Re: 'Selling' Eiffel



- Why (smart)eiffel? [I personaly like to see the generated toy-C-code,
Design by Contract, etc.]

Eiffel. Because there aren't so many object oriented garbage
collected,
native compiling (non VM), fast languages with generics (well in fact
there
was no other before the raise of D).

SmallEiffel was a good start but the move to SmartEiffel a total
desaster and the later
is complete unuseable because the compile time of serious programs (>
150-200 KLoc is insane).

Still using my own heavily patched (around 40% of the code changed/
replaced)
SmallEiffel clone. It works fine now but i had to spend all together
around a 3/4 year
full time work.

- What is (smart)eiffel famous of? [Coming from Smalltalk --> Everything
is an object?]

Having the most incompetent team with the most stupid design decision
in the history
of programming language. Breaking even simple things that the constant
'1' wasn't
an INTEGER anymore.

- What nice/strange stories can be told of the evaluation of the language?
- etc..

Sucked like OS/2 on marketing. Compilers were terrible and expensive
until late 90ies.
And it was to heavy to work on slower computers of this area
(compiling time not runtime).
Also GC was only real valued after the raise of Java in 2000 and at
this time
Eiffel was already doomed because Java kicked them out of the Business
like Smalltalk.

What story *must* I tell.

The story about DbC.

What features I *cannot* miss?

DbC. Good Generics and a consistent not feature overloaded language
design - it was called
a RISC language for a good reason.

Also talk about agents as a different flavour then closures (agents
are what i call
'call by value' Closures).

Eiffel is good but unfortunately there is no great future. Well for my
inhouse production
i don't have any intensions in the next decade to move away from
Eiffel (unless we have huge
changes in CPU technologies). But for all new projects you should
really look somewhere else
(my favorite for anything like this would be D 2.0 - 1.0 not useable
for me because they don't
have closures at the moment).

You can check out my program Arachno Ruby (http://www.ruby-ide.com) it
is written 90% in
SmallEiffel and 10% in C/C++. It's the only larger program and the
only serious GUI program
that exists in SmallEiffel.
.



Relevant Pages

  • Re: OOP Language for OS Development
    ... > Effectiveness on System Programming, ... Design by Contract - a process and a language, ... reasons why DbC alone puts Eiffel near the top of the list. ...
    (comp.lang.ada)
  • Re: OOP Language for OS Development
    ... > Effectiveness on System Programming, ... Design by Contract - a process and a language, ... reasons why DbC alone puts Eiffel near the top of the list. ...
    (comp.lang.cpp)
  • Re: OOP Language for OS Development
    ... > Effectiveness on System Programming, ... Design by Contract - a process and a language, ... reasons why DbC alone puts Eiffel near the top of the list. ...
    (comp.object)
  • Re: Newbie with some doubts.
    ... >> uses of OO language features, using a language that was designed ... He does tend to write like the Eiffel way of doing OO is the only ... >> you're liable to come back to Python and wonder "Why doesn't Python do ... design goals of Eiffel are different from the design goals of ...
    (comp.lang.python)
  • Re: ???
    ... It's about the defunct programming language Sather. ... It is not an off-shoot of Eiffel. ... There is a 99 bottles example in Sail on the same ...
    (comp.lang.sather)