Linux-Development-Sys Digest #622, Volume #7 Fri, 25 Feb 00 00:13:13 EST
Contents:
Re: Want to work with OSS for a living? (MA) (YtRabbit)
Re: Binary compatibility: what kind of crack are they smoking? (Donovan Rebbechi)
Re: Binary compatibility: what kind of crack are they smoking? (Donovan Rebbechi)
Re: Binary compatibility: what kind of crack are they smoking? (Donovan Rebbechi)
Linux and DVD ("Shazam")
Re: Binary compatibility: what kind of crack are they smoking?
Looking for a decent development environment ("Lord Petrosky")
RedHat 6.0 ("Sake")
TCP/IP socket from kernel ("Sake")
Re: Binary compatibility: what kind of crack are they smoking? (Colin Watson)
Re: Binary compatibility: what kind of crack are they smoking? (Albert Ulmer)
Re: Binary compatibility: what kind of crack are they smoking? (Donovan Rebbechi)
Can a previous handler be saved when requesting an IRQ? (Luis Miguel Pinho)
Re: Looking for a decent development environment (Grant Edwards)
Re: IP adresses not resolved - gethostbyname failed (Thomas F. Drescher)
Re: Does linux support DIRECT I/O? (Hong Shen)
Re: Binary compatibility: what kind of crack are they smoking? (Craig Kelley)
Re: Looking for a decent development environment ("Lord Petrosky")
Re: Binary compatibility: what kind of crack are they smoking? (Craig Kelley)
Re: Q: How good is Linux when the computer is suddenly loses power ? (Miguel Cruz)
Re: Binary compatibility: what kind of crack are they smoking? (Kaz Kylheku)
----------------------------------------------------------------------------
From: YtRabbit <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.development.apps
Subject: Re: Want to work with OSS for a living? (MA)
Date: Thu, 24 Feb 2000 23:21:53 GMT
In article <[EMAIL PROTECTED]>, mlw <[EMAIL PROTECTED]> wrote:
> I am currently working at a net service company. We need "web guys" you
> should know some C/C++, but mostly we need UNUX Admin (Freebsd), Apache,
> PHP, Perl, and the ability to rip apart poorly written (app generated)
> customer HTML, and make it pretty, functional, and problem free.
>
> Reply to my e-mail directly.
>
> Sorry to post this on the discussion groups, but we have been looking
> for people for months and the normal channels are not working, and this
> is a form of advocacy, right?
>
> The company is in Westborough Massachusetts.
>
>
I'd love to, and I have the qualifications, but I live on Long Island, NY, and I just
got back from a 2 year stint in the Twin Cities; besides, the Mrs. is fed up with the
nomadic lifestyle. I'll pass on the word to the few qualified people I know in your
neck of the woods
'nuff said,
YtRabbit.
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: 25 Feb 2000 00:57:40 GMT
On Thu, 24 Feb 2000 18:52:47 +0100, Mario Klebsch wrote:
>Craig Kelley <[EMAIL PROTECTED]> writes:
>The Windows plattforms have at least ABIs in common, like the win32
>ABI.
And what on earth is an "ABI" ?
>BTW, Windows did the same with supporting the win16 ABI on Win9x.
>
>But in Linux, there is no ABI, and this is the problem.
You gave no idea what you are talking about. The KDE and GNOME projects
are primarily about APIs. There is the QT/KDE API on Linux. This ships
with all distributions, as does GTK/GNOME.
So yes, in short, there are development tools on Linux that do the
same thing for Linux developers that MFC does for Windows developers.
--
Donovan
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: 25 Feb 2000 01:03:10 GMT
On Thu, 24 Feb 2000 18:42:33 +0100, Mario Klebsch wrote:
>Craig Kelley <[EMAIL PROTECTED]> writes:
>So if Linux will not be an OS, there sould be at least a way to define
>the ABI of the linux versions, and the build process of the individual
>libraries should be modified to guarantee, that the specified
>interface is implemented by the libarary.
If you're trying to say that the distributors
should standardise on core shared library versions, I agree.
Unfortunately, it's near impossible to get them to agree on *anything*,
and the result is that the attempt to standardise things, LSB, fell flat
on its face.
At this stage, this does not appear to be a major setback. The application
developer can simply require versions (x,y,z) of the appropriate shared
libraries and ship them with the application. No, this is not the most
efficient solution , indeed it's almost as bad as static linking. But
with large applications like office suites, the overhead of loading an extra
libc is kind of small in the big scheme of things.
--
Donovan
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: 25 Feb 2000 01:06:33 GMT
On Thu, 24 Feb 2000 18:30:52 +0100, Mario Klebsch wrote:
>>This comes from the fact that the C++ part is not as carefully kept up
>>to date as other parts of the system, e.g. libc. It is not a core part,
>>so many forget about it. Since getting the last libstdc++ together with
>>the last G++ is as easy as downloading the latest stable release of gcc,
>>issuing some 10 shell commands, perhaps you can do as others do with
>>libc: glibc2.1 required, basta.
>
>But getting new versions of the libraries is no help. I often had the
>problem, that I needed older versions of them. While it was not that
>hard to find the older verion, compiling it was only a minor problem,
Binary only software should ship with all the required shared libraries.
On the other hand, if your software is not binary only, you can just recompile
that instead of trying to futz with your shared library versions.
--
Donovan
------------------------------
From: "Shazam" <[EMAIL PROTECTED]>
Subject: Linux and DVD
Date: Thu, 24 Feb 2000 20:09:49 -0500
On Luxsonor's website is the following:
****************************
LuxSonor does not have the resources or even the proper people to do a port
to Windows NT 2000 anymore.
We have not been in the PC business for about 2 years. We have tried to
outsource our Windows 2000 driver port to people that did the previous ports
before. Unfortunately, no reasonable agreement could be arranged.
The best we can do is release our source code to the public.
****************************
This is bad news and great news at the same time! Bad news in that I was
looking forward to Windows 2000 drivers for our Luxsonor based DVD cards
(ie. CT7160, Labway, Procomp M410 etc). Great news in that Luxsonor has
decided to give us users the greatest gift of all: Open Source Drivers.
This means that if someone out there is smart enough (and there are a lot of
smart coders out there), we can expect to see continued Windows 98 driver
support for Luxsonor based DVD decoder cards! In addition, we can expect to
see Linux and Windows 2000 drivers for our cards too.
Let's give Luxsonor a big hand of applause for supporting its customers in
such a fashion! There are a lot of companies out there who wouldn't do
this.
------------------------------
From: [EMAIL PROTECTED] ()
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: 24 Feb 2000 20:31:16 -0500
Adam Ierymenko <[EMAIL PROTECTED]> spewed this unto the Network:
>Here is the output of 'ls /usr/lib/libstdc++*' on my system:
>
>/usr/lib/libstdc++-2-libc6.1-1-2.9.0.so
>/usr/lib/libstdc++-3-libc6.1-2-2.10.0.a
>/usr/lib/libstdc++-3-libc6.1-2-2.10.0.so
>/usr/lib/libstdc++-libc6.1-1.so.2
>/usr/lib/libstdc++-libc6.1-2.a.3
>/usr/lib/libstdc++-libc6.1-2.so.3
>/usr/lib/libstdc++.so.2.7.2
>/usr/lib/libstdc++.so.2.7.2.8
>/usr/lib/libstdc++.so.2.8
>/usr/lib/libstdc++.so.2.8.0
>
>Good Lord! Lets see... I need 2.7.2 for a couple things
>(Acrobat Reader I think).. 2.8 for a couple other things...
>I need one of the newer stdc++ libraries with the funky name
>for egcs-compiled binaries like the new Netscape 4.72 and
>stuff that I compile links with the other one.
