JASIST Information Architecture Column Oct/Nov 2004

Scary Messages and Content Creep: It's All in a Daily Digest.

Andrew Dillon, for the ASIST Bulletin, Oct/Nov 2004*

Despite the usual avalanche of smut and spam in my mail, it's been a quiet couple of months in IA-land, judging by the postings on the major IA lists. Clearly lots of people have been out of the office, as shown by their auto-reply messages that keep ending up in my digests. In one sense this calm is real progress – there has not been a flame-war of any note this summer on SIGIA-L. Perhaps a moderator is being kept busier than we think. Consequently I am more willing to read these daily digests rather than deleting reflexively at the sight of a certain person's name, and the results have me wondering more about convergence or, to put it less politely, content-creep.

I play a little game with myself every now and again when I read quickly through the postings on various IA, usability and web-design lists. The essential point is to see if I can determine which list I am on just by reading the content of the message. One has to play a little loose with the rules since you can always cheat and look at the header. But the goal here is not to score points; rather it is to identify the overlapping of discussions, the importing of answers, the posting of the same questions or even (ideally) the different treatment of the same issues across lists.

In my mind there are a limited number of characteristics by which we may describe list behavior and quality. We've all experienced the energy-waste of a group self-imploding over the rabid postings of one or two people (you know the type, the kind of poster who conflates freedom of speech with freedom to distort). Equally, many academics will be familiar with the other extreme – highly controlled, managed lists where anything remotely off topic (or even misspelled) will be bounced lest it cause offense. But these characteristics are impressionistic and somewhat obvious. I am more interested in the trajectory of lists, their natural lifecycle and their value to members. I sense there is an index of disciplinary health or development that can be derived here. If so, IA is at an interesting juncture.

I have not conducted a formal analysis of these characteristics, but I confess that what I had gathered as an informal impression is not immediately well supported by the data I have gathered. From scanning and reading the five lists to which I routinely subscribe, I had begun to get the impression that overlap was growing, that not only the same issues but even the same people were popping up. As it turns out, this duplication seems to be less the case than I imagined, giving me further pause for thought on the general lack of positive correlation between reality and what we think is out there. Yes, job announcements or calls for papers tend to be widely circulated, but it is rare that a discussion bleeds over the digital fence from one list to the other without considerable effort being made by one or more members of the respective lists. So while the closed IA group (sorry!) was busy discussing the topic of sub-domains one week, the more public SIGIA-L was engrossed in a lively exchange on open-source usability. This divergence is probably explained by the rather frosty relationship between these two lists (the closed list was viewed with suspicion by the open list once news of its existence emerged, not helped by the fact that most members of the closed list were once or still are members of the open group driven to distraction by the noise to signal ratio). But closed lists are more widespread than one might think – I am a member of one usability discussion group that threatens to excommunicate any members who reveal its existence – you didn't read about it here!

But content-creep can take other forms. The last few weeks of postings on SIGIA-L, while still being unique postings to that list, have covered a range of topics that one might expect to read on a usability list, such as seven postings one day in response to a usability test question, a discussion of pull-down menu design that extended over a week, a call for advice and subsequent discussion of how to make a design process more user-centered – all fodder for IAs and others. Interestingly, while that design process question was being raised in SIGIA-L, the nameless usability list was engaged in a similar discussion on how to incorporate user-centered design methods into software engineering models. How I wished these two sets of discussants could get together!

As a "big-IA" advocate I have always believed usability was part of IA anyhow. "Little IA" is just too small to justify the use of the A-word in my world. So the broad discussion of these topics on an IA list is further confirmation of this view as far as I am concerned. I do wonder though if lists are creating parallel universes where the same ideas are discussed among communities who remain oblivious to each other's existence. Clearly some of us are looking at multiple lists, but the politics of participation enforce division when some lists will not even announce their existence. Certainly this exclusiveness can be necessary, and I am often grateful for the nonsense it removes from discussions, but is anyone in IA even considering how such structures shape discourse? It's difficult for me to imagine an architect not being concerned with the effects of the social spaces she creates. Frank Lloyd Wright once mused that better architecture might inspire better government, and I cannot help but feel our information architectures may have equivalent potency, if only we knew how to build them appropriately.

Comments and reactions to this column can be emailed to adillon@ischool.utexas.edu

*This article originally appeared in the Bulletin of the American Society for Information Science and Technology, Vol. 31, No. 1, October/November 2004, a publication of the American Society for Information Science and Technology, 1320 Fenwick Lane, Suite 510, Silver Spring, MD 20910 USA. Phone 1 (301)495-0900 http://www.asis.org