Linux-Development-Sys Digest #899, Volume #7 Wed, 24 May 00 09:13:17 EDT
Contents:
Call BIOS from device diver ? (Robert Kaiser)
Oops in interrupt handlers / idle task ([EMAIL PROTECTED])
Oops message in interrupt handlers / idle task ([EMAIL PROTECTED])
Re: DEB: minimum install for compile ([EMAIL PROTECTED])
Re: Need ideas for university funded project for linux (Maciej Golebiewski)
Re: Need ideas for university funded project for linux (David T. Blake)
Re: Call BIOS from device diver ? (Josef Moellers)
Re: shm broken in recent (.99pre) kernels? (Aaltonen Timo)
adduser-shadow-1.6 and RedHat 6.2 ("Davy")
Re: high resolution timer? (Maciej Golebiewski)
970Cxi - HP refuses to help customer (Nicolas Eymerich)
Call stack in C (Joe Ceklosky)
shared libs (.so's) with dependencies (Kai =?iso-8859-1?Q?Hergenr=F6ther?=)
Re: Need ideas for university funded project for linux (Someone)
Re: Need ideas for university funded project for linux (Someone)
Linux Bootable CD (Jean-Claude MEILLAND)
Re: Need ideas for university funded project for linux (David Steuber)
Re: Need ideas for university funded project for linux (David Steuber)
Re: Need ideas for university funded project for linux (David Steuber)
Re: Need ideas for university funded project for linux (David Steuber)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Robert Kaiser)
Subject: Call BIOS from device diver ?
Date: 24 May 2000 08:26:12 GMT
Hi all,
is it possible to call a PC's BIOS from the kernel (e.g.
a device driver) ? Specifically, I would like to call int #13
for Disk I/O. I'm aware that this would be terribly inefficient
compared to a native Linux device driver, but would it be possible
at all ? If it is, can anyone point me to an example in the kernel
code ?
Thanks in advance
Rob
================================================================
Robert Kaiser email: rkaiser AT sysgo DOT de
Sysgo RTS GmbH
Klein-Winternheim / Germany
------------------------------
From: [EMAIL PROTECTED]
Subject: Oops in interrupt handlers / idle task
Date: Wed, 24 May 2000 09:28:39 GMT
Hello. Having read Rubini's book, the bit about using ksymoops, it
seems it should be a simple matter of feeding the oops text into the
program.
This works for usermode processes that fault in-kernel, as you retain
use of the shell, but when my interrupt handler or its bottom half
faults, nothing is recorded to syslogd, the kernel panics (although my
interrupt handler may still be invoked it seems) and obviously you lose
the shell.
Is it time for pen, paper and patience?
alan
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Oops message in interrupt handlers / idle task
Date: Wed, 24 May 2000 09:26:22 GMT
Hello. Having read Rubini's book, the bit about
using ksymoops, it seems it should be a simple
matter of feeding the oops text into the program.
This works for usermode processes that fault in-
kernel, as you retain use of the shell, but when
my interrupt handler or its bottom half faults,
nothing is recorded to syslogd, the kernel panics
(although my interrupt handler may still be
invoked it seems) and obviously you lose the
shell.
Is it time for pen, paper and patience?
alan
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: DEB: minimum install for compile
Date: Wed, 24 May 2000 09:35:41 GMT
> Hi anyone know what minimum .deb install + order of install
> I need to compile on debian 2.2. At the moment I got errors
> compiling c, such as ld cannot find crt0.o or something like that..
> I guess I'm missing some library .deb
>
Problem seems to be fixed by --force-depend installing
libstdc++2.10-dev_2.95.2-10 and g++_2.95.2-10 or gcc_2.95.2-10
(can't remember which exactly.. will check ~/.history)
It seems both packages depend on each other?
L8R
Bill
mailto:[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Maciej Golebiewski <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
Date: Wed, 24 May 2000 12:58:20 +0200
David Steuber wrote:
> Do people really have trouble with ./configure, make, make install?
> It has _never_ been a problem for me. Maybe I am just lucky. Even
> though I changed my compiler, libc, and libtools.
How often the makefiles provide the "uninstall" target, too? Or at least
a script for un-installing? Otherwise, over the time and number of upgrades
to newer versions comprised of different files, you're accumulating
"abandoned"
files.
RPM is not perfect but it is quite OK. It's just that the guys creating rpm
not always can get the dependencies right. Personally one of my favourite
query options in rpm is -q -f to instantly get the name of the package
"owning" a specific file. I love it.
Maciej
------------------------------
From: [EMAIL PROTECTED] (David T. Blake)
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
Date: 23 May 2000 13:07:01 GMT
Reply-To: [EMAIL PROTECTED]
David Steuber <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] (David T. Blake) writes:
> ' Section 3b) (on modifications to QT)
> ' When modifications to the Software are released under this
> ' license, a non-exclusive royalty-free right is granted to the
> ' initial developer of the Software to distribute your modification.
>
> The right is non-exlusive. That means everyone can get that right. I
> think TrollTech is just trying to prevent forking of the Qt library
> here.
No, they are ensuring they can continue a revenue stream based
on contributions from outside the company. They will take your
modification and include it in QT Pro.
> As I said previously, if you don't like the Qt license, you can
> create your own library. There is no one to stop you. You can
> also use one of the other available libraries.
I was not arguing I should create a library. I was not arguing
against QTs right to use whatever license they like. I was
arguing that people should think twice before referring to QT
licensing as substantially free or "open source". The right to
fork is absent, the right not to have your contributions included
in proprietary works (such as QT Pro) is gone, and QT gets a copy
of EVERYTHING that even links to their code, even if it is not
publicly available.
--
Dave Blake
[EMAIL PROTECTED]
------------------------------
From: Josef Moellers <[EMAIL PROTECTED]>
Subject: Re: Call BIOS from device diver ?
Date: Wed, 24 May 2000 13:19:42 +0200
Robert Kaiser wrote:
> =
> Hi all,
> =
> is it possible to call a PC's BIOS from the kernel (e.g.
> a device driver) ? Specifically, I would like to call int #13
> for Disk I/O. I'm aware that this would be terribly inefficient
> compared to a native Linux device driver, but would it be possible
> at all ? If it is, can anyone point me to an example in the kernel
> code ?
No, the PC BIOS runs in real mode, while the Linux kernel runs in
protected mode.
-- =
Josef M=F6llers
Fujitsu Siemens Computers
SHV Server DS 1
------------------------------
From: Aaltonen Timo <[EMAIL PROTECTED]>
Crossposted-To: linux.dev.kernel
Subject: Re: shm broken in recent (.99pre) kernels?
Date: Wed, 24 May 2000 14:10:26 +0000
Aaltonen Timo wrote:
>
> Hi!
>
> I have a nasty problem in my box (Debian woody, 2.3.99pre8,
> Pentium200MMX, 64MB). After a clean boot everything works fine, but
> after awhile disk access gets very slow. Hdparm gives timing
> buffer-cache read -result of appr. 1.5MB/s...
>
> It is most noticeable when I copy large chunks of data from a drive to
> another. I�ve heard some people having the same problem... Copying speed
> slows down to 200kB/s, despite the drives are large and fast (Seagate
> 17GB and a brand new Maxtor 40GB).
Hooray! Problem solved with pre9!
Still I wonder why my 17GB Seagate (ST317242A) is relatively slow, 4MB/s
compared to my old 4GB IBM which scored 9.5MB/s with hdparm. And yes,
dma was on..
------------------------------
From: "Davy" <[EMAIL PROTECTED]>
Subject: adduser-shadow-1.6 and RedHat 6.2
Date: Wed, 24 May 2000 11:35:10 GMT
Hi,
it's impossible for me to make a workable adduser from adduser-shadow-1.6
(ftp://ftp.lip6.fr/pub12/linux/sunsite/system/admin/accounts/adduser-shadow-
1.6.tar.gz) on my RedHat 6.2 with glibc 2.1.3, egcs 1.1.2 (same with gcc
2.7.2.3).
Compilation works without any prb, but when i run adduser, just after type a
user name and the "getpwnam" function is called for check it in /etc/passwd,
i've a core (Illegal Instruction). Prb is the same with getpwuid. Does
anyone known a bug with getpwnam ?
