Wave thoughts from a physics student

0 views
Skip to first unread message

Nick Stenning

unread,
Jul 20, 2009, 3:57:55 PM7/20/09
to Knowledge Waves
Hi all!

I don't know anyone round here yet, so "Hi!" ... I would say *waves*
but then you'd know how bad my sense of humour is.

Anyway, I don't have a Wave account yet, so I can't really say how
well the following idea would work, but I've just been pondering what
would be useful to me as a science student.

# Working-out "notebooks"

When one fires up Mathematica, the read-eval-print-loop that runs
isn't solely an evaluation environment. It's also a notebook that you
can save if you decide that your scribblings are of any worth. It
strikes me that Wave would be perfect for similar functionality,
namely:

- line-by-line evaluation of calculations, by something like the
Wolfram Alpha API [1] or Google Calculator (no API, unfortunately, so
that'd have to be scraping).

- additionally, the ability for the remote service to store the
results of each line, for reuse in follow-on calculations.

- the ability to annotate the calculations with comments, and also to
hide answers and recalculate them on request, for use in a teaching
environment.

(As an aside, how cool would it be to have scientific papers shipped
with raw data, and the graph-drawing and statistical analysis codes
shipped also, so that they can be looked at by anyone looking at the
data or its visualisations? A way out of the PDF issue, in the long
run? http://bit.ly/12H838)

I have a question for anyone that does know about such things: how
would one get structured data out of a Wave? Would this be a matter of
sending the contents of the wave to a remote server for processing, or
is there any support within wave for parsing predefined formats (HTML
tables, RDF) into, say, CSV or JSON or YAML?

Regards,
Nick

[1]: http://www.wolframalpha.com/developers.html

Oliver Hofmann

unread,
Jul 20, 2009, 5:25:24 PM7/20/09
to knowled...@googlegroups.com
> (As an aside, how cool would it be to have scientific papers shipped
> with raw data, and the graph-drawing and statistical analysis codes
> shipped also, so that they can be looked at by anyone looking at the
> data or its visualisations? A way out of the PDF issue, in the long
> run? http://bit.ly/12H838)

Yep! A literate programming framework such as Sweave
(http://en.wikipedia.org/wiki/Sweave) would be excellent to have. Not
going to happen for quite a while due to differences in formats and
algorithms, but for some data types it might be quite feasible.

Best, Oliver


--
Research Associate Department of Biostatistics
Associate Director Bioinformatics Core
Harvard School of Public Health
Skype: ohofmann Phone: +1 (617) 365 0984

Cameron Neylon

unread,
Jul 24, 2009, 2:35:58 AM7/24/09
to knowled...@googlegroups.com
Nick

Hi, sorry for the delay in replying. Good to have you on board. At one level
what you're talking about here is actually very easy. The idea of having a
Robot that responds programmatically to what you type is almost trivial.
There may be some issues with needing to enforce a single flow of commands
and responses but I'm sure that could be handled. The difficult part here is
connecting it back to a web service that actually processes the commands.

Having said that, there is already a Robot (which I haven't actually got
working myself yet, but I've seen it in operation) called Monty which
interprets Python commands and gives answers. If there is a web service that
can be polled with a command and returns the answer then the rest is pretty
easy.

Shipping papers with data and live charts is exactly one of the things I am
most excited about. For me this is important on two fronts, firstly that it
make live manipulation possible which I think is exciting but probably more
importantly it provides the history of what was done to the data and where
it came from.

Cheers

Cameron
--
Scanned by iCritical.
Reply all
Reply to author
Forward
0 new messages