tategallery/collection

Acquisition Data

zenlan opened this issue · 10 comments

zenlan commented

Thanks for a great initiative and dataset!

Wondering if it is possible to enhance the data with acquisition info. I was hoping to produce some stats about purchases, loans, bequeaths, disposal etc.

I can't see any such info in the json files, although I have only sampled a few random ones.

richbs commented

Greetings, @zenlan. Thanks for checking in! The only acquisition information we have is in the creditLine field in both the CSV and JSON. It's a free-text string so you'd probably be doing a bit of munging to generate stats but I hope it's something.

If it's missing in any cases, please let me know and I'll investigate. Here's an example where it shows up:

https://github.com/tategallery/collection/blob/master/artworks/n/007/n00762-13492.json#L21

Thanks again for your message!

zenlan commented

Thanks richbs, the creditLine is proving to be of some use, it does help that the acquisition year is always at the end of the (trimmed) string.

Have found only the following records to have an empty or null creditLine:

https://github.com/tategallery/collection/blob/master/artworks/d/415/d41533-117313.json
https://github.com/tategallery/collection/blob/master/artworks/d/415/d41534-117314.json
https://github.com/tategallery/collection/blob/master/artworks/p/060/p06007-545.json
https://github.com/tategallery/collection/blob/master/artworks/t/118/t11893-78181.json

The small number of nulls is hardly an urgent issue for my purposes however. :)

zenlan commented

Spoke too soon about the acquisition year "always" being at the end of the creditLine. ;)

"Purchased 200303" is a typo in this one http://www.tate.org.uk/art/artworks/schutte-no-title-p79033

richbs commented

Hello @zenlan! Looks like we will have the acquisitionYear as a separate field in the next release of data. Watch this space.

zenlan commented

Fantastic! :)

zenlan commented

That was quick! Meanwhile, just in case you're interested, this is what it looks like with the old 'munged' data... http://zenlan.com/tate/rickshaw.html

EDIT: old 'munged' data now replaced with new data in the above link

richbs commented

Those are some tasty graphs. I have finally updated all the JSON records with acquisitionYear too, in case you were using those. I see you're using Rickshaw for the front-end, what's powering it server-side, if you don't mind me asking?

Let me know if you're happy for me to link to your charts from the README.md.

zenlan commented

Imported the CSV data into MySQL, a few lines of PHP extracts the records to JSON files which feed the Rickshaw JSONP graph. (Could just as easily use Rickshaw.Graph.Ajax)

Sure go ahead and link to the graph, I'll try not to break it :)

Will stick the scripts and json in a repo at some point soon. They need a tidy up, embarrassingly hacky atm.

tatedev commented

Thanks for sharing!

Richard Barrett-Small
http://richbs.org
@richbs

On 9 Jan 2014, at 23:21, Zenlan notifications@github.com wrote:

Repo created here... https://github.com/zenlan/tate-data-doodles

—
Reply to this email directly or view it on GitHub.