PS : If i change getpwnam(buf) in source code by getpwnam("davy"), the
program works. This source code, without, any change, works fine on a
Slackware 4.6.
Thanks,
Davy.
------------------------------
From: Maciej Golebiewski <[EMAIL PROTECTED]>
Subject: Re: high resolution timer?
Date: Wed, 24 May 2000 13:35:01 +0200
Robert Redelmeier wrote:
>
> David Peter wrote:
> >
> > Not much good I'm afraid. This typically gives resolution only down to
> > 10ms or so. I'm looking for something in the microsecond or nanosecond
> > range.
>
> Well, if you're running 586+, then you can always use the `rdtsc`
> instruction to read the 64 bit CPU clock counter. On a 500 MHz
> CPU, that will give you 2 ns resolution.
Doesn't gettimeofday on the aforementioned platform use the TSC by default?
------------------------------
From: Nicolas Eymerich <[EMAIL PROTECTED]>
Subject: 970Cxi - HP refuses to help customer
Date: Wed, 24 May 2000 11:53:43 GMT
> in Ghostscript. Last time I looked, support consisted of a driver for
> either the cdj500 or the cdj550. Pretty ancient stuff!
The graphic output is sadisfactory with these settings. Unfortunately,
by asking in a HP customer forum the moderator replies that HP does not
release the codes for 2 sided printing.
Anybody knows if is possible to install a parallel port spy utility to
extract the control code used?
--
http://carini.tripod.com/
http://www.carini.net/
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Joe Ceklosky <[EMAIL PROTECTED]>
Subject: Call stack in C
Date: Wed, 24 May 2000 12:22:57 GMT
How can I get a call stack strace from a
running program. I want something like pstack
on Solaris for Linux
Sample pstack output:
24347: ksh
df6396c8 waitid (7, 0, dffff020, 7)
df65d4d8 _libc_waitpid (0, dffff120, 7, 2, 0, 7) + 58
000229c8 job_wait (c5, 0, 52918, 53578, 5358c, 4be54) + 1dc
0002ed7c sh_exec (c5, 0, 34, 0, 4cc00, 34) + fec
00026e90 ???????? (52918, 508f4, c, 0, 434, 508f4)
000265dc main (330020, 330020, 20, 420, 4cc00, 330020) + ba4
00014d4c _start (0, 0, 0, 0, 0, 0) + 5c
Thanks,
Joe Ceklosky
------------------------------
From: Kai =?iso-8859-1?Q?Hergenr=F6ther?= <[EMAIL PROTECTED]>
Subject: shared libs (.so's) with dependencies
Date: Wed, 24 May 2000 14:15:41 +0200
Reply-To: [EMAIL PROTECTED]
Hello,
Assumed I have a C program prg.c that has to be linked with several
given shared libraries, say libaa.so, libbb.so, libcc.so. To build it I
wold use
gcc -o prg prg.c -laa -lbb -lcc
and it would work fine. Now I remove the main routine from prg.c and
rename the rest to prg_lib.c. My problem now is to create a shared
library from prg_lib.c that incorporates the dependencies on libaa,
libbb, libcc. From what I know (and a great deal of naive intuition) it
would be
gcc -shared -o libprg.so -fPIC prg_lib.c -laa -lbb -lcc =
Now I take a test program, say testprg.c, that uses functions from
libprg and try
gcc -o testprg testprg.c libprg.so =
and I get
undefined reference to `__pure_virtual'
undefined reference to `__ucmpdi2'
I think I have located the implementation of these symbols in libgcc
and libg++ but I don't see how they can end up unresolved at this point.
Is there something wrong with the shared libs libaa, libbb, libcc? Or am
I missing some fundamental detail that has to be taken care of when
building shared libs? I'm new to compiling code to libraries and any
help would really be greatly appreciated.
I need the dependencies in libprg.so since I need to load it as native
code into a Java class using JNI. =
=
Thanks
Kai
-- =
=3D=3D Kai Hergenr=F6ther =
=3D=3D [EMAIL PROTECTED] / [EMAIL PROTECTED]
=3D=3D URL =3D http://www.student.uni-kl.de/~hergen
=3D=3D q.e.d. -KL/M-
------------------------------
From: Someone <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: Need ideas for university funded project for linux
Date: Wed, 24 May 2000 08:25:35 -0400
blah, blah, blah, more hobbies for Computer Science geeks. I have several
hobbies two of them are Digitally recording my band and another one is playing
with Linux. And for me the question is: Pay big bucks for a multitrack cd
pressing software for a Win box, Pay Bigger bucks for a Mac and App, or spend
the rest of my life figuring out how string together linuxs text apps, TCL, and
various X apps. I happen to be a musician and computer literate, but I only
have so much time. I guess the though of his thread is lost. I was just trying
to suggest a killer app that would bring hords of Linux users, sorry if I rained
on your power outlet.
David Steuber wrote:
> Someone <[EMAIL PROTECTED]> writes:
>
> ' Mr Steuber is an excellent example of my point. In my message can't you
> ' tell I LIKE linux, I have Linux, I have used Linux for several years. It
> ' is a great hobby to configure but once it is working what are you going to
> ' use it for? Why don't any of the free ISPs support Linux? That would be
>
> No, I could not tell.
>
> I also don't see how support by free ISPs is any kind of indicator of
> the quality of technology. I wouldn't touch those free ISPs with a 20
> meter, high voltage catle prod.
>
> As for what are you going to use if for, you are clearly limited by
> your imagination. What is it you are doing on Windows? Why can't you
> do it with Linux or *BSD?
>
> I spend a fair amount of my leasure time on usenet and mail. Linux
> seems to do those functions pretty well. I also develop software.
> Linux seems to have some pretty good tools for that too. Linux also
> seems to have more text processing tools than non-unix type systems.
> Also, there is at least one good image processing tool for Linux.
> There are also tools for digital signal processing, content
> distribution, type setting, data storage and retrieval, network
> management, network services, security, data encryption, numerical
> analysis, etc, etc, etc.
>
> It seems to me that anything you could possibly want to do on a
> computer can be done on Linux. If some feature is missing, the tools
> are provided to implement it. Saying that you can't do anything on
> Linux is utterly absurd.
>
> --
> David Steuber | Hi! My name is David Steuber, and I am
> NRA Member | a hoploholic.
>
> All bits are significant. Some bits are more significant than others.
> -- Charles Babbage Orwell
------------------------------
From: Someone <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: Need ideas for university funded project for linux
Date: Wed, 24 May 2000 08:33:40 -0400
I have tried simular searches with no Luck, those have included postings to every
newsgroup I could fine. Nobody has ever been so helpful before. Thanks!!! I am
now going to try this and if it works I will finally be able to leave my server
on all the time. I know what you are thinking "Why ever shut it off, Linux can
run for ever!". Well the SCSII drives are noisy and my wife pulls the plug on
the server everytime she goes into the basement. What she can't hear she wont
shut off!
BTW, your attitude just proved my point.
[EMAIL PROTECTED] wrote:
> Someone <[EMAIL PROTECTED]> writes:
>
> >How can I spin down my SCSI drives when they are not in use? I've
> >been after that for years.
>
> And in all those years, you never went to Google and typed
>
> scsi linux spin down
>
> as your search terms? Because if you did, you'd have found several references
> to scsi-idle, and the 9th hit is the "homepage" for it, appropriately titled
> "Starting and Stopping SCSI drives under linux".
>
> Or you go to Freshmeat, and enter
>
> scsi spin down
>
> in the search box --- and number 6, "noflushd", is what you are after.
> It describes how to handle SCSI disks in its README file.
>
> Now, granted, it sounds like neither solution is completely stable with
> the latest development kernels (the last time I used scsi-idle was in '96),
> but at least it's there, and it's fairly easy to find.
>
> Bernie
> --
> Gentleman, I am a Catholic... If you reject me on account of my
> religion, I shall thank God that He has spared me the indignity of
> being your representative
> Hilaire Belloc
------------------------------
From: Jean-Claude MEILLAND <[EMAIL PROTECTED]>
Subject: Linux Bootable CD
Date: Wed, 24 May 2000 12:52:06 GMT
Hello,
Does anybody know how to make a linux bootable CD with an entire linux
system installed and a ram disk for the system to be usable. Just like
Suse CD Live.