Why don't you do 'ls -F /usr/lib/libstdc++*' to see how many of
those are symlinks? On a lot of Unix systems, all of the older
versions of an SO are symlinks to the newest binary-compatible
version. For example, on my system, libstdc++.so.2.8 is a
symlink to libstdc++.so.2.8.0 (and libstdc++.so.2.8.0 is the
newest libstdc++ version on my system).
--
Guns don't kill people, cops do!
------------------------------
From: "Lord Petrosky" <[EMAIL PROTECTED]>
Subject: Looking for a decent development environment
Date: Thu, 24 Feb 2000 20:12:12 -0500
hi,
i'm getting into some programming (C, Perl, Java maybe) and i figure there
must be certain "environments" with integrated "tools" that make the job
interesting. i don't want anything too automatic; i like to do things
manually when i'm learning something new. at the same time, though, i don't
want to reinvent the wheel.
any ideas?
tia,
LP
------------------------------
From: "Sake" <[EMAIL PROTECTED]>
Subject: RedHat 6.0
Date: Thu, 24 Feb 2000 21:14:24 -0500
Hi there,
Given the popularity of RedHat Linux, there probably should be a RedHat
group.
Anyway, I'll use this for now.
Problem that I have is that, cat on /proc files and 'dmesg' both show 64MB
of RAM although
my machine is equipted with 256MB of RAM. I'm using RedHat 6.0 fresh out of
box.
Where is my memory ?? :(
Any hints are appreciated.
------------------------------
From: "Sake" <[EMAIL PROTECTED]>
Subject: TCP/IP socket from kernel
Date: Thu, 24 Feb 2000 21:23:27 -0500
Hi,
What is the "right" way to use TCP/IP socket functions in a driver/module ?
I try to write a module that serves requests from TCP.
Thanks
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: 25 Feb 2000 03:10:34 GMT
In article <[EMAIL PROTECTED]>,
Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>On Thu, 24 Feb 2000 18:52:47 +0100, Mario Klebsch wrote:
>>The Windows plattforms have at least ABIs in common, like the win32
>>ABI.
>
>And what on earth is an "ABI" ?
Application Binary Interface: I don't know the precise details, but it
involves things like calling conventions.
>>BTW, Windows did the same with supporting the win16 ABI on Win9x.
>>
>>But in Linux, there is no ABI, and this is the problem.
>
>You gave no idea what you are talking about. The KDE and GNOME projects
>are primarily about APIs. There is the QT/KDE API on Linux. This ships
>with all distributions, as does GTK/GNOME.
API stands for Application Program[ming] Interface. This is something
entirely different; it usually involves a specification of various
function calls that are exported from libraries to applications that
link against them.
Qt and KDE don't ship with all distributions, by the way. In particular,
Debian won't include them because of licensing issues (the QPL has fatal
incompatibilities with the GPL).
--
Colin Watson [[EMAIL PROTECTED]]
"And I forgot the next verse / Oh well, I guess it pays to rehearse
The music sheet's so hard to find / What are the words? Oh nevermind"
- "Smells Like Nirvana", Weird Al
------------------------------
From: Albert Ulmer <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: Fri, 25 Feb 2000 03:19:35 +0000
Adam Ierymenko wrote:
> > There is no such thing as a Linux OS! There is RedHat Linux,
> > SUSE-Linux, debian Linux, and lots of other. Face it, Linux already
> > has fragmented.
> Actually, if all the distros would just standardize on one
> package manager format, it would be OK. Right now it's
> RPM and DPKG... or maybe make DPKG and RPM able to
> install each others' packages?
But they already are! Search freshmeat.net for alien.
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Date: 25 Feb 2000 03:24:39 GMT
On 25 Feb 2000 03:10:34 GMT, Colin Watson wrote:
>In article <[EMAIL PROTECTED]>,
>Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>Application Binary Interface: I don't know the precise details, but it
>involves things like calling conventions.
Oh, I see.
