Presentations + Pretty Things: metamodel for library data

It has been a long lapse since my last post and this lo-fl blog is becoming borderline lo-fun. Summer has brought sunny weather and a super-mix of exciting events for me: EVA in Firenze, THATCamp in NYC, Eyeo in Minneapolis, and finally ALA in Anaheim (from where I’m currently crafting this posting). So in honor of Alan Turing’s b-day  * h1p_h1p_H-00-RAY* I am posting a much belated update. As a follow-up to my talk last month at EVA Florence on Preservation Metadata for Electronic Art Installations (#mouthfulofnerdy), I decided to author a summary of my recent program adventures from sunny and swell Firenze, followed by some comments on good design for library data. As per uszhe, I included lot of planets and pictures of space to power you through what promises to be a long posting.

Firstly, my presentation discussed a portion of my thesis topic on preservation Metamatics pdf coverpagemetadata and conservation practice for new media and, coupled with a decidedly New York-centric focus (as was encouraged since I represented ½  of the American constituency in attendance). As I promised to orchestrate a slide-share of sorts, I’ll post an abbreviated but attractive version of my slides, minus some video, citation marginalia, and more advanced descriptive content as a pdf of a ppt. For those who dug my formatting (hello, I love you), sorry, but the Consolas + Calibri typeface magnifimashup does not translate with a downgrade some of my color spaces are also skewed; I think you’re seeing a florescent version of Calibri text.  But I’m sure the people skimming my resource page really could careless than I do about that.

Download the pdf hereEVAPresMetadata

Now that I’m over the conference hump (EVA + EYEO+ ALA ohmy!), this short pocket of blissful break from the blog has made me restless so I’ve decided to pick-up a few pbsdatavis01lost threads that I’ve left trailing along the way. This will be a post about the need for data visualization nerds to jump on the metadata and electronic media bandwagon. In recent news, a few informal speaking engagements within the library community have left me frustrated with the slow pace of cultural institutions in the face of data vis; these organizations have only the best data at their disposal. Modest projects for visualizing and presenting these data exist, to be sure, and are impressive, but a more ambitious push to participate in social visualization tools, or at least to enable these visual devs as open data APIs is needed.

The weighty rhetoric of the library and archives world (I’m looking at you “repositories”) loses a bit of the play and processing snazziness of partner projects in news and advertising environments. As libraries, we’ve become woefully slow on the uptake of participation in animation and visualization initiatives even when the libraries and the github links are there (hello Timeline.js, and Fusion Tables). There’s still a persistent gap between how libraries present and illustrate their data, and how corporate entities organize and manage product promotions by establishing an attractive online presence; libraries lack the visual. There’s a lingering wall of intimidation between the library world and the IA/Developer world, and we need to close that gap.

Case Study: Conference Sites

Look at the ALA-Anaheim conference website:

Now try to navigate.


I’m not going to hate too much, since criticism should beget solutions rather that pbsdatavis_take2whine and cheese. But in comparison to Eyeo [] and SxSW [] (which boast a pretty profound data set of events and information), the libraryland conferences are a cocktail of cumbersome and creatively-challenged. To soften the brutality of that screencap, I am also peppering this post with some pretty attractive data vis coming out of a pbs program you may have heard about: America Revealed.  Mapping arial data about internet usage and across the US produces some beautiful swirly planet patternings. Even the pizza delivery routes in manhattan looke brilliant in blue light:

This echoed some other swell projects I’ve been seeing including this map (author unknown…maybe Jer Thorp?) of weather eruption patterns over time, projected in pbsdatavis_take3circular plumes (best if viewed in Chrome):

The first lesson in Processing is an education in how to render circles with animation software:

…so starting out with creative code and attractive vis is not that difficult, and as pbs and git can attest is furthermore pretty appealing on a universal scope. People like ellipses and curves, there’s a particular attraction to bubbles and bobbles as nodes of information, as planetary points on our navigation through an interweb solar system of impacted data sets. Even static vis shares a love of swirls and circles. See:


We Feel Fine []

The Dumpster : []

Bloom/Hodgin’s Planetary app: []

Take the above as examples of that. There are ways to render library data in pbsdatavis_take3informative bubblesets, There are ways of illustrating the popularity of certain dewey  100 blocks visavis checkouts rates, ways of ranking library “departments” and charting them as a newsmap  (hint: dewey just became RDF-enabled:

There is so much potential bundled in library data, and so much feasible datavis to build. There is a programmatic learning curve for some software applications, and it’s unfair to demand better datavis without acknowledging the limited budgets and programmatic expertise that affects library projects; but if Code for America can mobilize volunteers to push  government data out of cmd-line prison, there must be a comparable movement in America’s libraries * call-to-action face*. And anyway, perhaps I am a bit biased, but I am BIG on pedagogy in programming, less on the pretense. Some of the most fascinating things I’ve learned have been through free online tutorials and ad hoc apprenticeships with my coder friends who knew way more than I did and were willing to chaperone me as I tumbled my way through projects; and that includes data projects conducted at library venues. We no longer occupy a world of lone-gunmen genius…more a collaboration space where several geniuses combine skills to build brilliant things. Computers, with their open-ended potential as platforms for n projects foster that kind of networked intelligence and collaboration in obvious ways. So let’s go libraries, let’s collaborate and create pretty things!

Last month’s THATCamp Museums touched on new ways to make collections visible; harnessing the power of the timeline to create visualizations of information that map content to concrete nodes in time, and visualize those points:

So as a wrap-up to what has become a long read, let’s just close with some symmetry and some Alan Turing shout outs []. Just because we deal with ugly data sets doesn’t mean our representation of that information has to be likewise hideous. Even Turing’s most impressive mathematical and computational outputs (including ACE among others) were the product of design. Arguably, everything within the range of human interface is the product of design, and there’s no reason to promote unattractive and empty data vis, no reason to de-sign our signifier-rich data in libraries. HBD Alan Turing, HBD.

LEGO Turing Machine from ecalpemos on Vimeo.

Check out the following and give visualization a whirl:

Tagged , , , , , , ,
%d bloggers like this: