Reading :: After Method

Originally posted: Tue, 26 Oct 2004 21:56:59

After Method: Mess in Social Science Research

by John Law

Just over a year ago -- I remember it was on a plane to SIGDOC, so it must have been mid-October -- I wrote an email to a group with whom I've been discussing research methods. The email, based on my recent readings in science and technology studies, suggested that we should stop thinking of research in terms of hierarchical categories (paradigms contain methodologies which contain methods, etc.) and start thinking of all these components as enacted practices that occupy the same level of scale. In fact, I'll be presenting on the topic at CCCC 2005, and Kristie Fleckenstein is writing an article on this same topic (almost certainly better considered than my presentation will be).

As I say, the inspiration was the work being done in science and technology studies. And one of the most active scholars in that field is John Law. So I was happy to find out that Law was writing a book along roughly the same lines that our group had been discussing. I finally got the book in the mail the other week, and it doesn't disappoint. Law really gets that "mess" in social science research -- that is, the inability to draw this research into hard boundaries with unequivocal answers -- is insoluble and ultimately undesirable. He also gets the problem that I've attempted to wrestle with elsewhere and that I've called the problem of unintegrated scope: the impulse to impose levels of scale so that one level can be construed as a master level of which the others are just effects. In After Method, he takes on this problem by describing what he calls "method assemblages." He gives two definitions:

In chapter 2 I defined this for the case of representation, as the enactment of a bundle of ramifying relations that shapes, mediates, and separates representations in-here, represented realities out-there, and invisible out-there relations, processes, and contexts necessary to in-here. In chapter 3, I offered a parallel definition appropriate to objects: that method assemblage is also the crafting of relations that shape, mediate and separate our object in-here, its relevant context out-there, and then an endless set of out-there relations, processes and all the rest that are a necessary part of the assemblage but at the same time have disappeared from it. As is obvious, the two are similar in form. But the post-structuralist philosophical tradition suggests a different vocabulary. If we use this then method assemblage becomes the enactment of presence, manifest absence, and absence as otherness. More specifically, method assemblage becomes the crafting or bundling of relations or hinterland into three parts: (a) whatever is in-here or present; (b) whatever is absent but is also manifest in its absence; and (c) whatever is absent but is Other because, while it is necessary to presence, it is not or cannot be made manifest. Note that it is the emphasis on presence that distinguishes method from any other form of assemblage. Note also that to talk of crafting is not necessarily to imply human agency or skill. The various ethnographies we have explored suggest that people, machines, traces, resources of all kinds -- and we might in other contexts extend the list to include spirits or angels or muses -- are all involved with the process of crafting. (p.84)

Clear enough? It might help to know that Law says that paradigms (the most abstract level of the research hierarchy) are embedded in craft skills (the most concrete level). That is, the hierarchy of research is an attempt to make sense out of a flat, uniscalar set of practices/enactments. And once we get that, it makes sense to think of assemblages as self-assembling elements that are entangled rather than constructed (p.42). When we hear people talking about, say, ethnography as a methodology, a method, or a paradigm, we may become agitated if we are used to thinking of these as separate levels of a hierarchy; but if we think of the whole assemblage as uniscalar, it makes a lot of sense that ethnography would slip across these categories. It makes a lot of sense that methods such as experiments can lose and gain elements until they become, say, usability testing.

I hesitate to call Law's book a Copernican revolution, but it does make sense of these incoherences that we so often see in research when people try to define and operationalize it. Despite Law's style -- which varies from lucid to obtuse and self-indulgent, alas -- this book is an important book that draws together and reinterprets many of the STS threads that have been developing since the early work of Latour and Woolgar. Buy it, read it, quote it. >

Blogged with Flock

Reading :: Reshaping Technical Communication

Originally posted: Tue, 26 Oct 2004 09:18:36

Reshaping Technical Communication: New Directions and Challenges for the 21st Century

by Barbara Mirel, Rachel Spilka

My brother in law has a map on his wall depicting the Republic of Texas between its secession from Mexico (1836) and its annexation by the US (1845). In that map, Texas overflows its current boundaries; a long tendril stretches northward into what is currently Colorado.

The story behind this improbable configuration, as my brother in law tells it, is that a group of Texans went northward to claim as much land as possible in the unsettled territories. So this expedition genuinely traveled that far, turned around, and came back -- losing heavy casualties along the way. That is to say, the map is a fiction: the long tendril northward was traveled by, but never held by, Texans.

