IL2007: Mashups and data visualizations

Darlene Fichter (who is wearing an M&M costume)

(I like listening to Fichter talk — she’s personable and good at reading her audience, in addition to the fact that she generally and regularly gives good, organized, useful presentations with a good mix of examples and analysis. Theory and practice, and presented in a way that gets you motivated to learn more.)

Web 2.0 changes what we can do on the web — it’s a read-write world, now, without the need for a huge stock of tech knowledge. Also, customization is suddenly supremely easy, again without tech knowledge. “Sometimes you have to register, point, and click. If that’s a barrier for you, then mashups probably aren’t for you.”

Definition: mashup is a web application that uses content from more than one source to create a new service, typically sourced via an API or an RSS feed. “People can take the same ingredients and produce different things”.

Mashup ecosystem: Relies on open content, and open services. “A lot of what libraries are all about is making content be open and available to everybody”, so isn’t this a natural extension? We can be a source of data, we can be creators of data, and we can be users of data.

Programmable Web Mashup Dashboard — great place to track this phenomenon. Maps are huge, photos are big, and so is shopping. And bunches of other stuff — “Celebrity mashups”, etc.

Mashups are among the fastest growing ecosystems on the web, and so what’s that mean for us? Well, we’re probably using mashups on a regular basis. “One of the most popular mashups in Toronto is where the beer stores are, and whether or not they’re open now.” Which is all data entered by users, and then used by users. [No approval necessary to identify an information need, approach a solution to the information need, and then solve the information need through access. O noes, they iz stealin’ our authority!]

Some big mashups:

  • Maps are huge, in large part because of Google API for its map files
    • Frappr maps.
    • Google My Maps. Easy to use and personalize, and can be embedded in other sites.
    • Chocomap.com, coffee maps, bathroom maps, etc.
    • Western Springs History, from Thomas Ford Memorial Library — locations of important historical events

“90% of what’s out there has been repurposed and borrowed from someone else”

Yahoo Pipes — tutorial, tools, etc. “Best way to learn how to use them is to find one you like and clone it and then make it your own”

“This wins the pretty prize for mashups” Using Syndetics book cover data and combining into a slider visual that shows the top 20 new books. “I would prefer to show the last 20 books returned, since they’ll actually be on the shelves”. Hah.

Unintended consequences of mashups: “Garbage in, garbage out”, because the mashup is only as good as the data going in and the programming processing it.

Visualizations are side-effects of mashup culture in information — we can now take data and make a visual display of it. Population maps, website structures, etc.

As she was demonstrating one nifty data visualization, she said, “it’s got a lot of text, so if you like text, it’s great, and librarians like text. I don’t know that our users like text so much.” [Ya think?!]

Social sites for data visualization — allows new kinds of data analysis, will probably attract curious and serious statisticians and citizens. Swivel — “Tasty data goodies”. IBM Many Eyes. Gapminder/Trendalyzer.

“Metadata got hot and suddenly catalogers were cool.” LOL.

Leave a Reply

%d bloggers like this: