The data's on the Wall

I've been having a bit of a poke around different digital collection interfaces lately, and one of the most appealing I've found is the National Museum of Australia's History Wall.



The History Wall is developed by Tim Sherrat (@wragge). It brings together a range of data sources: the National Library's People Australia and Australian Newspapers projects, the Australian Dictionary of Biography, the collection of the National Museum of Australia, the Australian Bureau of Statistics, the Flickr collections of the National Archives of Australia and the Powerhouse Museum, and the discussions of a historians' workshop held at the Museum in November 2009.

Items from these sources are presented in the context of 100 'defining moments'. Each time you visit the site or refresh the page the items that are presented change, giving you a different view of history. I can quite easily imagine the presentation writ large as a touchscreen interactive.

The thing I like most about this project - apart from the polished design - is that it lightly mixes curated, handcrafted content (the 100 defining moments) with data pulled directly from the content sources (ie. straight from catalogues and websites) (I assume it's pulled directly, there may be some magic happening behind the scenes). Digital collections rise and fall on the state of their metadata - this lightweight mixed approach provides anchors of carefully prepared material that collections items can be clustered around.

Tasty new stuff from the Tate

I've been having a play this week with the Tate's beta collection site, Art & Artists.

I'm liking the way the Tate have built in lots of powerful search tools, while also making the browse experience enjoyable. The thumbnails on the homepage, for example, are generated from the items with the most views, meaning the most popular collections items appear (and allowing for sudden peaks of interest to be recorded and communicated to site visitors) (and, yes, creating a self-fulfilling prophecy, if you want to be picky about it).




The Artists tab is also ordered by Views. It has the nice option to let you order artists by number of works in the collection, which is how I discovered that there are nearly 42,000 items by JMW Turner - which seems astounding.



The above screenshot shows the homepage of the site with all the initial facets expanded. Facets and metadata currently play quite a large role in my life, and I appreciate how the Tate has tucked them away. They're there for the power user, but hidden from the casual browser. I do find some of the facets odd. For example, you can split artists by gender, and I can't think of a use case for that (apart from doing a comparison of number of works in the collection by men and women). Nationality and time period might be more informative here (but then, of course, how do you deal with someone born in Italy who worked in Belgium and France and died in England, or who painted between 1888 and 1934?).






Individual item pages have a simple and roomy display, and once again you can also hide and expand more filters/links/facets. I quite like the stepped approach taken towards the subject descriptions.



Some collection items have multiple views - a basic info page, a detailed catalogue entry, and another page titled 'summary' but which is more like 'interesting context-rich short essay'. The Tate has traded off keeping the first page simple and focused on the image with the chance that people might not click through to the following pages for even more, even richer information (and god knows we're all about even more information).

As this is a beta release, I'm expecting incremental changes and the addition of more functionality over time. In particular, all forms of user engagement and social media integration (from Facebook like buttons to commenting) are currently missing, and I betcha they'll start to dribble out.

Masses and mazes

An article on Read Write Web last week, titled I Don't Know Much About Art But I Know What's Online, has ruffled a few feathers in the English digital collection community (and also provoked some thoughtful responses).

In the RWW article, Curt Hopkins opens by stating that a digital experience of an art institution's can never replicate a physical visit. He then surveys six online collections, focusing on their comprehensiveness, the image resolution offered, and their usability.

Hopkins ends by observing

The collections of museums are making their way online, if for no other reason than they serve as a kind of advertisement. I have yet, however, to come across an outfit, small or large, whose goal was to make their entire collection, or even a substantial majority of it, available online. The few that tried did not hit the trifecta of navigational ease, resolution and information that would make it the most useful.

It was these points that got people agitated, rather than his commentary on the websites he critiqued. Clairey Ross wrote a self-titled 'rant', stating:

My work revolves around understanding what users think of online museum collections, why museums use them, their purpose their usability and god damn it, not one person I have interviewed, observed, surveyed and stalked ever suggested that an museum online collection was an ‘advertisment’ and there are several museums whose goal is to make their entire collection available online.

Frankie Roberto used the RWW article as a launch pad for thinking about what Hopkins' 'trifecta' could feel like:

When I visit an art museum, my primary experience is of wandering. I never know that much the art works, or what to expect, so I rarely have a specific destination in mind – merely just a desire to see some interesting/provoking stuff.

I think it’s this experience which is so hard to replicate online. We’re good at building web experiences which are optimised at getting users to the thing they’re after (usually information) as quickly as possible, via carefully considered navigation and relevance-optimised search. What we’re less good at is building web experiences where the user sits back and is simply entertained/amazed/enthralled by things they wouldn’t have otherwise come across.

One of the things I admire about Frankie is his ability to look at other models and use them to generate new ideas. He points to Phil Gyford's Today's Paper (an alternative - beautiful - layout of the Guardian's content) and the Guardian's own EyeWitness app for the iPad and asks how you could build on these ways of manipulating content in order to enhance the experience of looking at digital collections.