I'd agree that Linux needs something like this. If it wasn't for all
the petty squabling, we'd have LSB out the door. Unfortunately, we have libc
du jour instead.
>Qt and KDE don't ship with all distributions, by the way. In particular,
>Debian won't include them because of licensing issues (the QPL has fatal
>incompatibilities with the GPL).
They don't ? Hmmm ...
--
Donovan
------------------------------
From: Luis Miguel Pinho <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development
Subject: Can a previous handler be saved when requesting an IRQ?
Date: Wed, 23 Feb 2000 17:00:56 +0000
Hi,
I'm making some experiments with kernel interrupts (I'm
using redhat 6.0, kernel 2.2.5-15 and redhat 5.2, kernl
2.0.36), but there is one thing that is puzzling me.
If I free a currently used irq, is there a way to save which
was the handler?
I can't find a kernel function to do it. My only guess is by
manipulating irqaction structures.
Thanks
Luis Miguel Pinho
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: Looking for a decent development environment
Date: Fri, 25 Feb 2000 03:44:34 GMT
On Thu, 24 Feb 2000 20:12:12 -0500, Lord Petrosky <[EMAIL PROTECTED]> wrote:
>i'm getting into some programming (C, Perl, Java maybe) and i figure there
>must be certain "environments" with integrated "tools" that make the job
>interesting.
Yup. It's called "Unix". There are hundreds of tools ranging
from grep to gdb to ctags to Emacs. Most of the tools can be
hooked to gether using pipes, and automated using
shell-scripts.
--
Grant Edwards grante Yow! What's the MATTER
at Sid?... Is your BEVERAGE
visi.com unsatisfactory?
------------------------------
From: Thomas F. Drescher <[EMAIL PROTECTED]>
Subject: Re: IP adresses not resolved - gethostbyname failed
Date: Fri, 25 Feb 2000 05:45:24 GMT
Thanks William,
when i put IP numbers in the /etc/hosts.allow it works !?! I still=20
wonder why the translation does not work. The /etc/hosts contains the=20=
necessary definitions.
P.S. Sorry for the =3D20=3D's... dunno what the software does...
>>>>>>>>>>>>>>>>>> Urspr=FCngliche Nachricht <<<<<<<<<<<<<<<<<<
Am 24.02.00, 08:41:13, schrieb [EMAIL PROTECTED] (William Park) zum=20
Thema Re: IP adresses not resolved - gethostbyname failed:
...
> It's difficult to read your post, with all that '=3D...' quotes. But,=
=20
try
> using IP in your /etc/hosts.allow.
> --William
> --
------------------------------
From: Hong Shen <[EMAIL PROTECTED]>
Subject: Re: Does linux support DIRECT I/O?
Date: Fri, 25 Feb 2000 03:30:18 GMT
I am working on a video editing program on linux. Because of the high
bitrate of the video file, the caching seems no use and may add some
overhead.
Thank you very much!
Marc SCHAEFER wrote:
>
>
> Hong Shen <[EMAIL PROTECTED]> wrote:
> : Can anybody tell me if there is DIRECT I/O support in Linux? I read the
> : man page of "open", these isn't O_DIRECT support. I am running Redhat
6.0
> : on an Intel PC.
>
> SGI supports on IRIX a O_DIRECT flag which means that data should not
> be cached on that file and that data should be DMAed directly to user
space.
> To my knowledge, this means you must ensure that your buffer is locked
> in memory and that it's page-aligned.
>
> On Linux, the closest which is implemented is the raw device access,
> available as a patch in 2.2.x. Raw devices are accessed through /dev/rawX
> devices, mapped to real devices with a special utility.
>
> Now, why do you want O_DIRECT for ?
