Eiffel better than Python ?

[Tim Peters]

[Tomasz Stochmal]
> Should I stop using Python and jump on Eiffel wagon ?

[Tim]
> You didn't say what you want to accomplish.  If the idea of "provably
> correct" programs appeals to you, Eiffel will give you more help than
> any other practical language I know of.

[Alex Martelli]
> I still think Haskell would be of higher help.  Immutable data
> make proving program properties easier (while not _easy_).  (If
> you're implying Haskell isn't "practical", I'll unchain the
> slavering hordes of FP groupies against you... scary threat,
> innit?).

Consider the questioner.  Eiffel is an embodiment of a formal theory of
software construction, and has many language features specifically designed
to make application of that theory straightforward.  Thomasz likely doesn't
know either what to prove or how to go about it, in a vacuum; most
programmers wouldn't; Eiffel helps them by providing a well-documented
theory and executable framework; doing it in Haskell instead requires
sucking the theory of functional programming out of scattered papers,
building the proof support machinery by hand, and then persistent commitment
to use it.

Immutability isn't really much help when the *interesting* questions are
about how dozens of subsystems interact with each other; documented and
automagically enforced preconditions, postconditions, and class invariants
do help, and in large part by forcing the programmer to *think* about them.
If you try to do something similar in Haskell, you have to bring the
motivation and support tools with you.  DBC never catches on in Python for
much the same reasons, I fear.

too-clumsy-==-won't-be-used-except-by-motivated-experts-and-those-are-
    the-ones-who-need-it-least-ly y'rs  - tim