[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: information <--



> > > r-to-l support to X-Windows is not possible without significant changes
> > > and additions to X-Windows protocol.
> > Eli Marmor does it for a living. not at the X protocol's level, but at the
> > level of the MOTIF library. he can do it for LessTiff if he had the
> > funding I suppose.
> 
> Will the hypothetic sponsor then release the source? Will Eli permit it?
> Doesn't Lesstif has GNU license, which inhibits proprietary works based on
> it? Though maybe if it doesn't require Lesstif modification, it may still
> be possible. Theoretically.

Relax. It was not my idea, I was not the one who raised it (although
I really appreciate Ira and what he did for us and for Linux), and I
don't have any intention and/or time to do this project. Moreover, I
don't believe in this idea. Although I am aware of the drawbacks of
Motif, Lesstif is so inferior in relative to it, and is not going to
replace it seriously in the following years. And please, before you
argue automatically, look at the sources of at least libXm, even not
the whole ~10 MB, but at least one module, and try to understand the
complexities. Otherwise, you will be exactly as one who attacks Ori
Orr without reading the original interview in HaAretz (hey, don't
flame me, I read the original interview, and besides - I am half
Sepharadic :-)
And even if the developers of Lesstif would be more brilliant than
the developers of Motif (and they are not - believe me - I know most
of the Motif original developers), it is not the same to develop
something and to develop a clone. In order to develop a 100% clone,
you must be 10 times more genious than the developers of the original
stuff. This is the reason why Wine will start to run Win 3.11 apps
when Microsoft will replace Windows 98 by its successor - Windows
2000 (I know, Wine already runs some 95 applications - even Solitaire
and MineSweeper! ;-).

Contrary, I see that another one (Gal) understood my point EXACTLY.
Lesstif is totally different than Motif, and of course my code is
very specific to Motif. You cannot drag this code into Lesstif. All
the data structures are different, the functions and the methods
are different, and developing Hebrew for Lesstif is like beginning
all the work from scratch, with years to pass till the code will be
stable enough:

> We can't drag Eli to all discussions about Hebrew support, Eli CAN NOT write
> code for LessTiff. He is contaminated by the Motif Code, his code if he ever
> write it, will never make it to the LessTiff source tree!

Well, "contaminated" is not the reason, but the conclusion is same.

> So, you got his Hebrew fonts, what more can you ask from him? 

The rest 98% of Hebrew support?

> GTK & QT are planed to have support for r-to-l, let's hope it will
> be better then motif 2.1.

Oh, Gal, I was so happy that somebody understood me, but you spoiled
it in your last words, or you are too optimistic: In order to have a
good Hebrew support, you must invest much time (much more than the
staff of QT), much experience, having various unique technologies,
and be an Israeli with expecience with the various Hebrew supports.
The people there, even after having the documents of Unicode, don't
understand the language and its requirements and behavior.
And even a BiDi (a.k.a. r-to-l by some of you) support which millions
of dollars were invested in it, staff of 60 developers, unique
technologies, and much experience, is too less for you (I referred to
Microsoft's BiDi support, if it was not understood...). So what can
you expect from the QT?  I can promise you that it will be a complete
disaster, much more than Motif 2.1  :-(

And to another quotation from other one:

> Go for Gtk. The only free serious application using Motif that we
> are interested in is Netscape/Mozilla

So please, stop mention all the big names who decide to move to Linux!
Don't be proud that Oracle moves to Linux. Don't be proud for Sybase.
Don't be proud for Informix. Don't be proud for CA/Ingres. Or for
Interbase. Or for many other new players in the Linux field. All of
them are going to use Motif (for their front-end, of course; The back-
end doesn't need any GUI). None of them will use QT or GTK. or even
Lesstif. I know them. Porting is very easy and safe, as long as you
don't replace key components such as your GUI toolkit. Replacing the
GUI toolkit of your package, causes the porting to be 10 times to 100
times harder, with beginning the whole process of early unstable
versions etc., again.

Chevre', we (i.e. Linux) finally get attention, please don't spoil it!
None of these companies is taking Linux seriously just because of QT
or GTK or Lesstif. Moreover, one of their fears is that after having a
Linux version, users will press them to support one of these toolkits.
I know them: many of them are among my customers, and believe me, they
love Open Source and Linux, but don't love extreme evangelist users
who only cause damages for Linux chances to beat the "mainstream"
operating systems. Did you know that while we waste time for argumets
which only cause damages to Linux, Oracle Israel donates 3,000 (three
thousands!) NetBSD based computers to Israel?  The Vice President of
Oracle visited Israel and met a few days ago with PM Netanyahu and
Minister Eitan (yes, these two who were blamed for letting Microsoft
occuppying Israel...) and agreed to help the Network Computing in
Israel, and maybe even to stop Microsoft from taking the whole
country (as Eitan said before: You are against Microsoft step?  No
problem! Join them, and donate YOUR technology too; We will be glad
to have a party and/or annoucements with you too...).

There are thousands of Motif applications there, hundreds of them are
key components for the success of Linux, and nobody is going to port
them to GTK/QT/Lesstif. I know, there are some dozens of GTK/QT
applications, and there are even one serious GTK-based and one
serious QT-based. But what is ONE vs. hundreds of key applications?

I am very concerned. We have a great opportunity, but becuase of
our "ideology" we may miss it. And who is this "enemy" that we all
try to fight?  Motif???  This open software which was developed by
the (previously known as) "Open Software Foundation" as a contra
against proprietary code (Microsoft and also - in the past - Sun)?
The first one to develop an open and PORTABLE toolkit?  The first to
give source code for cheap prices?  (just try to ask Microsoft to
sell you its sources for $2000. Or even $200,000). What will be the
next step?  To fight QT?  You probably didn't read what Linus had to
say about all these stupid wars. And I agree with him. These stupid
wars destroyed UNIX, and will hurt Linux as well. So please, don't
saw off the bough that we are all sitting on.

> 1. Get the Unicode book.

I have it. So what?  If it would be thrown at the face of The Bill
(instead of the cream pie), the status of all of us would be much
better, except for the assacinator who would be executed for the
electric chair. But using this book will not make the development
instead of you. It will only save you the definition of one stupid
algorithm. About 0.1 prumil of the work.

Regards,
-- 
Eli Marmor