A Python application server?

Duncan Smith buzzard at urubu.freeserve.co.uk
Mon Jun 2 11:49:54 EDT 2003


"Andrew Walkingshaw" <andrew-usenet at lexical.org.uk> wrote in message
news:slrnbdmq44.i3p.andrew-usenet at athena.jcn.srcf.net...
> In article <bbfmfo$ulf$1 at newsg2.svr.pol.co.uk>, Duncan Smith wrote:
> > Hello,
> >          I'm trying to put together a (basic) design for an online
> > examination system.  The system already exists, but is based on Java
> > technologies.  I have (probably prematurely) assured the author that the
> > system could have been developed in Python.  This is based in faith
rather
> > than reason, as I have never had cause to consider such a thing before
> > (although I can now envisage that I might need to develop a simple
> > application server within the next year or so).
>
> In terms of prior art, I and a fellow student, as holiday employees,
> wrote an exam server in Python (mod_python/apache, postgres backend),
> conformant to the QTI Lite spec (http://www.imsproject.org/ - basically
> multiple-choice visual/audio/textual questions), in about three months:
> we didn't have strong authentication or any sort of eavesdropping
> protection, though (we were doing everything via unencrypted HTTP). This
> was meant to be released as BSD license software at some point, but it
> never appeared. :(
>
> This makes me believe more experienced programmers could put together a
> rather better system in the same sort of timeframe.
>
> Is there any reason (type of question, etc) why you can't aim for a
> web-browser based solution on the client side, using SSL authentication
> or similar to counter MITM/eavesdropping/impersionation attacks?
>

Not as far as I'm aware.  I don't see the need for anything like RMI +
applets (but I'm *far* from an expert in this area).  If there's no
relatively pure Python equivalent, I can live with that.  But I'd like to be
armed with some good reasons why we don't need it :-).  I'm not a computer
scientist (and of course my associate is) so I'm guessing he might be able
to come up with some 'problems' I haven't even thought of.  But he's a
reasonable guy, so I reckon he could be convinced.  Cheers.

Duncan






More information about the Python-list mailing list