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

Re: Python or Perl for a Debian maintainance project?

>>>>> "Andrew" == Andrew Suffield <asuffield@debian.org> writes:

    Andrew> What are you talking about? The *point* was that handling an
    Andrew> error is better than not handling an error.

I don't quite agree with even this claim.  Usually leaving errors unhandled
(and crash the program right away) is better, simply because there will be
less code to maintain.  Not every user will feel it better, but many will
once they understand the consequence.  If you have to choose, will you want
the software you use to have every frequently demanded features implemented
twice as quickly, and every frequently pointed bug fixed twice as quickly?
Or do you prefer them to be twice slower, so that you can write and maintain
code to catch every exception (or inspect every error flag), and give a
"correct" and "user-friendly" explanation to every possible error that
results from corner cases, (say having your computer crashed just when
saving your file, or when the user manually override a dependency so that
the package that you expect to be in the system is now not), and then find
the best reasonable way to continue the program execution?  Of course your
answer will depend on whether you have recently get such an experience, but
for all I concern I'll take for the former.

Put it another way, I want useful software, not full human-machine
interrogation software.  In general we are not creating software that is as
mission-critical to require those.  User will retry the failed thing, and
choose the best recourse anyway when it fail again.


Reply to: