llothar <llothar@[EMAIL PROTECTED]
> writes:
>>
>> >> I can't believe that's the only reason. There has to be more.
>>
>> > You have no idea what the community did for insane things.
>> > This was indeed enough.
>>
>> Well it would be nice what to know what you'd understand under the
>> community. I can not see the Eiffel users having playes havoc with it
>> but well those providing the "base functionality"
>
>
> Do you remember Uwe Sander and his wife? They tried hard to give
> SmallEiffel
> a GUI framework to make it useable for applications in there company.
>
> But Mr. Colnet and his team didn't care about it. They pissed on all
> the users,
> changed there compiler and breaked backward compatibility every two
> months. Just
> with the argument i like this better then the other (And they had
> always very
> questionable changes about what is better).
Well I do not have to check that but it does now show why the users
have broken it, you wrote Uwe wants to make a GUI toolkit but the base
tools was changed which broke their code. So what have this "users"
done to break Eiffel?
I'm quite aware about the way backward-compatiblity was treated. We
have seen it also. We once had added Eiffel support to our IDE and it
was to that time the only one I can remember which was able to debug
SmallEiffel on Windows. But that has not last very long....
>
> Always telling us that it is his research project and he don't care
> about
> anything them himself and his Loria friends.
>
> After Uwe was pissed off enough he left and Greg C and others who
> worked on ELK's
> gave up, because the Sanders were the workhorse of the project.
Again I can not see why the users have broken Eiffel. It's very easy
if the tools you use change in a backward-incompatible way you just
can keep up some time. And you just can keep up, with anything if you
get enough money out of your product.
So for me it seems, the problem is that there never was a market for
in Eiffel written tools. Well I just can say that Bertrand Meyer has
quite another opinion on how Eiffel could have changed that.
Howerver it seems that this is just another point in which one does
not just have to query Eiffel for it's values but what OO really
gives us. And for me this reflects back to Eiffel also. Till 2000
Eiffel has virtually no support for anything else but objects.
There was no abstraction which has existed since the 1950ies, the use
of functions. If you wanted that in Eiffel pre 2000 you simple did not
get it. Howerver the Smalltalkers got that right from the
beginning. They have the blocks which are extermly useful for
"functional abstraction"
And Eiffel could not deny that, I just propose to find anything about
blocks, function etc in Eiffel the langauge II or even Object oriented
software construction 2 nd Edition.
It was simply not there. Then there were things which were written
down in ETL II which never had worked in any Eiffel ever on the
market.
But then suddenly everythin has changed. They could not deny the
usefulness of functions any longer and invented "agents", and from
there on Eiffel as Language was history. Not because that was a bad
move but because not every Eiffel has got them. And well too all
Eiffelists before ETL II was the "standard Eiffel" but we do not have
this Standard any longer. We have a family of Eiffels which do work
differently have different tool support and without GOBO it would be
a hell of work to get some stuff portable between Sma(ll,rt)Eiffel and
ISE-Eiffel.
>
> I don't know about ISE and VE because they had never a public
> community. Seems there was
> just a little bit Gobo which is Eric's project and was opened IMHO
> after it was already
> too late.
It was never a problem to modify GOBO to your liking. We've done that,
and I just can say GOBO is one of the most excellent tools I ever
found in the computer business. The library just "works". And if
nothing else from Eiffel ever will be remembered. I for my part think
that the libraries were the best ever developed. And they are the
easiest to understand IMHO.
>
>> You can still try to "revive it
from"http://sourceforge.net/projects/visualeiffel/
>
> I could do a lot. If somebody decide to pay my living (please a little
> bit over social
> wellfare level). Without this it is just one more '0' and '1' sequence
> in the digital
> garbage dump called sourceforge.
Well yes you can see it that way. I just can tell we had tried to get
some decent Eiffel and wanted to pay for it. But it was not enough,
and so it vanished completly.
>
> Putting it under a BSD license would have given it a little more
> chance. But the attitude
> to dump it and make sure that nobody else can ever make money with it
> (aka GPL'ed it) makes
> me not even to think about it.
Well I'm not in the mood to discuss the "merits" of the GPL. I just
can see that the market for "development tools" on the most used PC
operating systems (Windows, Linux) has simply vanished. We once had a
flouring industry and people were willing to pay for tools and so they
got choices. today it is reduced to more or less
MSVC, Eclipse, gcc, gdb.
And the tool situaion on commercial unices is even worse (hard to
believe but try to get a decent IDE running under AIX). The things the
Unix vendors have tried have just vanished (who the hell knows why it
is named KDE?)
>
>
>> > and SmartEiffel is not an Eiffel anymore.
>>
>> Well I guess the SmartEiffel developers will disagree, and tell you
>> that they have the only true Eiffel.....
>
> Yes, unfortunately i'm pretty sure that you are right here.
Well one of the biggest problems we had and have is that (but that's
true for other languages also) is that the market was and is small but
the competition harder then nearly everywhere else. And so man decades
were spend on parallel development. All the Eiffel vendors could have
spared a lot of money, while helping there users developing portable
libraries. But neither side has grant the other side much good will.
And here we are....
Regards
Friedrich
--
Please remove just-for-news- to reply via e-mail.
|