Mike Ellis is another person I admire, mostly for his willingness to cut to the heart of the matter and ask hard questions. In his post, he asks three questions:

1. Is getting your entire collection online really a good goal?
2. If you achieve that goal, how do you make it a good user experience?
3. Why do it at all?
   

With some digital collections, width is everything. Take Papers Past for example, the National Library's website providing access to digitised historical New Zealand newspapers. Users of this site care far less about speed or navigation than they do about content - the more papers digitised and the more years covered the better.

With other collections though, I wonder if focusing on a great experience with your most useful, most interesting, and most unique collection items might be a better goal?

UPDATE

An interesting comment from Mike Ellis (thanks Mike!) that I thought I'd bring up into the post:

"One thing I didn't manage to get in that post - something I'll maybe do in the future - is how I suspect this could work best. Seems to me that there should be a way of (for instance, making this up as I go along..):

1. Only digitising the 100 (or 1,000) or so "iconic" objects, and providing a huge amount of detail, photography, context, story for these;

2. Spend the rest of the time focussing resource on amending or creating Wikipedia entries

This kind of approach, (or something more thought about..) would probably maximise number of eyeballs as well as hitting the more in-depth stuff asked for by some users"

Search and serendipity

The sting in the tail of all the work art institutions have been doing to digitise their collections is that the more stuff you have, the harder it is to find any single thing.

To my mind, collection sites have two core needs:

1. Search

This is for the people who know what they want, and want to get to it fast. Academic and professional image researchers may benefit from advanced search options. Experienced searchers might use filters, or other tools that allow them to expand and limit their searches easily (say by date, region, maker, materials, subject matter). It's also good to cater for people who get their search terms nearly-but-not-quite right (collin mcahon) with fuzzy search or suggested search options.

One of my favourite solutions to the search-right-now demand is the IMA's mega-menus: from absolutely any page on the site you can start a new collection search, but because the tool is tucked away in the site navigation, it doesn't have to be visually present on every page.



This blog post by the IMA's Charlie Moad explains the tools that have been put in place for collection searchers

2. Serendipity

This is for the people who want to know what you have, but don't know where to start. A search box is scary if your question is "what is there?".

Yesterday LACMA's Tom Drury blogged about findings from a survey of website visitors. People said they LACMA should let images rule the website (rather than the more proasaic info, like opening hours, that LACMA expected).

As a result, LACMA have introduced two new collection display techniques.

First up, collection splash pages - like this one for photography - provide entry points to works, blog posts, exhibitions and events related to different collection areas.



Second, a new collection homepage (in Flash) based on the idea of remixing. Clicking on an item in a tag cloud opens up three works grouped according to a word or short phrase.




The initial sets are been curated by staff, although visitors have been invited to email suggested sets in. Although the first thought I had was that this should work more like Flickr's galleries - with people being able to sign up and create their own groups for sharing or personal reference - on reflection I actually like the very lightweight implementation.

Take me there

Humans make patterns out of information, even when patterns don't really exist. Hence lucky undies, and this post.

Over the past few weeks I've seen posts and articles about 'curation' popping up all over the place.

First up, Aaron Straup Cope at this year's Museums and the Web presented a paper titled 'Buckets and Vessels', looking at curatorial practices ranging from Flickr Galleries to Newspaper Club. (See the slides from Cope's presentation) He argued:

People are not suddenly self-identifying as curators. Rather, what we are seeing is a growth of tools being made available to allow people to exercise a curatorial muscle many people never knew existed, even if the results don’t always look like something we’re used to.

Cope cited Pete of New Curator's March 2010 article 'You are not a curator'. Pete argues for the difference between filtering (sifting and aggregating data) and curating, which in the comments he says begins with "providing meaning".

The difference between aggregating (picking) and curating (adding meaning) is a sub-current of Robert Scoble's lengthy post 'The seven needs of real-time curators'. Scoble sees a market gap to build tools that will let people aggregate, weight, and visually represent data from all over the web in a way that's meaningful to them and to others.

Then a post a few weeks ago on the Nieman Lab site caught my eye: Calmness, curation, cat porn: Dave Eggers’ joys of print:

“I want a printed newspaper curated and edited by professionals that have been in the trade for decades,” Eggers said. “I am in eternal deference to expertise.” Eggers likes the idea of branded — which is to say, vetted — news outlets shaping his view of the world. “I entrust my daily news input to the professionals,” he said.

Finally - another MW2010 paper, this one by Nate Solas on 'Hiding Our Collections in Plain Site: Interface Strategies for "Findability"' about a website re-design intended to help with the problems faced with surfacing large amounts of digitised collection items.

What's the pattern that I see here? You could say "ways to defeat information overload", but I refute that term. Close your eyes. Switch off the tv. Shut down the browser. Or - find people and tools that reliably take you to things that are valuable, entertaining, expanding.