Thougths on the gnu project

Thougths on the gnu project

Post by Christopher B. Brow » Sun, 27 Sep 1998 04:00:00





>> 1. From the gnu homepage:
>> "The GNU Project started in 1984 to develop a complete free Unix-like
>> operating system."

>> 2. From the Highest Priority list:
>> "Develop a substitute, which runs on GNU systems, for some very popular
>> or very important application that
>>     many non-programmers use on Windows, and which has no comparable
>> free equivalent now. "

>> I think that all the components for a really great OS exist:
>> GNU/Linux , as the kernel and base, use debian,redhat,slackware or
>> whatever
>> XFree86 , the most widely used windowing system on unix
>> KDE , a new great gui that runs on top of X

>KDE can't hold a candle to GNOME.

>> The only problem with this is that KDE depends on QT which is not free,
>> this raises various problems (see
>> http://www.redhat.com/redhat/qtlicense.html ).

>> Wouldn't writing a GPL:ed replacement for QT be the ideal GNU project
>> and among the highest in priority ??

>Even if there were a free qt replacement, kde would be crappy, cos it lacks
>support for Guile scripting and the SCWM. Thus Kde is obsolete.

I thought that they were both crappy because they used graphics, which
is evil.

And it is indeed a well-known fact that KDE includes a daemon that
looks for any processes called anything similar to "guile" and kills
it, as well as doing a regular "find" to find and stomp out any code
that is neither written in C++ nor linked to Qt.  SCWM would get
"nuked" from several perspectives under that policy.

(Before flaming, think about how likely this is to be true, and
what it is responding to...)
--
Those who do not understand Unix are condemned to reinvent it, poorly.  
-- Henry Spencer          <http://www.hex.net/~cbbrowne/lsf.html>

 
 
 

Thougths on the gnu project

Post by David Joff » Sun, 04 Oct 1998 04:00:00




> >> The only problem with this is that KDE depends on QT which is not free,
> >> this raises various problems (see
> >> http://www.redhat.com/redhat/qtlicense.html ).

> >> Wouldn't writing a GPL:ed replacement for QT be the ideal GNU project
> >> and among the highest in priority ??

There is a project underway to develop a Qt-compatible replacement for
Qt. It's called "Harmony"; theres a link to it from the main KDE page
http://www.kde.org/ .

 - David

--
-----------------------------------------
David Joffe  djoffe at icon dot co dot za
http://www.geocities.com/SoHo/Lofts/2018/
-----------------------------------------

 
 
 

Thougths on the gnu project

Post by Joe Bu » Tue, 06 Oct 1998 04:00:00



Quote:>There is a project underway to develop a Qt-compatible replacement for
>Qt. It's called "Harmony"; theres a link to it from the main KDE page
>http://www.kde.org/ .

I hope that these guys can pull it off.  I would advise them to get
legal papers from all contributors certifying that they have not looked
at Qt source (their FAQ says they don't, they need to be able to make
this rock-solid).

--
-- Joe Buck

http://www.welsh-buck.org/

 
 
 

Thougths on the gnu project

Post by David Kastru » Wed, 07 Oct 1998 04:00:00




> >There is a project underway to develop a Qt-compatible replacement for
> >Qt. It's called "Harmony"; theres a link to it from the main KDE page
> >http://www.kde.org/ .

> I hope that these guys can pull it off.

I am not sure that I agree here.  Cloning a moving commercial target
with volunteer work is not generally working out very convincing.
Look at the Wine and Lesstiff projects.

One problem is that such projects are designed with the confidence
that one has man-years to spare in maintaining them.  Thus they often
tend to be pretty maintenance-intensive due to design reasons.

In contrast, free projects where people are only working intermittedly
or on freetime tend to have much more thinking in them before doing,
because doing it right the first time is very much important (with
commercial development, the first objective is getting something
running fast, then do it correctly later (for which there never is
time enough)).

Lesstiff has not made Motif programming popular or easy.  Wine has not
made Windows programming popular or easy.  Yes, Harmony is an
essential component for being able to push KDE as a standard desktop
also for free systems.  I am just not certain that it will be enough.

--
David Kastrup                                     Phone: +49-234-700-5570

Institut fr Neuroinformatik, Universit?tsstr. 150, 44780 Bochum, Germany

 
 
 

Thougths on the gnu project

Post by Wern » Wed, 07 Oct 1998 04:00:00



>I am not sure that I agree here.  Cloning a moving commercial target
>with volunteer work is not generally working out very convincing.
>Look at the Wine and Lesstiff projects.

Well, Wine is a different thing. Qt is just a class library for
windowing stuff, whereas Wine is a complete windows emulator which
partially has to emulate dos stuff, driver stuff etc... It's almost a
miracle that Wine works at all. I have pretty much faith that they can
clone Qt so that KDE is working properly.

Werner

-----
Memory Dragon

Check out ftp://ftp.gmd.de/if-archive for something which has
been forgotten years ago.

 
 
 

Thougths on the gnu project

Post by David Kastru » Wed, 07 Oct 1998 04:00:00




> >I am not sure that I agree here.  Cloning a moving commercial target
> >with volunteer work is not generally working out very convincing.
> >Look at the Wine and Lesstiff projects.
> Well, Wine is a different thing. Qt is just a class library for
> windowing stuff, whereas Wine is a complete windows emulator which
> partially has to emulate dos stuff, driver stuff etc... It's almost a
> miracle that Wine works at all. I have pretty much faith that they can
> clone Qt so that KDE is working properly.

I did not deny that.  What I said is that I doubt they can add
attractiveness or sex appeal or whatever to Qt programming.  Lesstiff
did not in any groundbreaking way broaden the acceptance of the Motif
interface in the free world.  I doubt that Harmony will be able to
significantly increase the enthusiasm for Qt programming in the free
software world when "the real thing" will still be commercial.

--
David Kastrup                                     Phone: +49-234-700-5570

Institut fr Neuroinformatik, Universit?tsstr. 150, 44780 Bochum, Germany

 
 
 

Thougths on the gnu project

Post by j.. » Wed, 07 Oct 1998 04:00:00





>> >I am not sure that I agree here.  Cloning a moving commercial target
>> >with volunteer work is not generally working out very convincing.
>> >Look at the Wine and Lesstiff projects.
>> Well, Wine is a different thing. Qt is just a class library for
>> windowing stuff, whereas Wine is a complete windows emulator which
>> partially has to emulate dos stuff, driver stuff etc... It's almost a
>> miracle that Wine works at all. I have pretty much faith that they can
>> clone Qt so that KDE is working properly.

>I did not deny that.  What I said is that I doubt they can add
>attractiveness or sex appeal or whatever to Qt programming.  Lesstiff
>did not in any groundbreaking way broaden the acceptance of the Motif
>interface in the free world.  I doubt that Harmony will be able to
>significantly increase the enthusiasm for Qt programming in the free
>software world when "the real thing" will still be commercial.

        What they really need to do is get off of this bloatware
        kick, stop trying to come up with the uber library, stop
        trying to play Microsoft's game and try to draw more on the
        strengths of Unix (philosophically).

        Widget Libs, DnD & remote object libs should all be distinct
        entities with attempts to generate the ultimate one of each
        instead of the ultimate one of all.

        A lib to bring together all of the DnD standards as seamlessly
        as possible would be nice such that the OffiX DnD in WM and
        dfm can interact with the XDnD in Gnome/gtk and KDE/gt.

--
Hardly. Microsoft has brought the microcomputer OS to
the point where it is more bloated than even OSes from          |||
what was previously larger classes of machines altogether.     / | \
This is perhaps Bill's single greatest accomplishment.    

        In search of sane PPP docs? Try http://penguin.lvcm.com

 
 
 

Thougths on the gnu project

Post by j.. » Wed, 07 Oct 1998 04:00:00





>>What they really need to do is get off of this bloatware
>>kick, stop trying to come up with the uber library, stop
>>trying to play Microsoft's game and try to draw more on the
>>strengths of Unix (philosophically).

>To be sure...

>>Widget Libs, DnD & remote object libs should all be distinct
>>entities with attempts to generate the ultimate one of each
>>instead of the ultimate one of all.

>I'll disagree.  There should be attempts to provide a couple of good,
>interoperable options for each.

>It is reasonable to hope for ones that optimize in different
>directions, whether:
>a) Optimizing for size,
>b) Optimizing for speed,
>c) Optimizing for flexibility.

>>A lib to bring together all of the DnD standards as seamlessly
>>as possible would be nice such that the OffiX DnD in WM and
>>dfm can interact with the XDnD in Gnome/gtk and KDE/gt.

