Christian Crumlish Q&A

Christian Crumlish is the curator of the Yahoo! Design Pattern Library and co-author of Designing Social Interfaces. Christian spoke in London and Berlin in January 2010, when he kindly agreed to be interviewed:

CS: What was your route into technology? What were your founding interests that have proven useful over the years?

CC: I was fascinated by computers at a young age, although I had a hazy idea of how they actually worked. In fourth grade (age 9) I “designed” a combined computer and fortune-telling machine. My schools were very progressive with technology so I was able to play with a RSTS time share system on a DEC PDP 11/30 when I was still quite young, and it only got nerdlier from there.

In college I studied philosophy, focusing on language and logic and mathematics, writing a thesis on the function of metaphor, and while it was far from a vocational degree I find that it has stood me in good stead over the years.

You talk about 5 principles that developed into Designing Social Interfaces – Pave the Cowpaths, Talk Like a Person, Play Well With Others, Learn from Games, Respect the Ethical Dimension. How did you come by these priniciples, and how did you realise they could be expanded into a book?

Hmm, it was more the other way around. First there was a project of trying to identify and organize social design patterns. I had this taxonomy that I took around to various events and got people to help me work with and it kept growing. At some point I think Havi Hoffman from Yahoo! Press suggested it might make a good addition to the classic O’Reilly interface pattern books that have come before and Mary Treseler at O’Reilly agreed and we turned it into a book proposal.

In the meantime, I had this long extended lunchtime conversation with George Oates, one of the original designers of Flickr, to try to suss out some over-arching principles of social design, and some of those ideas (particularly the parts about talking like a person) came from that conversation. Other principles more or less presented themselves, cropping up in numerous contexts. There are actually more than five principles in the relevant book chapter but you can’t overload people in a talk!

When you wrote Designing Social Interfaces, whose work was most influential to you?

Well, beyond my co-author (because we really got intertwined in our thinking and writing styles) I’d say that I was influenced by Jenifer Tidwell (who wrote Designing Interfaces for O’Reilly), Christopher Alexander (who wrote A Pattern Language and kicked off the patterns movement), and Ward Cunningham and his collaborators at the Portland Pattern Repository. Outside of technical and instructional writing my influences come primarily from fiction (John O’Hara, Borges, Nabokov, DeLillo) and music (Robert Hunter, Elvis Costello).

How were your own ideas for design patterns and social interfaces shaped by your experience?

Almost entirely! That is to say, I stumbled into design when the Web came around so I wasn’t trained in a studio. I was influenced by the stunning graphic design environment of the New York in the late sixties and seventies but all of my thinking about design up to that point (the mid-90s) was empirical and intuitive. As I learned more about actually making interfaces and planning experiences for people I took myself back to school, so to speak, by reading whatever I could find that was recommended to me and by grilling anyone I ever met who seemed to have a clue.

When I started on the Web, it was hard to incorporate social experiences. I started a magazine with some friends and wished there was an easy way to add comments to each article, to enable readers to join the community and perhaps even contribute to the ‘zine, etc., but at the time I was literally hand-coding all the pages and I didn’t have the first idea how to add those other features.

I stayed on top of CMS evolution and blogs and wikis and community/collaborative-filtering applications always looking for the suite of tools that would someday help me recreate the sort of content-driven community I had been unable to sustain in the 90s. Along the way I learned that there was more to it than simply integrating the correct code module, and that kindling and supporting a community is a lot harder than it looks.

Over the past ten years a lot of smart people have written essays and blog posts and books and magazine articles about the elements of social design and I’ve been reading it all.

When I came to Yahoo! we were sitting on these various disconnected social properties – some homegrown, others acquired – and it occurred to me that I might make good use of my tenure as curator of the pattern library by trying to wrap my arms around the current thinking on the key elements of social design and how it differs from designing for individual users, so if anything it felt past time to get to work on this project by the time I finally had a good excuse (my job!) to do so.

Open Source is about scratching your own itch – how do Yahoo! use the Yahoo! Pattern Library?

Well, the library began as a non-heavyhanded way of circulating the best of our knowledge about how to design common elements in our interfaces, and to try to prevent the constant reinvention of the wheel that can happen in a large, loosely connected organization. It still functions that way. We also share more and more of it with the web at large, which I find tends to drive better feedback and helps position us as a contributor to the ongoing public conversation about how to make a better web.

You’re the 3rd curator of the Yahoo! Pattern Library – how does your custodianship differ from your predecessors, and how do you see them developing in the future?

Well, first of all Erin Malone founded and presided over the library for the first three or four years of its existence and established an indelible mark on how it unfolded. She recruited Matt Leacock, a brilliant community and gamer designer, to be the first curator and he did the yeoman work to establish the fundamental set of patterns, to build the library (along with Drupal maven Chanel Wheeler) and to establish initial processes for writing, revising, and vetting patterns.

The second curator, Bill Scott, is a legendary user interface engineer who co-wrote O’Reilly’s Designing Web Interfaces, and who played a big role in evangelizing intelligent use of rich interaction (using Ajax and other clusters of technologies) on the web. He is now running the UI team at Netflix and I bounced ideas off him regularly. Bill presided over the process of taking the library public, alongside its sister resource, the Yahoo! User Interface Library, as well.

When I came on board there had been a slight gap and the pattern process had ground somewhat to a halt, so one thing I think about a lot now is succession planning and putting in place a handbook and some tools that will enable the next curator to get up to speed and manage the mechanics of publishing patterns without having to do quite so much of the command-line work and hand-editing of JSON files that I do today.

Aside from modernizing the processes and focusing on the social design patterns collection, I think my other legacy is bringing more and more of the internal library out to the public through the Yahoo! Developer Network. When I started I believe we had published 24 patterns (I should check that) and as of today we have 59 in there and plenty more coming.

Design patterns came out of the world of architecture – why weren’t design patterns adopted as a discipline by architects, and why did they become so important to technology and computing? In that respect, what could architects learn from the working practices of computer programmers?

That’s a great question. I often ask architects about this and the sense I get is that they view Alexander as a throwback to a modernist approach that is somewhat prescriptive and strict in certain ways, and perhaps a bit idealistic and unrealistic in others. They often claim that “no one uses the patterns today.” This is probably true but I’ve heard that some builders and landscape architects, for example, actually do. Another view is that Alexander was trying to demystify a profession and relate it to a language that anyone could learn, a sort of western feng shui. If this is the case, then perhaps this idea would be threatening to people who are trained and authorized to do this work? Or perhaps they simply wouldn’t believe that this is a good idea or even possible?

Alexander himself has been pretty savvy about computer technology over the years and he personally exhorted the developer community to strive for the sort of “timeless qualities” that he reaches for in his own work. I think some programmers took up this challenge and were inspired. I’ve been to some pattern language conferences and people take this stuff very seriously! There is an almost mystical bent to the way some people really try to get down to the bedrock principles and forces at work in defining these patterns.

I honestly don’t know why it took root with software except that I imagine it met some unspoken need for a kind of meta-structure to help decide when to use various techniques and to understand the consequences of those choices.

For user interface development and product design, I think patterns also help as a sort of lingua franca that communicates to software developers that design isn’t simply an intuitive, magical, artsy thing to do but that it also has principles and reasons for things and systems. Kind of a, “Hey, we have design patterns too” message. And frankly, I think if anything the social design patterns we’ve mapped out in our book and wiki probably come closest yet to the sorts of patterns Alexander identified. In a virtual way, they get at the same sort of challenge: how to create livable “spaces” in which people will relate to each other on a human scale. I mean, don’t our web sites also need “small public squares” and “dancing in the street”?