Python *is* hyped, and many novice (and less novice) programmers will try and contribute stuff – I’m sure – if the system tools backends were written into Python.

but let’s try to imagine the same situation in the late nineties (pre- and nearly-dot-bomb era): instead of Perl you would have had sh scripting; instead of Python you would have had Perl.

what I’m saying – and what I tried to say in my blog – is that following the hype-du-jour is not an answer. if python gains us something *in the long run* I’ll be the first in line to say to ditch perl as the language of choice.

by the way: never though about replacing part of the perl machinery with gobjects implemented using the Glib bindings? that would make the s-t-b “scale” better, and would allow changing backends more easily. just a thought.

I also believe that is impractical to re-write g-s-t in python and throwing away the perl code just for the sake of having everything written in python.

i’m not so crazy about this mono-language motto we’re getting into nowadays. maybe it’s because my language of choice is the ruby underdog. anyway i hope g-s-t flourishes forward… it’s very useful as it is and rewriting seems a useless waste of energy…