>I ran dfm for a little while on Saturday; it looked pretty neat.  

>The significant downside, from my perspective, was that it seems to take
>over the desktop.  Menus normally provided by the WM seem to go away.

        You can configure that. Just dig around the desktop
        icon menus until you find Desktop Context Menu. Then
        uncheck it.

Quote:>(Happily, I was running GNOME panel at the time, which meant that it's
>functionality remained...  wmx was pretty much "toast," though...)

--
Hardly. Microsoft has brought the microcomputer OS to
the point where it is more bloated than even OSes from          |||
what was previously larger classes of machines altogether.     / | \
This is perhaps Bill's single greatest accomplishment.    

        In search of sane PPP docs? Try http://penguin.lvcm.com

 
 
 

Thougths on the gnu project

Post by David Joff » Thu, 08 Oct 1998 04:00:00





> > >There is a project underway to develop a Qt-compatible replacement for
> > >Qt. It's called "Harmony"; theres a link to it from the main KDE page
> > >http://www.kde.org/ .

> > I hope that these guys can pull it off.

> I am not sure that I agree here.  Cloning a moving commercial target
> with volunteer work is not generally working out very convincing.
> Look at the Wine and Lesstiff projects.

Wine is different because it *has* to implement the 'moving commercial
target'. Once a Qt clone is developed, and KDE compiles with it,
KDE/Harmony no longer need to follow whatever changes happen to the
"real Qt"; they can even add extensions. Wine must always follow the
real Win16/32 family of API's.

 - David

--
-----------------------------------------
David Joffe  djoffe at icon dot co dot za
http://www.geocities.com/SoHo/Lofts/2018/
-----------------------------------------

 
 
 

Thougths on the gnu project

Post by Christopher Brow » Thu, 08 Oct 1998 04:00:00




>What they really need to do is get off of this bloatware
>kick, stop trying to come up with the uber library, stop
>trying to play Microsoft's game and try to draw more on the
>strengths of Unix (philosophically).

To be sure...

Quote:>Widget Libs, DnD & remote object libs should all be distinct
>entities with attempts to generate the ultimate one of each
>instead of the ultimate one of all.

I'll disagree.  There should be attempts to provide a couple of good,
interoperable options for each.

It is reasonable to hope for ones that optimize in different
directions, whether:
a) Optimizing for size,
b) Optimizing for speed,
c) Optimizing for flexibility.

Quote:>A lib to bring together all of the DnD standards as seamlessly
>as possible would be nice such that the OffiX DnD in WM and
>dfm can interact with the XDnD in Gnome/gtk and KDE/gt.

I ran dfm for a little while on Saturday; it looked pretty neat.  

The significant downside, from my perspective, was that it seems to take
over the desktop.  Menus normally provided by the WM seem to go away.
(Happily, I was running GNOME panel at the time, which meant that it's
functionality remained...  wmx was pretty much "toast," though...)

--
----==-- _                     / /  \
---==---(_)__  __ ____  __    / / /\ \
--==---/ / _ \/ // /\ \/ /   / /_/\ \ \
-=====/_/_//_/\_,_/ /_/\_\  /______\ \ \
A proud member of TeamLinux \_________\/


 
 
 

Thougths on the gnu project

Post by Roger Espel Lli » Thu, 08 Oct 1998 04:00:00




Quote:

>Wine is different because it *has* to implement the 'moving commercial
>target'. Once a Qt clone is developed, and KDE compiles with it,
>KDE/Harmony no longer need to follow whatever changes happen to the
>"real Qt"; they can even add extensions. Wine must always follow the
>real Win16/32 family of API's.

Except that:

1) KDE is not the only piece of software that uses Qt, and that people
   may want to compile with Harmony instead, and

2) Some of KDE's developpers are close enough to TT that they may not
   want to forget Qt and go for Harmony; hopefully they'll at least
   remain compatible, even if that means not using newer Qt features.

I have hopes in Harmony, but we'll see...

--

http://www.eleves.ens.fr:8080/home/espel/index.html

 
 
 

Thougths on the gnu project

Post by Klaus Schillin » Thu, 08 Oct 1998 04:00:00





> >Wine is different because it *has* to implement the 'moving commercial
> >target'. Once a Qt clone is developed, and KDE compiles with it,
> >KDE/Harmony no longer need to follow whatever changes happen to the
> >"real Qt"; they can even add extensions. Wine must always follow the
> >real Win16/32 family of API's.