I kept thinking of this map and this story as I read Reshaping Technical Communication. Technical communication is an interdisciplinary field (though some people regard it as a jobs program for surplus literature PhDs) and technical communicators seem to be forever going on forays upriver to claim more and more territory for their field. Graphic design? Check. Usability? Check. Information design, user-centered design, participatory design, document design, typography, color theory, project management, database management, content management? Sure, why not.

This problem of claiming vast swathes of territory based on small expeditions, though endemic in the tech comm literature, highlighted in this book to a startling degree. That's because the book is, above all, about reshaping tech comm: figuring out what it is now and what it should be next. The results are inconclusive, a fact that perfectly mirrors the state of the field. This is, after all, a field in which people with literature PhDs talk animatedly about how Melville's chapter on cetology makes a model text for tech comm students (no, I'm not making this up -- it's happened at two separate conferences) -- and simultaneously the single-sourcing crowd is claiming that you have to know XML if you're going to be a member of this profession. (See Borland's chapter, p.194.) In short, it's a field with a serious identity crisis. The book reflects this crisis quite well, but I don't think it explores, critiques, or diagnoses the crisis incisively. One exception -- and it is glancing at that -- is provided in Faber and Johnson-Eilola's chapter, in which they charge that "practitioners would be hard pressed to identify a robust core body of knowledge" (p.140).

One way that crisis manifests itself, of course, is through the familiar and comfortable academy-industry split, a split that shows up in a number of chapters. (Full disclosure: my second article, published in 1996, discusses the split as well and talks about how to "bridge" it. This is the only article of mine to be reprinted in a collection.) This split begins to address the varying expectations and interpretations of our fractionally coherent field, but it provides only one axis, whereas I am convinced that the variations are multidimensional.

Despite the fractioning/fragmenting of the field, some of the chapters really stood out. Faber and Johnson-Eilola's essay was interesting in that it brought in the economic dimension and tried to engage it in a way that we usually don't. Pare's chapter on participatory research was valuable. And Mirel's discussion of usability was, as always, engaging.

Overall, I found this book to be thought provoking. I think we'll look back at it in ten years and find a summary of the embryonic trends in technical communication.

Blogged with Flock

Reading :: Getting Things Done

Originally posted: Mon, 18 Oct 2004 20:27:23

Getting Things Done: The Art of Stress-Free Productivity

by David Allen

A few days ago I mentioned a "paradigm shift" I recently had. I had realized that my category-based system for organizing my life had not scaled well, and I had to change if I was to stay on top of my many projects. My wife recommended Getting Things done, a personal productivity book that has become somewhat of a fetish object.

So I skimmed the book on the bus a few days ago. That very day, I had a catastrophic data loss and my calendar items and todos irrecoverably disappeared from my handheld. Fortunately, I had mapped out my projects recently in Excel, so my life did not utterly disintegrate. But since they're gone, I thought, why not reconfigure them in terms of GTD?

GTD, by the way, is essentially a way to distribute one's tasks and projects into artifacts that are organized so that they do the remembering for you. Socrates may have griped that writing destroys memory, but Socrates' work was not nearly as fragmented or spliced as ours is, and GTD is a testament to the increasing difficulty we encounter in juggling our roles and obligations. Allen suggests that these many roles and obligations are impossible to keep "in our heads," and that doing so keeps us from actually getting things done. So he develops a relatively simple set of rules for dealing with or delegating work. At the heart of the system is the in box, which should be cleared out periodically. Folders include a set of time-based folders -- a "ticker" file marked with the days of the month and the months of the year -- for placing physical reminders of timed tasks. The calendar should have "hard edges": you calendar in things you will get done for sure, not things you think you'll do if you have time. And the to do list is undated; each item is marked by the occasion in which you'll get it done (e.g., @computer, @home, @office) so that when you have a few minutes, you can go down the list and accomplish items. Finally, projects should be broken down into easily accomplishable steps. (Project management, as Allen states, is beyond the scope of the book.)

What intrigue me, of course, are the distributed cognition aspect and the commentary on fragmented work. I'm going to try a modified version of the system and see how it works.

Blogged with Flock

Reading :: Qualitative Inquiry and Research Design

Originally posted: Wed, 13 Oct 2004 06:54:11

Qualitative Inquiry and Research Design : Choosing among Five Traditions

by John W. Creswell

Well, it turns out I'm not crazy. I've been trying to nail down exactly what people mean by qualitative research for a while now, particularly because most of the time people simply call it "ethnography." But, I wondered, is grounded theory simply a kind of ethnography? How about case studies --- can they be considered ethnography when Yin emphatically says that they're not? How does action research relate to these two? Much of what I read about "qualitative research," especially in my home discipline of rhetoric, seems theoretically and methodologically mushed up. It turns out that other fields have this problem as well, and that is what motivated Creswell to write this book.

To the extent possible, Creswell separates, compares, and contrasts five different traditions of qualitative research: biography, phenomenology, ethnography, grounded theory, and case studies. (He also notes omissions such as action research.) Like me, Creswell loves taxonomies and comparisons and tables, and he provides plenty here. Using thoughtfully considered examples of each kind of research, he traces through design, traditions of inquiry, philosophical and theoretical frameworks, research questions, data collection, data analysis and representation, report writing, and issues of quality and verification. By the end of the book, I could articulate the differences among the traditions, see how they had grown out of different disciplines, and determine which one would be most useful for answering a given question. And Creswell's direct language, clear illustrations, and focused organization allowed me to quickly understand, apply, and navigate through the material.

I also appreciated Creswell's discussion of software for supporting qualitative research. He focuses on NUD*IST, a qualitative theory-building program designed for aiding grounded theory (and not coincidentally published by SAGE, the company that Creswell's book). Through text and diagrams, he describes how NUD*IST could be used to support each of the five qualitative approaches. He also discusses other types of software in an evenhanded manner.

Finally, the book includes a thick glossary that includes definitions by approach -- especially useful since different traditions sometimes use similar words for different phenomena, different words for similar phenomena, etc. It also includes article-length examples of each approach, something that Creswell uses to great advantage when he provides illustrations throughout the book. The one thing that bothered me about the examples was that in their subject matter -- indigence, sexual assault, violent assault -- they tended to emphasize the heartrending pathos of victimhood, and I worry that this will lead readers (especially students) to seek out those sorts of projects over the less exciting, more quotidian, but still valuable aspects of life.

Creswell's text is not going to provide the detailed information that one might get from reading more narrowly defined texts (e.g., Yin's book on case study research or Strauss and Corbin's book on grounded theory). But it provides good strong overviews. I've been looking for a good solid text to anchor my grad-level qualitative research course this spring, and I think this one is it.

Blogged with Flock

Reading :: Becoming Qualitative Researchers

Originally posted: Wed, 13 Oct 2004 07:24:56

Becoming Qualitative Researchers: An Introduction (2nd Edition)

by Corrine Glesne

I recalled this book from the library because I had seen it mentioned so frequently in the qualitative research texts I've been reading. And the title, of course, marked it as a good introductory book for the graduate course I'm planning to teach. When I picked it up from the library last week, I was surprised by how thin and how friendly it looked. And as I read the first chapter on the way back to the office (a bad habit, I know), I thought: I have found a real winner here.

Due to other obligations, I put it away a while and finally read the rest of it today on the plane to Memphis. Unfortunately, the book didn't manage to sustain my initial enthusiasm, in large part due to its contrast to Creswell's book (which I finished on the same flight). Where Creswell's book gave us a broad scope of different methods, Glesne focuses on variations of ethnography. Where Creswell's book was tightly organized, Glesne's meanders. Where Creswell's book used focused, plain language, Glesne's often seems to be a forum for creative use of language. In fact, in Glesne's second to last chapter, "Improvising a Song of the World: Language and Representation," she makes much of the fact that she takes creative writing classes and encourages her students to render their research in alternative forms of expression such as poems, plays, etc. I am not particularly interested in this sort of reporting. If you are, you really should buy this book immediately.

I do appreciate how Glesne differentiates among traditional, critical, and feminist ethnographies and how she discusses action research. The discussion of postmodernism in research is more than worthwhile. Unlike Creswell, Glesne also includes a helpful discussion of research proposals and institutional review boards. And her discussion of computer-assisted data management and analysis, though shorter than Creswell's, is broader. In her best passages, she is engaging and exciting. (Of course, in her worst passages, she is self-focused, self-indulgent, and vague about specifics.) She dedicates a considerable number of pages to writing up the report, including advice from Strunk and White (use active voice; eliminate constructions such as "it is" and "there is").

Is the book worthwhile? It doesn't represent a big time investment. But I didn't find the payoff to be as high as some of the other books I've reviewed on the list. I wouldn't suggest using it as a class "introduction," though, because I don't think its meandering style will appeal to busy students. >

Blogged with Flock

(My own personal paradigm shift)

Originally posted: Wed, 13 Oct 2004 07:27:22

This weekend I had the realization that I haven't been managing projects well. So I'm in the middle of a paradigm shift. If you had looked at my computer and handheld on Friday, you might have been struck by the elaborate Aristotelian directory structure; same with my email program, in which I have carefully stored every piece of email since spring 1999. Same with my handheld, which not only had a dozen categories for pigeonholing events and todos, but also had an elaborate directory structure in its own right. But these categories, which I had conceived as ways to manage my data and my projects, were actually fragmenting my data and projects pretty severely. I could see parts, but not the whole picture. That played havoc not just with my personal projects but also with my delegation at the CWRL.

I seriously considered project management software, like Entourage, that would give me back the whole picture. But after test-driving Entourage, I rejected it: it basically did what I was doing already, throwing things into containers. Entourage might have allowed me to scale up, but ...

But when I want something on the web, I do what everyone else does. I Google it. Or I Clusty it. The last thing I want to do is go to a category-based site like Yahoo. So why not take that search paradigm and apply it to the rest of my life?

So I took a second look at some software my wife had asked me to test earlier this month. Instead of using MacJournal for notes -- a great program that allows you to store text notes in categories -- I transferred everything to Notational Velocity, a fantastic app that allows you to find-as-you-type (kind of like how your browser tries to guess the URL as you type it in). No categories, just searching. I downloaded Quicksilver, a launcher app that allows me to find any document or app by name and launch it -- just hit ctrl-spacebar, find-as-you-type, then hit return. All documents went into my Documents folder; the elaborate file structure went away. Same with my email client, which now has only a few basic categories rather than hundreds; it has an elaborate search engine built in.

Same story on my handheld. I deleted all the categories -- they don't sync with iCal anyway, for some reason. And I rediscovered the search function on my iPaq, which is actually quite well developed.

For more elaborate project management, I now use a spreadsheet. Projects in column 1; concrete steps in column 2; delegation, if necessary, in column 3; target date in column 4. The spreadsheet transfers to my iPaq too. The concrete steps can be copied and pasted into the todo list if appropriate, but I suspect that won't have to be done very often.

I'm also beginning to read 43 Folders, a site which someone (I forget who) recently described as time management porn.

This really is a significant change for me; let's see how it works out.

Blogged with Flock

Reading :: Handbook of Usability Testing

Originally posted: Wed, 13 Oct 2004 06:26:10

Handbook of Usability Testing: How to Plan, Design, and Conduct Effective Tests

by Jeffrey Rubin

I teach a usability course nearly every year -- this time it was on accessibility and usability -- and every year I dither about what book to use. Should I use Carol Barnum's book Usability Testing and Research, a book for which I have much admiration, even though I contriibuted perhaps the most awkward recommendation statement ever? Should I instead use Dumas and Redish's A Practical Guide to Usability Testing, a venerable old tome? Or should I pick up something by Jakob Nielsen?

This year I returned to Jeff Rubin's book, which has been around for a while but still seems to have the most concise and readable discussion out there. It doesn't have the high quality of examples that Barnum has, nor the up-to-date cachet of Nielsen's work. But it does get into the guts of usability methodology, and in a more rigorous way than the others do. And since my students have a usability lab for testing websites, something with a focus on methods seemed like the best choice.

Rigor, of course, is not something I normally associate with usability testing. Contemporary usability tests are descended from orthodox experimental psychology (in fact, Rubin's training was in that area), and it used to be that a "real" usability test was essentially an experiment oriented to a product, with N>=8 for each condition, multiple conditions, etc. (e.g., the Bellcore work on SuperBook). But different encrustations have developed, the most noteworthy being that testing often occurs with relatively small Ns and often no alternate conditions at all. In describing why the small N has caught on, Rubin repeats a classic fallacy first perpetrated by Jakob Nielsen. Let's call it the Easter egg fallacy.

The Easter egg fallacy goes something like this. Any given product has a finite number of problems with it. Chances are that you won't find every single problem with a single test, since users have varying levels of experience and varying backgrounds. But you'll find a good number of them. With N=5 or 6, you'll find upward of 80% of the problems. So it's not necessarily cost-effective to run tests with larger Ns -- at some point you're bringing in people who aren't really telling you anything you don't already know. Better to run multiple iterative tests with small Ns.

I don't have anything against multiple iterative tests with small Ns, but I do have a problem with the underlying logic. The assumption is that problems are hidden in the product, like Easter eggs are hidden on a dewy green lawn. How many can we find? But problems are not simply hidden in products; problems are problems because of how the product is enacted in a given environment by given actors. When programmers say with only a little irony that "it's not a bug, it's a feature," they're onto something: problems can be shifted out of their frames, applied differently (not necessarily even idiosyncratically), and can become solutions to still other problems. I recall a study sometime back -- by Rosson, I think -- of an email feature that included a bug that let people do something they wanted to do. That bug actually did turn into a "feature" in the next release, with not much more change than a write-up in the documentation. What I am saying is that the Easter egg fallacy is a structuralist reading, and it doesn't work as well as a constructivist one.

Most usability books have a version of the Easter egg fallacy. But to Rubin's credit, he does encourage larger Ns and multiple conditions, and he does take other measures to ensure rigor (as much as you'll get in a usability test, anyway). His writing is characteristically terse and parsimonious, and that extends to his examples of test plans and test reports. For the money, this book gives a quick, readable, and surprisingly complete introduction to usability testing.

Blogged with Flock

Reading :: Writing Winning Business Proposals

Originally posted: Wed, 13 Oct 2004 05:57:09

Writing Winning Business Proposals: Your Guide to Landing the Client, Making the Sale and Persuading the Boss

by Richard C. Freed, Joe Romano

Full disclosure: I took the grad level proposal writing course from Rich Freed back in 1994. It was my first semester in the Ph.D. program at Iowa State. The book was in press, so we used a course pack that was essentially the book's manuscript. I really wasn't so sure about the course: the things that Freed called proposals didn't look much like what I had been taught proposals were supposed to be (i.e., recommendation reports). And I'm a little embarrassed to admit that I didn't really "get" the methodology or warm to it until after the semester was over.

In fact, I don't think it really sank in until I taught the undergrad version of the course a few years later. I reread the book and suddenly it clicked. Perhaps that's because I had become more theoretically sophisticated by then, or because I had grown to appreciate method. Because, even though the book is written in the sort of gladhanding voice that one associates with business guides, it is quite sophisticated in its methodology. Although the business proposal is a relatively simple genre, it requires a complex -- and, more problematically, a sustained -- argument, one that will build bridges to several different stakeholders simultaneously. That's not easy, and it's next to impossible to do well the way that students are inclined to do things: procrastinate, sit down and outline the argument in one sitting, and give it a thorough copyediting. To be fair, it's also nearly impossible using the boilerplate and models of the professional proposal writer, although they tend to have experience on their side.

Freed's take is -- as I emphasize repeatedly to my students -- that writing is not expression (i.e., something that emerges from one's head) so much as it is construction, the careful building of a sustained argument out of good building materials. The first half of Freed's book is all about generating those building materials. Writing teachers may be taken aback by those materials -- which mainly consist of forms that students fill out and Aristotelian diagrams representing methods and qualifications subarguments. But they do work, they do act as enormously useful heuristics that interact to generate a multidimensional view of the many readers one's proposal can be expected to have.

Once these building blocks have been generated, assembling the proposal involves drawing heavily and in structured ways from them. It's surprising how quickly a rough draft can be assembled. Freed discusses how to make sure the argument is "aligned" and how to develop its style, displaying in the latter discussion the willingness to bend accepted mechanics rules if the result is a more persuasive document. Throughout, he reminds us that persuasion should have an ethical component: you find the best arguments, but argument building itself is a method of inquiry, a way to determine how your abilities fit the clients' needs. If you can't make that argument well, he says, you shouldn't make it at all.

In this second edition, Freed includes new and revised material: a method for guiding collaborative criticism of proposals, an expanded discussion of Situation, and a better final overview of the process. But the second edition also has the strengths of the first one. This book taught me much of what I know about sustaining arguments; I expect that I'll continue to use it for my proposal writing classes for some time.

Blogged with Flock

Reading :: Participatory IT Design

Originally posted: Wed, 06 Oct 2004 08:39:36

Participatory IT Design : Designing for Business and Workplace Realities

by Keld Bodker, Finn Kensing, Jesper Simonsen

Bodker, Kensing, and Simonsen have been involved in Scandinavian participatory design for a long time, and they've done a lot to work out a systematic participatory design approach. This approach, called MUST, is the descendant of the early and relatively unstructured PD work and was presaged by the Cooperative Design approach described in the 1991 classic Design at Work. Understandably, I was eager to see their new book. And as I read Participatory IT Design, I saw that they had indeed systematized and developed those participatory design themes while keeping the political and ethical orientation of the early Scandinavian work. The many techniques that were developed in the UTOPIA project and similar projects -- prototyping, future workshops, organizational games, observations, document and functional analyses -- are here, further developed and given coherence in the MUST approach. And MUST provides a far more worked out understanding of organizations, projects, and project phases.

And yet. Have you ever met a scholar whose work you greatly admired, perhaps at a conference, and gotten to know him or her? And have you suddenly realized, this person is incredibly boring? That's how I felt about 30 pages in. The authors have provided too much structure, too much detail, and far too many scenarios to keep things straight; I could tell that there's something to the method, but it gets bogged down in the many, many details and principles and phases. Further, the methods and techniques are separated from the phases, so we don't get a solid sense of how the two relate. I can't keep from comparing this book to Beyer and Holtzblatt's Contextual Design, which -- although I have my issues with it -- really does a remarkable job of explicating phases, methods, and techniques in a more integrated manner.

All in all, I expect to refer to Participatory IT Design quite a bit in my scholarly work. But I can't imagine using it in a classroom, as I regularly use Beyer and Holtzblatt's book.

Blogged with Flock

Reading :: Networks of Power

Originally posted: Mon, 04 Oct 2004 21:38:56
Networks of Power: Electrification in Western Society, 1880-1930
by Thomas P. Hughes
Spend any time at all in the sociotechnical studies literature (broadly speaking) and you'll run across the title of this book. Latour cites it, Law cites it, Bazerman draws liberally from it, and Pickering has a detailed critique. The book is a historian's examination of how Western society became electrified between 1880 and 1930, focusing on the US, England, and Germany. Since my current project (on telecommunications) has something to do with the construction of similar networks, I decided I had to read it. So I ordered a used copy from Amazon.
The book, frankly, is daunting because it is so immense. This thing is the size of a telephone directory -- 465 pages. It's taken a while to get through it, especially since I ended up reading other books in the meantime to avoid being bogged down. But the book is worth it -- even though I have to agree with Pickering's sharp critiques of it. Networks of Power provides an unusually thorough discussion of sociotechnical networks, supported by painstaking research and some strong theoretical work. Perhaps because he is a historian, Hughes often provides more than sufficient evidence to support his points, and sometimes I think the points get lost in the narrative, but I strongly suspect the difference is one of discipline rather than unfocused writing per se.
This book is rich enough that I'm not even going to try to summarize it. Instead, I want to draw out a couple of key points.
One key point is that sociotechnical systems -- which Hughes regards as cultural artifacts -- have what he calls unique styles. These styles are shaped by cultural factors such as "geographical, economic, organizational, legislative, contingent historical, and entrepreneurial conditions" (p.405). These factors shape the style of a sociotechnical system so profoundly that styles are easily recognized (p.443) and a system of one style simply can't work elsewhere if lifted and transplanted whole. Chapter 14 is devoted to this proposition and makes the case in great detail. The foundation of the case is laid much earlier, though, in Chapter 3: Hughes argues with great success that when one applies a sociotechnical system from one set of conditions to another (e.g., when Edison tried to reproduce his New York system in England and Germany), one must see the work as technology transfer rather than as simply applying the same solution elsewhere (p.47; see also Bowker's book on Schlumberger, Akrich's work with the gazogene, etc.). Hughes' real contribution here is in doing the hard work of demonstrating this point across the mentioned factors, providing us with several examples and with a set of factors to apply to other cases. I'll certainly be using them.
Another helpful point is what Hughes calls the reverse salient, a lag in development of part of the system (p.79). (Hughes prefers this term to "bottleneck" or "disequilibrium" because of problems with those two metaphors.) "The reverse salient usually appears as a result of accidents and confluences that persons presiding over or managing the system do not foresee, or, if they do foresee them, are unable to counter expiditiously" (p.80). One example included the problem of the high cost of transmission that was faced in the direct current system.
Now to the chief problem. Like Pickering, I think that Hughes makes a mistake in separating technology and politics. He argues that in Chicago, technology was preeminent; in London, politics were; and in Germany, both participated equally. The problem, of course, is that it's not especially illuminating to separate the two, as Hughes seems at the verge of admitting several times. As his analysis suggests over and over -- particularly in the chapter on style -- technology is political and politics are technological.
In all, this is a really useful (though overly long) study and I'll certainly be using it a lot.

