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

Re: Motif v lesstif - was Re: Netscape Communicator 4.01b6 ...



On Sun, 6 Jul 1997, Alex Yukhimets wrote:
>
> and david teague responded
> > If one has the library call interface and specifications, it seems to me
> > that it should be possible to write a library that is functionally
> > identical to Motif. If this is not true, I'd like for someone to explain
> > to me why.
> > 
> > --David

Alex:
> I guess by functionally identical you mean binary compatible.

Yes, I do mean binary compatible.

> OK, one rude example. If among the set of library functions we also have
> some global variables (for completely internal use, not documented), they
> have to be linked to your program at compile-time, not run-time. 

Isn't Motif available in shared libs?  If so, this scenario might not be
the case. How do we find out about globals?

Phoenix wrote legal code for the IBM BIOS by a clean room techinque, one
set of engineers examining the code, writing specs, and another, clean,
set of engineers wrote code. WE could do this, but I think at this time
it is nearly impossible for most of us to contribute that much time.

> But even the goal of source code compatibility with Motif I consider
> unattainable (without examining Motif source code) due to the nature
> of Xlib - the layer behind Motif. 

> Each call to its routine changes
> internal state of Xlib. If the same Motif library call implemented even
> by different order of Xlib calls in Motif and Lesstif, the internal 
> state of Xlib might very well be different after this call. If the
> programmer also uses direct Xlib call after calling some Motif routine
> (absolutely legal thing), the result of this call may also be different
> in Lesstif and Motif implementations, which immediately breaks even source
> code compatibility.

I think this could be done by reverse engineering Motif. Is that legal?

At any rate, I HOPE you are wrong! I am willing to bet that the first
scenario is not valid, but that the second may well be a problem,
adn hope you are being unnecessarily pessimistic!

At any rate, thanks for the insights!

--David
-------------------------------------------------------------------------
           LINUX: the FREE 32 bit OS for [345]86 PC's available NOW!
David B Teague | User interface copyrights & software patents make 
teague@wcu.edu | programing a dangerous business. Ask me or lpf@lpf.org



--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-user-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .


Reply to: