[okfn-help] http://www.openeconomics.net/

[Rufus Pollock]

On 25 June 2010 14:26, Tim Hubbard <timjph at gmail.com> wrote:
> Hi Rufus,
>
> Another question
>
> what is the state of http://www.openeconomics.net/ ?
>
> I looks like its been sleeping for a couple of years.  Is it living or dead?

Basic answer: it's very much alive (though not all of it, the atlas is
bordering on comatose).

This is a great time to ask since we've just had someone else (Alex
Hagon -- in cc) getting actively involved with this.

To answer your question in detail:

  * openeconomics.net is actively monitored and being kept running and online
  * have not done as much work on codebase (or datasets) over last 9
months though did fix up http://openeconomics.net/hwi just last week
(browser for cia factbook at the moment but supposed to be a human
welfare indicator thing)
  * some material which was in openeconomics is now elsewhere e.g.
econdata in wdmmg repo
  * we've got lots of ideas for what to do here it is just finding
time to execute on them.

One reason for slight slow-down as I would like to migrate (and have
wanted to for a while) the core data store functionality into ckan.net
+ storage (basically generalize it). CKAN already has got a good
metadata stuff so why duplicate it? What CKAN lacks is storage. I/We
have been thinking for a long time about sorting this out generally
but I feel, right now, it would be better to do something quick and
dirty and get something running (which we could do *very* quickly)

Finding a way to run openeconomics.net data store off CKAN + Storage
would also beyou'll need to install the python app in usual manner
(README.txt covers this poorly) and then run paster webserver. If you
want to look at the modelling code (not that developed but there is
some) see stuff  a really useful demo of CKAN functionality in its own
right.

Several possible solutions for storage:

  * Build our own app: could build something to at least store simple csv
    * My original plan ~ 2-3 years ago but my feeling now is why not
just use google docs
  * Use mercurial repos for each data package: this is increasingly
attractive and what i've started doing personally and with wdmmg over
the last year. However heavy tech requirement on users.
    * this is what openeconomics does already but just in a single
repo -- see econdata directory
  * Use external, in particular just use google docs/archive.org/...
    * Increasingly inclined to this solution as a great "low
tech"/buy-off-the-shelf approach

At the moment working with Alex is a good example: he's uploading
stuff to google docs and we'll then probably created a CKAN package
and then find a way to integrate into the store from there.

> I poked around but couldn't really find much documentation.  How do you
> setup and use?

What do you want to do?

If you just want to start hacking on datasets check out a dataset in
econdata/ if you want to run the web frontend (econ/www) you'll need
to install the python app in usual manner (README.txt covers this
poorly) and then run paster webserver

If you want to look at the modelling code (not that developed but
there is some) see econ/model.

Rufus

PS: we're also thinking of starting an economics working group:
<http://wiki.okfn.org/wg/economics/> ...