Blogged with Flock

Reading :: Maximum Accessibility

Originally posted: Fri, 24 Sep 2004 20:12:55

Maximum Accessibility: Making Your Web Site More Usable for Everyone

by John M. Slatin, Sharron Rush

I've been coding websites since 1994, and I've always been concerned about simplicity and elegance. Where others used graphics, I used none; where others used GIFs for spot color, I used colored tables. My pages loaded quickly, scaled well, and were primarily text. So when I taught my first usability course at UT in fall 2002, I was confident that my stripped-down course site would pass muster in terms of accessibility. Then I read this book.

It's not often that a book is a life-changing experience. But this one changed forever the way I see websites and web development. And as I quickly, frantically began to recode my own site to better comply with accessibility guidelines (you can see that first effort at accessible design here), I also began to understand the uneasy relationship between structural markup and visual presentation -- and how vital it is to get down exactly how to mediate that relationship.

Before we get into exactly why this book is life-changing, let me note that John Slatin is my colleague and sometimes collaborator, and Sharron Rush is someone with whom I've discussed research. Both are great people, and perhaps that fact colors this review. But even if it weren't true, this book is really quite strong. The first half offers narrative accounts of different common problems that disabled users have with website features: graphics, tables, text rendering, plug-ins, pop-up windows, frames, and forms. The second half offers best practices for dealing with these issues. Both halves describe accessibility guidelines in a way that's easy to understand, and emphasize that the best solution is a single design that is entirely accessible while still being attractive and usable for those who are not disabled. Separate but equal, Slatin and Rush emphasize, wasn't a good solution in 1963 and isn't a good one now.

And that's really the crux of the book. More than offering an understanding of accessibility challenges, more than offering how-tos, the book makes a compelling argument that accessibility is a civil rights issue and that we must attend to it morally and ethically, not just in terms of regulations (although regulations are also a concern). Just as we provide curb cuts and ramps in our buildings, we should provide accessibility measures on our websites. And, the authors are quick to point out, such measures benefit everyone: just as curb cuts serve anyone who has to use a wheeled device (from wheelchairs to hand trucks), accessible sites serve anyone who has to use a website (from those with screen readers to those using text-only browsers to those using handheld devices and news aggregators). This book, and my continued discussions with Slatin, are what have led me to emphasizing accessibility in the CWRL and to explore its possibilities in my own development work. (My current website is a testbed for accessibility measures, among other things.)

To enact this civil rights focus, Slatin and Rush provide really remarkable insights, examples, and suggestions for redesign. The book is easy to read and follow, even for those without much HTML experience -- like my current students, who have just finished reading it for my accessibility and usability class. I'm happy to report that the book serves as a good textbook as well as provoking interesting class discussions -- discussions that mirror the ones you'll have with yourself as you read the book and imagine what it's like to use adaptive technologies.

Blogged with Flock

Reading :: Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory

Originally posted: Fri, 24 Sep 2004 19:04:06

Basics of Qualitative Research : Techniques and Procedures for Developing Grounded Theory

by Anselm Strauss, Juliet M. Corbin

The title of this book is somewhat misleading, but it's corrected by the subtitle. This book is a grad-level introduction to grounded theory in particular, not qualitative research in general: competing qualitative research traditions such as participatory action research, ethnography, and case studies are not discussed. Granted, they often have much in common with grounded theory, but there are also a lot of differences. Unfortunately they're often simply lumped together.

That was certainly the case in the research methods courses I took in my PhD program, several years ago. I'm sure that was out of necessity, but I'd like to more strongly differentiate them in the course I'll be teaching this spring -- and to get them straight in my own mind. Similar terminologies, similar methods, and similar aims tend to introduce a lot of slippage into discussions of qualitative research. And I should add that there's sometimes a frustrating lack of detail about exactly how to deploy them, a lack of detail that I think points to the depth of experience and the level of practice it takes to do qualitative research well. Diana Forsythe's last article, on how ethnography is something done by people with PhDs in anthropology (as opposed to something one can learn in a three-day contextual design course), resonates more strongly with me every time I read one of these books.