>
>
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
From: Craig Kelley <[EMAIL PROTECTED]>
Date: 24 Feb 2000 21:09:59 -0700
[EMAIL PROTECTED] (Mario Klebsch) writes:
> Craig Kelley <[EMAIL PROTECTED]> writes:
>
> >[EMAIL PROTECTED] (Mario Klebsch) writes:
>
> >> Craig Kelley <[EMAIL PROTECTED]> writes:
> >> >It is a difficult problem. UNIX is up-front with the user, Windows
> >> >and MacOS hide it and then certain applications start "acting funny";
> >> >hence fun tools like "Conflict Catcher" and "First Aid".
> >>
> >> But Linux is not UNIX, unfortunately. I'd love to have the UNIX
> >> aproach, but the Linux aproach is braindead.
>
> >Linux certainly behaves like UNIX (if are a pedigree bigot) in this
> >regard. If you don't believe me, then explain how Linux differs from
> >"real" UNIX (whatever that is) for library support.
>
> That is easy to explain. When I write programs for e.g. Solaris 7, I
> can be shure to know a set of shared libraries, that will be available
> on the target system. And I can be sure, they have the correct ABI.
>
> When writing a Program for e.g. Linux 2.2.13, I cannot rely on
> anything except the system calls. This is an ABI, too, but far from
> sufficient for most programs.
>
> Since Solairs is an operating system, specifying its version does
> specify the version (and th4e interface) of all shared libraries,
> too. There are patches, but the pates usually do not change the
> interface of the libs, not the API, nor the ABI.
I've never used Solaris, but I have had many library problems with
AIX; even having to install updated packages from IBM to get Oracle to
work. Oracle, at one time, even had different versions of their
package for different OS levels of AIX *within* the 3x series.
I've had extensive problems with libraries under Windows NT; and many
programs require a certain service pack or even that a certain version
of IE be installed.
It is a problem, but it isn't indiginous to Linux.
> In linux, everything is on its own. I can have libz.so.1.1.3, but even
> this specification (altough including a version number) does not
> specify it all. I can have one libz.so.1.1.3 linked agains libc5, and
> one libz.so.1.1.3 linked against libc6.
>
> So if Linux will not be an OS, there sould be at least a way to define
> the ABI of the linux versions, and the build process of the individual
> libraries should be modified to guarantee, that the specified
> interface is implemented by the libarary.
>
> Linux could include a set of ABI specifications for the Linux set of
> libraries, and when linking the libraries, the linker can check,
> wehter the library maches the specified ABI.
Instead of doing this, Linux has dependency-checking package
management systems. It may not be the best solution, but it certainly
isn't worse than others, in my experience.
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: "Lord Petrosky" <[EMAIL PROTECTED]>
Subject: Re: Looking for a decent development environment
Date: Thu, 24 Feb 2000 22:53:19 -0500
err, how about in english now? seriously, i have gotten over the beginner's
hump with Linux and am presently getting deeper into Samba and Apache. i
want to program mainly for the web as well as get into some (non web)
network programming. do you know how to start? throwing out acronyms
doesn't really help; i am looking for maybe a site or a downloadable package
that explains things in a logical manner.
thanks anyways for your reply,
LP
>
> >i'm getting into some programming (C, Perl, Java maybe) and i figure
there
> >must be certain "environments" with integrated "tools" that make the job
> >interesting.
>
> Yup. It's called "Unix". There are hundreds of tools ranging
> from grep to gdb to ctags to Emacs. Most of the tools can be
> hooked to gether using pipes, and automated using
> shell-scripts.
>
> --
> Grant Edwards
------------------------------
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
From: Craig Kelley <[EMAIL PROTECTED]>
Date: 24 Feb 2000 21:17:18 -0700
[EMAIL PROTECTED] (Mario Klebsch) writes:
> Craig Kelley <[EMAIL PROTECTED]> writes:
>
> >[EMAIL PROTECTED] (Mario Klebsch) writes:
>
> >> There is no such thing as a Linux OS! There is RedHat Linux,
> >> SUSE-Linux, debian Linux, and lots of other. Face it, Linux already
> >> has fragmented.
