Smithsonian Data Fun

I futzed with the code that Matt Miller developed, and all I got was this dumb word cloud.

What is this image? It's all the extracted names from within the "indexedStructured" section of the 11M open metadata records that were part of the Smithsonian Open Access event, placed into a wordcloud that resembles the right tower of the Smithsonian Castle, with (rough) weighting by font size for each name.

What I did was:

• run Matt's code to parse the 11M Smithsonian open metadata records
• extracted all the names found in all the records using pandas (but jq is also super fun and useful to use with the output of Matt's scripts)
• then I used Matt's fanart drawing of the right tower of the Smithsonian Castle as an image mask for a word cloud
• then I made a static IIIF representation of that image
• then I added it to Github and then I tweeted about it, and that's how we got here.

But seriously, here are some fun links of Matt's work and others that I was able to sew together in a few hours to fun effect:

• The Smithsonian Open Access release that started it all
• Matt's Tweet about his data analysis and data viz efforts that started it all more specifically
• Matt's blog post about his process and viz examples, also including fun easter eggs at the bottom which is where I got the fanart image of the Smithsonian Castle
• The Python wordcloud library I used
• The static IIIF tile generator I used
• My wordcloud as IIIF