[Open-data-census] Questions about Index Methodology

Tue Mar 25 17:02:06 UTC 2014

On Tuesday, 25 March 2014, Greg Michener <greg.michener at gmail.com> wrote:

> Nice going on the Open Data Index - very useful to have these sorts of
> metrics, especially because some countries tout their `open data` without
> really being that open i.e. Brazil - Where I'm based.
>
> I had three queries for you:
> 1. I am wondering why there is no scoring for the last three items at
> https://index.okfn.org/about/#criteria
>

Those are not criteria but additional information we collect. We perhaps
shouldn't have them in that table!

2. I'm also wondering why 'open licensing' is worth an inordinate score.
> You can still use the data, generate some great facts and figures and
> have those facts and figures published. Available in bulk, on the other
> hand, is only worth 10 points, but without being able to process large bulk
> datasets, open data becomes hugely limiting...
>

There is always going to be some debate about scoring as it isn't a perfect
science but the logic was as follows:

- There was a 50:50 split of points between what one could crudely term
"access" and "reuse" (access was questions likely openly licensed,
available for free whilst reuse was "machine readable", bulk, up to date
etc).

- Openly licensing ~= bulk + machine readable so in fact weighting of legal
and technical aspects pretty close

> 3. Isn't it silly to attribute points for governments having data (any
> data) in paper form? If it is a precondition for scoring at all, why this
> question itself included in the score? Same goes for number two - if it's
> not in digital form, you might just call it `active transparency` and not
> open data, correct?
>

Idea of score was that it built up. We could have given zero for some items
but we felt it was worth giving some credit. Remember, open data as per
open definition requires *all* criteria other than last one (up-to-date /
timely - which isn't an aspect of "openness" per se but is an aspect of
good data provision).

> I mean no offense, but I do think a healthy debate might improve metrics.
>

None taken at all - critique and debate is the way to improve things :-)
Thanks for sharing your comments and thoughts and keep them coming!

rufus

>
>
> --
> Gregory Michener
>
> Brasil +(55) Work (21) 3799-5778  |  Cell (21) 8231-7777  |  Home (21)
> 3576-4116  |  Twitter @gregmichener  <https://twitter.com/gregmichener>|  skype greg_michener  |
> http://gregmichener.com  |  http://observingbrazil.com |
>

-- 

*Rufus PollockFounder and CEO | skype: rufuspollock | @rufuspollock
<https://twitter.com/rufuspollock>The Open Knowledge Foundation
<http://okfn.org/>Empowering through Open Knowledgehttp://okfn.org/
<http://okfn.org/> | @okfn <http://twitter.com/OKFN> | OKF on Facebook
<https://www.facebook.com/OKFNetwork> |  Blog <http://blog.okfn.org/>  |
 Newsletter <http://okfn.org/about/newsletter>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.okfn.org/pipermail/open-data-census/attachments/20140325/84d6787f/attachment-0001.html>