I was hoping to enter; indeed, I got about 1/3 of the way through my project but
sheer lack of time and expertise has got in the way.
The main problem was parsing the XML. I'd developed the beginnings of the
interface, and I was about to approach parsing the XML. Unfortunately, I ended
up writing the application in Python which is not quite my forte (ie: I
practically knew none of it when I began). The difficulty of parsing the XML
was really what was holding me back - that's why I asked if anyone had
developed a TV-anytime parser for Python.
Anyhow, I may as well blow open what I was working on because the deadline's
passed: an AIM chatbot that you could talk to in natural language eg ("what's
on bbc1 or bbc2 between 2030 and 2230"). The bit I've done is most of the
language-parsing and creating the userid (the bot is called whatsonthebbc). The
bit I haven't done is parsing the XML.
It's not even really the XML that was tricky - my XPath syntax is competent, and
that's all I'd need; it's just handling the gzip'd file, decompressing to a temp
directory, and joining the dots between three seperate files for each channel.
I can see the advantages of TV-Anytime, no doubt, but the file-handling was just
getting a little beyond me.
Still, if anyone's developed a TV-Anytime parser in Python, that'd ease the
cogs.
Also, as I said, the main constraint was time - not the length of the deadline,
but merely my summer being busier than I thought.
Still, I've discovered that I rather like Python as a result.
t.
Quoting Ben Metcalfe <[EMAIL PROTECTED]>:
> Dear all,
>
> I'm writing to let you know that the inaugural backstage.bbc.co.uk
> competition hasn't gone as well as I had hoped. In fact, at the time of
> sending this we haven't received any entries at all.
>
> backstage.bbc.co.uk is very much about the BBC experimenting with new
> ways of engaging with it's expert user base, and clearly this specific
> exercise hasn't worked. backstage.bbc.co.uk also strives to be a
> publicly open and transparent project, which is why I am writing to
> communicate this to you all.
>
> Moving forward, I've been trying to think about why this has happened -
> and my guess is that it comes to one of two possibilities:
>
> * The TV schedule data we provided over-complicated and in an alien
> format that was difficult to parse, or
> * The idea of developing around a BBC-led theme, even for a prize, isn't
> an approach that is of interest to the backstage.bbc.co.uk community.
>
> I'm keen to gather whether either/both of these reasons are the case, or
> maybe there's something else I've completely missed?
>
> All of your thoughts and views are very subject are very much
> appreciated, so I'd be really grateful if you could let me know what you
> think - either publicly on this mailing list or privately (ben.metcalfe
> [at] bbc.co.uk).
>
> I don't want to pre-empt your views on this, so I will get back to you
> with some more thoughts and action points on my part, once I am able to
> gauge where we stand (and thus what we need to do differently next
> time).
>
> Many thanks
>
>
> Ben Metcalfe
> Project Lead, backstage.bbc.co.uk
>
>
>
>
-
Sent via the backstage.bbc.co.uk discussion group. To unsubscribe, please
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.
Unofficial list archive: http://www.mail-archive.com/[email protected]/
