Linux is spreading already
primarily by word of mouth - and the best marketing in the world
cannot compete with real world experience. 
	Linux should not, and in my opinion can not, be viewed as
something that could be used as someone's tool to be manipulated.
Linux will be linux. If that helps some company, it will help them.
If they like it, they can give back by funding kernel developers
(like Red Hat), or providing resources (like Cygnus for egcs), or
writing excellent drivers (like SuSE). 
	And even if the corporate world never embraces linux, it would
still be the best open source GPL kernel ever, and it would still be
the most stable full featured POSIX system ever, and it would still
be a great developer environment, and it would still run the world's
web servers.

---
Dave Blake
dblake@phy.ucsf.edu

 

 

 
Eklektix, Inc. Linux powered! Copyright © 1998 Eklektix, Inc., all rights reserved
Linux ® is a registered trademark of Linus Torvalds