CSE 271: User Interface Design: Social and Technical Issues
Notes for Third Meeting
Notes on Readings

The Ackerman piece is a short summary of a large literature on CSCW (Computer Supported Cooperative Work), and although fairly easy to read, it should be read more than once, because of the density of its ideas and their unfamiliarity (to many of you). We will go over this in class.

Karp is a professional "architect" for websites; his advice is most appropriate for commercial websites, because only such sites generate the funds to pay professionals. Some of his guidelines are actually opposite to what should (probably) be said for academic sites, and some of his information is out of date (e.g., about lynx on unix). Commercial sites must cater to lowest-common-denominator browsers, but academic sites do not have this restriction. So this document must be read with a critical mind, and not simply accepted without further thought. (By the way, it has nothing to do with Zen ("Chan" in Chinese) Buddhism; the title is inspired by the book Zen and the Art of Motorcycle Maintenance by Pirsig, which does have something to do Zen, the title of which in turn was inspired by Zen and the Art of Archery by Herrigal, which really is about Zen.)

Chapter 3 of Shneiderman is his best so far, at least for this course. The introductory remark (p.96) that "programmer's intuitions may be inappropriate" for large classes of other users is very important for motivating much of his book, and this course. Shneiderman again emphasizes that design should be based on "careful observation of current users" and that iterative design methods are very important (both p.97). "Three Es" management strategies are interesting (p.98). The paragraph on p.99 giving economic justifications for usability is important. But with all the methods, guidelines, checklists, etc., we should not forget the following:

Design is inherently creative and unpredictable. Interactive system designers must blend a thorough knowledge of technical feasibility with a mystical esthetic sense of what attracts users.
The following 4 bullets make some aspects of this a little more explicit.

The material in section 3.3.1 on what to consider having guidelines for is valuable and Figure 3.1 is worth a thousand words. The paragraph at the beginning of section 3.3.2 is really about how tacit knowledge arises in interface design. The first three sentences of section 3.4 (p.104) are very important: spending more think time at the beginning of project is usually a very good investment. The LUCID stages in Table 3.1 correspond closely to stages conventionally considered in software engineering; and organizational change is something that really does have to be considered. (By the way, the LUCID method is an outgrowth of UCSD (User Centered System Design) method developed at UCSD.) The list of deliverables on p.107 is useful if you ever get involved in a really large project.

Sections 3.5 and 3.6 quickly sketch two important methods for learning more about users. The checklist for ethnography on p.108 is very good. The last paragraph of 3.5 and the last sentence of the preceeding paragraph sum up the case for ethnography well (p.109). The first sentence of the next to last paragraph of 3.6 is important (why controlled experiment does not work), and the last paragraph of 3.6 (p.110) makes clear the need for user participation in design, but much more could be said here. The 4th paragraph of 3.7 on scenarios explains why this is a really important technique, and the National Digital Library is a good example. Social impact statements are a terrific idea, and should be part of the early work on any large system; the outline for a social impact statement on pp.114-5 is useful but should not be taken too literally. Section 3.9 does not say nearly enough about privacy; safety and reliability also merit more than just one paragraph. Designers can be sued for problems in these areas. The last paragraph of 3.9 is good advice, but I would go further: legal disputes can be very unpleasant and the best strategy is often to seek ways to avoid them, e.g. through cooperative agreements.

Notes on Class Discussion

The third meeting began with a short lecture. The first point was that change is inevitable, unending, and unpredictable, especially in the user interface field; this was illustrated by noting that several points in Art and the Zen of Web Sites by Tony Karp were already badly out of date after less than one year. The rest of the lecture concerned some specific changes, mostly driven by (what has been called) convergence, the merging of computers with communications. In user interface design, this is reflected by the evolution of HCI towards CSCW: whereas traditional HCI focuses on an individual user of a single system, CSCW addresses the social issues that necessarily arise when communication places individuals into communities. An important enabling technology is increased bandwidth, which is surely happening (newspapers of the morning of the meeting announced the ambitious plans of a consortium of Compaq, MicroSoft, Intel and 5 of the 6 "baby Bells" for modems with a 30 fold (!) speed increase by Christmas 1998).

There was a brief discussion of the "HTML philosophy" that texts should use only structuring commands and never layout commands, so that browsers can choose the best presentation compatible with their capabilities, e.g., audio output for blind users. The construction of a "button" from a table with background was discussed as an example:

See the class discussion page for more on this example. The best way to learn more about HTML, Java, etc. is to look at source files, which is well supported by standard webbrowsers like Netscape.

Then we tried to discuss Communication and Collaboration from a CSCW Perspective by Mark Ackerman, but this was not very successful, probably due to the unfamiliarity and difficulty of the ideas. I asked everyone to read it once more so that we could try to discuss it again later, and then gave some examples of context and of nuance in natural language, noting that nuance is subtle contextualization that comes along with signs (e.g., speech). Click here for details on speech act theory and mitigation. We also discussed the concept of recipient design, which says that all natural speech is designed for its audience, and bears specific signs of that design. (If there were a final exam for this course, material from this paragraph would undoubtedly be on it!)

Finally we considered Aramis. I was disappointed that no one noticed the bogus (i.e. fictional) press release on p.22. Clearly we will have to work hard to understand what Latour is trying to do in this book. Hint: This passage is in italics.

To CSE 271 homepage
To my home page
29 January 1998