> >>
> >> This will kill Linux, because if you want to support Linux, you have
> >> to support more than a hand foll of operating systems, that only
> >> differ slightly. And although you officially are supporting Linux, you
> >> are not supporting all thos linux systems, that are not just installed
> >> from CD, but self compiled.
>
> >Oh? Like Windows 95, 98, NT4, NT4/Alpha, NT2k, and the CE variants?
>
> The Windows plattforms have at least ABIs in common, like the win32
> ABI.
And yet there are plethora of programs which only work under certain
service packs and certain versions of Windows. There are programs
that only work well/at all with certain hardware. It would be
difficult to point to any Win32 product more complex than WinZip which
*doesn't* have some sort of issue with binary compatibility under
certain circumstances.
> CE does not need to have anything in common, since binary
> compatibility to anything is not its purpose. We are talking about
> ABIs, the interface between the OS and binaries running on it.
True, but then again, you can't just go and buy a CE product and
expect it to work flawlessly.
> >Or MacOS/68k, MacOS/PPC, Carbon, Cocoa?
>
> But MacOS/PPC included tha ABI of MacOS/68k, and the new Macs with
> Carbon ABI still do support the Macintosh ABI. IMHO this is an
> excellent example on how older ABIs can be supportet on new
> architectures.
I see you've never tried to install Cricket Graph on a G3.
> BTW, Windows did the same with supporting the win16 ABI on Win9x.
With erratic results. We had to have a Windows 3.11 machine around to
run minisoft for quite some time before a 32-bit version came out.
> But in Linux, there is no ABI, and this is the problem.
True, but it's an industry-wide problem, not inherent to Linux.
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
Crossposted-To:
comp.os.linux.hardware,comp.os.linux.development.apps,linux.redhat.devel,linux.redhat.development
Subject: Re: Q: How good is Linux when the computer is suddenly loses power ?
From: [EMAIL PROTECTED] (Miguel Cruz)
Date: Fri, 25 Feb 2000 04:36:35 GMT
Markus Wandel <[EMAIL PROTECTED]> wrote:
> (I have since gotten a longer power cord, eliminating the accidental
> powerdowns!)
Does that work the same way that a longer garden hose means you have a
little more water around when the main gets shut off?
miguel
------------------------------
From: [EMAIL PROTECTED] (Kaz Kylheku)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.setup
Subject: Re: Binary compatibility: what kind of crack are they smoking?
Reply-To: [EMAIL PROTECTED]
Date: Fri, 25 Feb 2000 05:07:32 GMT
On Thu, 24 Feb 2000 18:52:47 +0100, Mario Klebsch <[EMAIL PROTECTED]> wrote:
>>Oh? Like Windows 95, 98, NT4, NT4/Alpha, NT2k, and the CE variants?
>
>The Windows plattforms have at least ABIs in common, like the win32
>ABI. CE does not need to have anything in common, since binary
>compatibility to anything is not its purpose. We are talking about
>ABIs, the interface between the OS and binaries running on it.
The Intel Windows platforms break binary interfaces in subtle ways. You just
have to flip through pages of MSDN to see what I mean. You have functions that
are in one platform but not the others. Then there are some that behave
differently.
For example, take a look at the InterlockedIncrement() function. It does
one thing on Windows 95 and another on NT 4.
In my software team, we have had problems getting some executables and
libraries to work under 95 that were developed on NT. It boiled down to
searching the DLLs to find the offending missing symbol.
When you develop Windows applications, you must validate them on each platform
that you intend to ship on. You would never compile and test something on NT
only and then release it as Windows 95 or Windows 98 software.
So as a Windows user, the compatibility you see among the platforms is largely
thanks to the application developers who ensure that their stuff works across
these platforms. Not due to the exitence of any ABI standard!!!
Microsoft wouldn't know a standard if it was put into a four inch binder and
thrown at them. They are unable to standardize interfaces control of which is
in their sole posession.
------------------------------
** 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
******************************
