2008: Internationally renowned artist Gary Hill didn't have a website.

Role: Web Designer User Research, information architecture, content design, WordPress development 2008–2010, 2014

Gary Hill is widely considered a pioneering artist in the medium of video and video installation. Beginning in the 1970s his work explores the limits of technology, language and understanding. He came to me with a 30+ year body of work, extensive exhibition history and thousands of images to organize and present.

From our first meeting we knew two things, one was that we'd be using WordPress to organize ourselves and allow Gary to manage his own website into the future. The other thing was that we wanted users to be able to access to the information in the database in two different ways—the right brain and the left brain paths. The right brain experience would be unlike most websites. Something abstract and designed for discovery and mystery. We could come to that later, we needed to start with the left brain version what we described as "the library".

At the time Gary had a studio manager working for him who was planning to leave and they needed the website to take care of some amount of the requests that she had been fielding over email/phone. These were primarily requests from museums requesting to exhibit work, curators or researchers writing about the work, museum employees seeking technical information, and students trying to learn more.

I spoke with members of these groups, including employees of the local contemporary art museum and students from the University of Washington in the art and library sciences departments, to learn more about what would be of interest to them to learn and what they would expect to see upon a visit to a website for Gary Hill. After analyzing the data, a few content types became important to consider as:

Video! Gary Hill is a video artist so we knew we needed to have video. What the students, in particular, lamented about other video artists websites was a lack of video or a lack of full-length pieces.

Lots of images! At least a handful of stills or installation images to give the user a quick idea of what each piece looks like / how it is displayed.

Basic information, title, date, media and duration if it was a video

A transcription of the spoken or visually presented text in the video. Much of the work plays with words and meanings, signs and signifiers, and people wanted to be able to explore this.

An exhibition history per piece. I heard that while artist websites often had a general exhibition history page, which might tell the user which pieces were in that exhibition, it was often cumbersome to cross-reference with a particular piece.

A bibliography. Again, this was sometimes available as a cross-reference on other sites, but was often absent entirely.

Technical specifications for the creation of the piece, i.e., "how was it made?" This showed up only rarely elsewhere.

Technical specifications for the presentation of the piece, i.e., "if I'm installing it for the museum I work for, what do I need to know about doing that. No artist websites had this, it would often demand a call to the studio manager even if there was a website.

Credits for any on-camera talent or crew / technical assistants.

A general description of the work, as one might find in a book of an artist's work.

From a few of these user groups it became clear to me that a useful tool would be the ability to download the content of each artwork's online entry as a PDF for later reference offline, note taking, et cetera.

From my interviews, I learned that there was a strong interest in not only seeing the normal pages you might find on any website (work, about, and contact), but also a thorough bibliography and something that could quickly show the viewer where in the world which pieces of Gary's work were currently being displayed. I decided to call this "on view. I then created a navigation for maximum simplicity in finding your intended artwork and discovering others along the way. (Of course there was a search as well that would search all the WordPress custom fields to turn up results even if it was a researcher trying to remember "what as that piece I saw at Schaulager in Münchenstein a few years ago?")

With analysis, consultation, and affinity mapping complete we starting to consider the actual information architecture and needs for how to find and view each artwork. For each entry we'd include the following:

1) images/video 2) basic information 3) 'textbook' description 4) transcription 5) bibliography 6) exhibition history 7) notes (including cast, crew, and installation guide) and finally, a way to package the current content dynamically into a downloadable PDF.

Looking over the nearly 100 pieces of art we needed to catalog and design for, I felt displaying these could feel a little overwhelming and that we needed sub classification. Luckily some work was clearly different, rather than being video art presented in a gallery or museum, it was a live performance, spoken word, or video included as part of another artist or team's musical/theatrical performance. This was the minority of the work, however, so I tagged and catalogued the rest of the work to better understand the features and found that some work could be or had been presented as pure video whether on a screen in a gallery or in a theater space. The rest of the work was changed meaningfully by the way it was presented, or it was purely sculptural with no video at all.

For the single-artwork entry I decided to use the then relatively new jQuery to swap out content from a primary text frame, as well as images from a central image location with a collection of thumbnails. 

Post launch I received positive feedback from my interview participants on the ease of finding what they're looking for, and discovery of work with which they'd been previously unfamiliar. Gary found that he began referencing it himself as the easiest way for him to retrieve data about a piece.

• Identifying audiences carefully is BIG DEAL. A lot of great information came from the museum installation workers, who I would not have thought to interview had Gary's previous studio manager not brought them up. This was the first project for which I did in-depth research, and well before that type of UX consideration was normal for this kind of project.

• Get help next time. I did every aspect of this project, which meant I leaned on WordPress plugins for some fancy footwork, like the PDF export and custom fields. This worked OK, but some users reported a slower-than-hoped-for experience using the site. Some of that PHP was not as clean as it could have been if I'd had help to write something custom . . .