[Open-data-census] minor points regarding ease of access to OKFN index data

[Rob Davidson]

Dear OKFN,

I recently made a very light investigation of how the OKFN index of
countries correlates with other indices of open/good governance,
specifically Transparency International's Corruption Perception Index and
Reporters Without Borders Press Freedom Index.

I have a small slide show that I presented to Open Data Hong Kong, here:
http://slides.com/bobbledavidson/transparency-opendata

My points to be raised with you are related to how hard/easy I found it to
access the collated index itself.

Both Transparency International and Reporters without Borders provide their
final index as a .csv file, alongside their methods (in .pdf) and their
raw, unprocessed data (to varying degrees, .csv). These are linked to
directly from their main results page, making them very easy to find.

Although OKFN has the raw data available on datahub, I found it hard to
find due to there being no link from the country index page. The source
code for the whole website is available via a link from the country index
page but not the data. I found it hard to find the link, 2/3rds of the way
down the About page. Also, there is no direct access to the resulting index
itself. The raw data is useful if you want to recalculate the index from
scratch but I only wanted the final score for each country. Having this in
.csv would have been very helpful.

Of course, it's fairly easy to scrape the index from the website but my
initial attempts to copy and paste the table into a slightly old version of
Excel failed as it stumbled over the list entries in every cell. Some
people may not be so happy to resort to HTMLIMPORT in Google Docs (although
copy and paste even works in Google Docs).

My point being, I feel that it would help users if there was a clear link
to the data from the index page and if there was a .csv/json format version
of the final index table.

The only other point that I would mention is that it would also be helpful
if you would include a column that linked each country name to its ISO
short code. It took a bit of work to match up the country names across all
3 indices. Transparency Int. included the ISO short code but for 2013
neither you or Reporters without Borders did. South Korea, Korea (South)
and simply Korea was my favourite variation on country name... I note that
Reporters without Borders have added the ISO code into their 2014 results,
it would be very useful if OKFN did too.


One final suggestion that is not related to data access is: would it be
worth contacting Transparency International to see if they would be
interested in making use of the OKFN index in their collated 'perceived
corruption' index? Currently they collate about a dozen different indices
from economists, NGOs, the World Bank etc. to produce their final index.
I'm not entirely sure if OKFN's index fits in that list but there may be
some sort of collaboration that would work.

The general discussion here in HK after my presentation suggested a
consensus that Transparency was a 'political' decision whereas Open Data
was a technical one and that it would help Open Data as a movement if it
was distanced from the political question of what to make available,
national security, personal privacy etc. However, it was generally agreed
that Open Government could make use of Open Data to enhance transparency,
even if Open Data did not rely upon Open Government.

My thoughts being that Transparency International has a decent reputation
and may reach an audience that has yet to care about Open Data yet. Perhaps
some sort of cross over may help spread interest in Open Data.

Many thanks for all the hard work you do at OKFN.

Best wishes,

Rob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.okfn.org/pipermail/open-data-census/attachments/20140523/2cbd2203/attachment.html>