Archive for the ‘Libraries’ Category

Forte: the National Library of Australia’s sheet music collection iPad app

Tuesday, March 26th, 2013

A man walks into a reading room, hands over his iPad and says “Hi, my name is Jake and I’ve built this”…

Forte iPad

Let’s go back in time

In March 2011 the National Library placed a dataset of our sheet music collection up at data.gov.au. This was to be used as one of the datasets for the LibraryHack competition. The dataset contained around 11,000 items from our sheet music collection, all the metadata, all the details about every page within each piece of sheet music and most importantly references to each image of the digitised page. This was a bit of a handcrafted dataset. Our regular API’s, like the Trove API, will only return information for the “top level” of the item, not the lower level details of every page.

We didn’t quite know what would happen with the dataset & how it would end up being used in the competition. Unfortunately it didn’t really get used in a major way in any of the entries and the dataset sat there gathering electronic dust.

Nearly 12 months later, totally out of the blue, someone walked into the reading rooms at the library, approached the person on the desk (who, as luck would have it was Sarah who was the project manager for our iPhone catalogue app), introduced themselves and said “Hi, my name is Jake & I’ve built this” & showed off what was the first prototype of what went on to become Forte.

At the time Jake was looking for a large dataset to help him solve a problem he was working on. He found our sheet music collection & built a proof of concept that helped him to solve his problem. Jake spent a lot of time at the Library working (and using our free wi-fi) and never realised that we had such an interesting collection. In approaching the library he wanted to find out what we thought. He had taken it as far as he could, but wanted some advice on how to take it further.

Over time, the Library worked with Jake to fine tune the app. The app initially started as an A-Z list of 11,000 items. We broke it up into decade by decade & attempted to show how many items were contained within each decade.  We also added a feature to be able to limit it to display who were the active composers within each time period. One thing we decided quite early on in the discussions was that there was to be no search. This was to be a discovery tool that displayed everything. We wanted people to explore.

The final app really has kept the core of Jake’s initial idea. It was refreshing to see someone outside the Library who was unfamiliar with the collection, navigate it & pull out the meaning of what was important and didn’t have preconceived ideas of how a collection had to be accessed. Likewise it was great being able to work with Jake & get the benefits of how he had approached similar problems within other non-library projects.

In keeping with the initial goals of making the data openly available, the Library has released the source code to the app under an open source license. Others can build upon what we have done.

Lessons

To me there were three really important lessons to come out of this:

  1. Make your data available.
  2. Don’t expect things to happen immediately
  3. If you can, guide the developer to help them fine tune their product & give them insight into the intricacies of the data.

It was one of those perfect examples of everything falling into place with the right people being in the right place at the right time.

Forte is available now at the AppStore. I’m really excited about the story of how the app came into existence, our team that worked alongside Jake to take his idea & build something that I’m really proud of.  I hope you enjoy discovering our collections.

Flickr Commons turns 5

Thursday, January 17th, 2013

Today Flickr Commons turned 5 years old. The Commons has turned into an incredible resource of over 250,000 images from 56 different libraries, archives, and museums throughout the world. For me, the launch of Flickr Commons heralded what turned out to be a huge turning point for my career.

Flickr Commons allowed me to build my first proper mashup – my Google Streetview Then & Now. This was a bit of a breakthrough moment for me. It was my first demonstration of the power that comes from having interesting photos and associated data that was freely licensed and freely available to be shared through an API so others could do things with it. It made my day when George Oates at Flickr saw it and called me a superstar!

A bit to my surprise, people liked what they saw. I overcame my fear of public speaking and starting talking about what I had done and what could come from sharing and reusing data at conferences. People outside of work took an interest in what I had to say. People were willing to fly me places to listen to what I had to say & to listen to my ideas. Not only that, people referred to what I was doing in their conference papers and blogs. This led to me becoming the first Australian to be named as a Mover & Shaker of the library world by Library Journal. It’s allowed me to become a bit of an experimenter at work and

Along the way, my Commons experiences have introduced me to so many like minded people throughout the world. It’s these connections that can’t be measured by the number of views or comments an image contains.

In 2011 my Commons experience was complete when my work set up our own Commons account. I’m now on the other side trying to get interesting things from our collections out there to see what other clever people do with it. It’s a blast!

Little did I know that way back then that building a mashup at 3am one morning would influence my life and my career. I’m absolutely loving the journey it’s taking me on and can’t wait to see what happens next.

Thanks Flickr for what you’ve created.

Trove WordPress plugin

Monday, July 9th, 2012

One of the fantastic resources in Trove is the digitised newspaper collection. We’ve seen people using the newspapers for research & republishing their findings on their blogs. There’s a couple of issues with this:

  1. If the text that is copied hasn’t been corrected, if it ever gets updated in Trove it won’t get updated on their blogs
  2. Attribution is a bit hit and miss. Depending upon what the person does, they may or may not add a link back to Trove. We all know it’s nice to share the love and provide a link back to the resource where you’ve found something.

To make life easier for people using newspapers in Trove I’ve built a simple WordPress plugin based around the Trove API to allow you to embed the text from a newspaper article within your WordPress blog so that the text will reflect the current state of the text in Trove & it will provide attribution back to Trove. All it takes is three easy steps:

  1. Install the plugin
  2. Get the article ID of the text from the URL (e.g.:  the ID of http://trove.nla.gov.au/ndp/del/article/2760597 is 2760597)
  3. Where you would like that article to appear in your post us the following short code [trove newspaper=2760597]

That’s it!

I made up a quick screencast to demonstrate the use of it. I really hope that it’s useful for people. If you have any suggestions for improving it or extending it for other uses within Trove I would love to hear about it.

I have to give a big thanks to Kathryn Greenhill for helping me out with beta testing. This is the first thing I’ve built using the Trove API & the first thing I’ve released back into the WordPress community & it feels pretty good!

Puttin’ on the writs

Thursday, December 16th, 2010

This year we had a bit of fun at the National Library of Australia’s Christmas party with our own little take on copyright law. Thanks to Dereta for all her hard choreography work. It was great fun.

Australian Women’s Weekly visual timeline

Tuesday, November 30th, 2010

In the past I’ve spoken about moving our collections Beyond the search box, the colorful library and that libraries should be the provider of digital information but not control how we interact with that information.

Recently the National Library of Australia released digitized versions of the Australian Women’s Weekly. The existing way of accessing this collection is via a search box or a calendar. That’s a pretty traditional method for accessing library information, but I want to relate to the material in a different way. Back in the physical world, how do we view magazines in a newsagents? Are magazines hidden away or are the covers displayed to catch our interest and therefore purchase them? Some magazines like National Geographic are easily recognised by their iconic yellow and striking cover images.

Likewise, the covers of the Women’s Weekly are an iconic historical record of Australian society. I can remember what the covers looked like from when I was growing up, but I can’t easily remember the dates of any of the issues. How can I access this information in a visual manner?

With a bit of screen scraping I can build an alternative entry path into the collection. By extracting the relevant details for the year, month & issue I can repurpose the data into a visual timeline.

Screen shot of Women's Weekly visual timeline

50 years worth of issues is a bit much to be loading all at once, so I’ve built this to dynamically load in another year of covers as you scroll to the bottom of the screen. I like this interactivity as it encourages exploration without being too resource intensive. Of course the covers link to the relevant issue within Trove where you can explore the content further.

I hope this proves to be an interesting way of interacting with the collection. Enjoy.

Update 24 Dec 2010: My timeline has been integrated into Trove.