Advice on tools/technologies/books, etc.

Matt matthewcroberts at gmail.com
Sun Apr 13 03:48:52 CEST 2008


I would like to create a web-based tool for risk management. The tool
actually currently exists, but it was programmed in about 1998 using
old VB, etc, and we are updating it & moving it to the web. Basically,
as a first step, i'd like to create a basic web site that takes user
input, gets data from MySQL (i might pickle it, not yet sure) and then
runs some numpy routines & outputs the results. This will give us a
platform to develop the backend. My intermediate goal is to have an
asynchronous site in which as users adjust settings, the computations
are run & graphs updated. (the computations are pretty simple, btw).
My fantasy goal would be to combine that w/ google gears so the users
could use it online or offline, but that's probably just fantasy.

So, here are my constraints: I know Python & HTML (and lots of other
non-germane languages) but I don't know any javascript or ruby or XML.
What is the lowest cost path from here to there? I have been totally
out of the loop for this whole web 2.0 thing (I'm an economics
professor). Will it be possible for me to put together an async site
with only python? (I hesitate to use the term AJAX, b/c its unclear to
me how generic it is--do all async sites use javascript? can someone
clarify this?) If so, does it make sense to go ahead and start trying
to learn Turbogears or Pylons? Will they be able to create async
sites? Is there an easier way to do it? (Easy defined as me not having
to learn a 7th programming language) I have looked at Spyce, and that
seems an easy way to do the basic (step 1) site, but its not at all
clear that I can do async with it. CherryPy looks like it has a
steeper learning curve, but it also appears that the route to async is
clearer.

I know where I want to go, and I know what I can do now. I don't mind
getting deeper into Python, but I'd love not to have to learn a bunch
of other languages if I can avoid it. Any thoughts/comments?

TIA,
Matt.



More information about the Python-list mailing list