Thanks.
mailto:[EMAIL PROTECTED]
------------------------------
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
From: David Steuber <[EMAIL PROTECTED]>
Date: Wed, 24 May 2000 13:00:00 GMT
[EMAIL PROTECTED] (Dowe Keller) writes:
' [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
' >David Steuber <[EMAIL PROTECTED]> writes:
' >
' >> Do people really have trouble with ./configure, make, make install?
' >> It has _never_ been a problem for me. Maybe I am just lucky. Even
' >> though I changed my compiler, libc, and libtools.
' >
' >That precise process usually works out fine. However, a number of
' >these processes require manual modification of the Makefile or a
' >custom configuration file. I've also encountered several configure
' >scripts that break, and when that happens, you're doomed to rewriting
' >the Makefile by hand. And there are still a few programs that just
' >provide you with a grab-bag of Makefiles, and you get to pick which
' >one you want. Those are *always* disasters, but usually the Makefiles
' >are at least short enough that fixing them isn't impossible.
'
' Yow, you make it sound like brain surgery. I can count the number of
' times that I had to hack Makefiles to get a program to make on the
' fingers of one hand. And multiple make files are a good thing if you
' have to do special stuff to get a program to compile on a particular
' system. I'd rather there be three makefiles named:
'
' Makefile.AIX
' Makefile.Linux
' Makefile.BSD
'
' Than to have to edit a single Makefile to work.
Sorry to quote everything, but I thought that this is what
autoconf/automake was supposed to fix. Then there is TrollTech's
tmake. Life must get interesting at times.
--
David Steuber | Hi! My name is David Steuber, and I am
NRA Member | a hoploholic.
All bits are significant. Some bits are more significant than others.
-- Charles Babbage Orwell
------------------------------
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
From: David Steuber <[EMAIL PROTECTED]>
Date: Wed, 24 May 2000 13:00:01 GMT
Maciej Golebiewski <[EMAIL PROTECTED]> writes:
' How often the makefiles provide the "uninstall" target, too? Or at least
' a script for un-installing? Otherwise, over the time and number of upgrades
' to newer versions comprised of different files, you're accumulating
' "abandoned"
' files.
It does happen. I don't know what percentage of makefiles provide
that or a simiar target. Most produce a single executable image that
is easy to dispose of.
' RPM is not perfect but it is quite OK. It's just that the guys creating rpm
' not always can get the dependencies right. Personally one of my favourite
' query options in rpm is -q -f to instantly get the name of the package
' "owning" a specific file. I love it.
What do you do when two packages claim ownership of the same file?
--
David Steuber | Hi! My name is David Steuber, and I am
NRA Member | a hoploholic.
All bits are significant. Some bits are more significant than others.
-- Charles Babbage Orwell
------------------------------
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
From: David Steuber <[EMAIL PROTECTED]>
Date: Wed, 24 May 2000 12:59:59 GMT
[EMAIL PROTECTED] (Leslie Mikesell) writes:
' >Have I violated someone's license?
'
' Just putting the code on the same machine is not a problem.
' You must create a 'derived work', which has been commonly
' described as code linked in a single process. Linking
' GNU readline (GPL, not LGPL) into a database control
' program that used a commercial client library would be
' an example. While you can probably get away with building
' such a program yourself, distributing it would be at
' least questionable, especially if you redistributed a copy
' of the GPL'd readline with it. Now for something even
' more confusing, consider what happens if you have a
' perl script that dynamically loads readline and also
' uses DBI which can pull in an assortment of database
' client libraries at runtime, including commercial versions.
' If this ends up linking to (say) Oracle libs, does it
' become illegal to distribute the script?
I would think not. The dynamicly loaded modules are not part of the
script, they are just used by the script.
' For an even stranger case, consider mod_perl, compiled into apache,
' both of which can dynamically link other modules and the process
' lifetime spans many requests. If one web requests runs a script
' that links a GPL'd library, and another pulls in a commercial
' library, you now have a derived work (by some definition)
' that is not allowed, yet no single thing is responsible for it.
ibid.
' >I am not trying to create a situation where you or anyone else can't
' >use code that I write or code that is derived from code that I write.
' >I am trying to avoid the situation where improvements to my code are
' >not returned to me or to others. The whole point of FSF style free
' >software is to advance the state of the art by not shackling code with
' >proprietary licensing.
'
' The GPL instead shackles any other code linked into a derived
' work with its own restrictions, or in the cases where
' other code already has different restrictions it makes the
' combination impossible.
I think the only code shackled by GPL is the code that is derived from
GPL. In the case of linkage above, the top level code must be
compatible with GPL, but the libs it uses does not have to be.
If I have GPL lib A and proprietary lib B and my code C uses A and B,
then my code C is GPL. I can distribute my code without distributing
A or B. My code may even be able to work without A or B with reduced
or different functionality.
' >If there is a better way to achieve this goal, please tell me about
' >it.
'
' In the case of a complete stand-alone package, restrictions on
' derived works may not be a problem. For things that would
' be useful as a component, the LGPL or BSD style opens up
' more possibilities.
OK, you say reusable components like libs are better off with a less
restrictive license. That is fair. But if the lib is under a BSD
style license, Microsquish can take that library, change it, and not
release any of the changes back to the public. Once you use an
'enhanced' feature, you are tied to a non-free Microsquish library.
You are at the mercy of Microsquish. If the code is at least LGPL,
then the library code is still going to be free. Microsquish must
make its changes available as source.
Now if the library is GPL, then anything linking too it must also be
GPL or GPL compatible. What about software that makes system calls in
Linux? Must that also be GPL? Should I demand that Netscape ( now
AOL ) release _all_ the code for Navigator? Surly all programs must
make system calls at some level. Does the Linux kernel make
commercial software for Linux impossible? Make that closed source,
proprietary software. No Oracle or Sybase for Linux?
I guess that must mean running such software on *BSD instead. But
what if the application uses networking protocols to link to other
software? Is that linkage that GPL affects? How far does it go?
The possibility that closed source or non-GPL compatible software can
not be legally run on Linux does bother me. Although I would prefere
all software to be GPL, I do see the need to accommodate people who do
not like the GPL. Having a libc for each possible license category is
certainly wasteful if the duplication is soley to deal with license
concerns. I guess it's a good thing libc is LGPL.
--
David Steuber | Hi! My name is David Steuber, and I am
NRA Member | a hoploholic.
All bits are significant. Some bits are more significant than others.
-- Charles Babbage Orwell
------------------------------
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
From: David Steuber <[EMAIL PROTECTED]>
Date: Wed, 24 May 2000 13:00:00 GMT
[EMAIL PROTECTED] (Matthias Warkus) writes:
' It was the Tue, 23 May 2000 08:59:59 GMT...
' ...and David Steuber <[EMAIL PROTECTED]> wrote:
' > The right is non-exlusive. That means everyone can get that right. I
' > think TrollTech is just trying to prevent forking of the Qt library
' > here.
'
' Exactly that is which is bad IMHO. Real software freedom has always
' been the freedom to fork.
That's a good point. But what is the value in real forking? Do you
really want to have ten different major versions of GTK+ floating
around? Or even two? If an application says it uses GTK+ ver x.y,
wouldn't it be simpler to know that the application didn't really mean
FGTK+ x.y?
Linux is held together because people respect the opinion of Linus
Torvalds. Even so, there are Alan Cox diffs, RTLinux, and probably
others. The tendency is to stick with Linus Torvalds Linux as the
base. Will that be true for all GPL projects? What if some group
decides a certain feature is needed in GTK+, but another group of
equal size feels that feature doesn't belong, or should be implemented
in a different way, with a different interface?
I guess Qt doesn't really solve that problem since you can just use
inheritance to provide feature X in any number of ways.
Now the people at Microsquish might say that this is why you need one
company enforcing a single standard way of doing things. And yet they
too are stuck with feature creep and redundant implementations too.
It is a jungle out there.
--
David Steuber | Hi! My name is David Steuber, and I am
NRA Member | a hoploholic.
All bits are significant. Some bits are more significant than others.
-- Charles Babbage Orwell
------------------------------
** 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.development.system) 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-Development-System Digest
******************************