> Except that:

> 1) KDE is not the only piece of software that uses Qt, and that people
>    may want to compile with Harmony instead, and

> 2) Some of KDE's developpers are close enough to TT that they may not
>    want to forget Qt and go for Harmony; hopefully they'll at least
>    remain compatible, even if that means not using newer Qt features.

Nothing prevents them from allowing dynalinking with either harmony and qt,
while offering contract supports/bug fixes only for official qt-linked
versions..

                Klaus Schilling

 
 
 

Thougths on the gnu project

Post by Karsten M. Sel » Tue, 13 Oct 1998 04:00:00



> 2) Some of KDE's developpers are close enough to TT that they may not
>    want to forget Qt and go for Harmony; hopefully they'll at least
>    remain compatible, even if that means not using newer Qt features.

As KDE *is* GPLd, it would be possible to fork development to track
Harmony rather than Qt, at such a time as Harmony is sufficiently
developed to merit production-quality products.

While I sympathize strongly with the "keep the core free" crowd, I also
have to admit that KDE is a slick desktop.  I would like to see
alternatives for the Linux desktop -- not *just* KDE, GNOME, fvwm,
afterstep, etc., but a choice among them (hot swapable at that).

--

    What part of "Gestalt" don't you understand?
    Welchen Teil von "Gestalt" verstehen Sie nicht?

web:       http://www.netcom.com/~kmself
SAS/Linux: http://www.netcom.com/~kmself/SAS/SAS4Linux.html    

  2:21pm  up 26 days, 14:10,  3 users,  load average: 1.43, 1.38, 1.36

 
 
 

Thougths on the gnu project

Post by j.. » Tue, 13 Oct 1998 04:00:00




>> 2) Some of KDE's developpers are close enough to TT that they may not
>>    want to forget Qt and go for Harmony; hopefully they'll at least
>>    remain compatible, even if that means not using newer Qt features.

>As KDE *is* GPLd, it would be possible to fork development to track
>Harmony rather than Qt, at such a time as Harmony is sufficiently
>developed to merit production-quality products.

>While I sympathize strongly with the "keep the core free" crowd, I also
>have to admit that KDE is a slick desktop.  I would like to see
>alternatives for the Linux desktop -- not *just* KDE, GNOME, fvwm,
>afterstep, etc., but a choice among them (hot swapable at that).

        KDE is only lukewarm. It may make a nice base for
        developers (if you ignore licencing) but as a whole
        desktop package, its down there with fvwm95.

        WM & E on the otherhand (or even icewm) are considerably
        more intresting (even if the rest is KDE).

        It's a shame gnustep is taking so long. I suspect if it
        ever gets usable, it will seem like the second coming
        (in comparison to K).

--
Empowerment implies having the power to do for yourself
what needs done and the necessary confidence to try.            |||
Crutches do not bring empowerment; they only encourage         / | \
the opposite.
You are the thinking machine & quite capapble
of much more than you may realize.

        In search of sane PPP docs? Try http://penguin.lvcm.com

 
 
 

Thougths on the gnu project

Post by Daniel Tayl » Wed, 14 Oct 1998 04:00:00


On Mon, 12 Oct 1998 21:28:33 +0000,


>> 2) Some of KDE's developpers are close enough to TT that they may not
>>    want to forget Qt and go for Harmony; hopefully they'll at least
>>    remain compatible, even if that means not using newer Qt features.

>As KDE *is* GPLd, it would be possible to fork development to track
>Harmony rather than Qt, at such a time as Harmony is sufficiently
>developed to merit production-quality products.

>While I sympathize strongly with the "keep the core free" crowd, I also
>have to admit that KDE is a slick desktop.  I would like to see
>alternatives for the Linux desktop -- not *just* KDE, GNOME, fvwm,
>afterstep, etc., but a choice among them (hot swapable at that).

So if the Qt license is such a big problem, and KDE is under GPL,
why doesn't someone port KDE to another library (GTK+, Lesstif, Athena,
Imlib, whatever).  The KDE developers have already released
under GPL and by the terms of the license cannot prevent the
port.

(You would think that someone would have said this before now)

--
Daniel Taylor  
Unix is a Linux-like operating system.