A little background: Anselm Strauss developed grounded theory along with Barney Glaser. They deployed it in several studies and wrote several books on it. This one is meant for grad students, which probably explains why so much of it involves expressing empathy for the reader's struggles, discusses ways of managing one's emotions about one's project, and dispenses general advice. Unfortunately a lot of that advice seems obvious and not terribly helpful. Here's a sample from the last chapter, on writing up studies:

Our best counsel is to choose if possible a supportive yet critical advisor, and to write as good a manuscript as possible. If you produce solid research you are likely to earn your degree, unless none of the committee members can counternance qualitative research. If that is a possibility, then you should strive to keep the number of such potentially adverse critics on your committee to a minimum. (p.237)

Frustratingly, much of the book is taken up with similar text. (I suddenly realize that this text reminds me of another book I've reviewed.) That's unfortunate, because there's a lot of really good stuff in here -- if you're willing to dig.

What good stuff? Well, the book manages to lay out the basics of grounded theory quite well. We find that grounded theory emphasizes the inductive development of theories, first as substantive (situated in a particular context) and eventually as formal (examined in many such contexts) (p.174). We learn that there are four criteria for judging how well a theory applies to a phenomenon: fit, understanding, generality, and control (p.23). We find that -- just as in participatory design, incidentally -- "the research question in a grounded theory study is a statement that identifies the phenomenon to be studied" (p.38). And over and over we find that grounded theory is all about "integrating detail, procedures, and operational logic" (p.159) -- developing coherent, well integrated, and strongly connected theories through progressive waves of open coding and axial coding.

But then again, we also find that we can't reduce the approach to a set of steps that will necessarily produce a grounded theory study. I don't think this is the fault of the prose; rather, I think Forsythe was right in that these procedures and the judgment behind them must be learned through deep experience. This book outlines the methodological precepts and gives a lot of examples, but ultimately it doesn't seem to pin down the approach with much specificity, and I think that's because the approach doesn't lend itself to that level of operationalization. Which does put us as qualitative researchers in a difficult position: Strauss and Corbin talk often of grounded theory's rigor, but as an inductive method that doesn't and can't provide strong operational guidance, it seems that grounded theory would have difficulty making such a case for itself. I'll have to think about this further as I examine other qualitative methods.

Blogged with Flock

Reading :: Opening Spaces: Writing Technologies and Critical Research Practices

Originally posted: Sun, 19 Sep 2004 10:04:34

Opening Spaces : Writing Technologies and Critical Research Practices

by Patricia Sullivan, James Porter

Reading Yin's classic on case study research last week led me to dig up my copy of Sullivan and Porter's Opening Spaces, a 1997 book from the Computers and Composition series that is perhaps the only computers and writing book to deal primarily with research methodology. I remember having a love-hate relationship with this book when I first read it. And I'm here to report that this relationship is indeed intact.

Let's talk about the love first. Sullivan and Porter do a good job of demystifying research, starting with the unusual and illuminating example of a basketball game: the game can be observed and described in different ways, using different methods, and from different perspectives. I can imagine grad students, and even undergrads, catching quickly onto the basics of research by the end of this chapter. Clearly, Sullivan and Porter understand research as inquiry rather than as a set of steps that will result in transcendent knowledge. This understanding of research is forefronted all the way through the book, resulting in some valuable insights about the design-research-report cycle. Great. Research as praxis -- one of the central themes of the book -- seems like a valuable theme to develop.

But then we get to the problems. Some of these are due to the series, which has low production values and I think has been edited with a rather light hand. Certainly the chapters are uneven, the prose sometimes gets bogged down (more on that in a moment), and the book as a whole doesn't follow the arc set by that strong first chapter. I expected a book on methodology. I got a book on, well, critical research practices.

So what do they mean by critical research practices? Well, there's a disorienting mix of postmodern uncertainty and foundational certainty.

The foundational certainty is in Sullivan and Porter's embrace of liberation theology's precepts (pp.118-128). In their adoption and amplification of Paolo Freire's work in Pedagogy of the Oppressed, Sullivan and Porter enthusiastically elaborate on forms of oppression, differentiate between oppression and power, and confidently declare, "Once oppression is identified, there can be only one ethical stance toward it: Oppose it" (p.122). We're not just talking about overt forms of oppression such as genocide in Sudan, but also covert ones such as computer interfaces: "The Macintosh interface may represent a form of oppression (cultural imperialism), but it's certainly better than the old DOS command-line interface" (p.122). They go on to identify the foundational principle, "liberate the poor," and confidently state: "Given conditions of fundamental inequity, or faced with a situation of oppression, this is the only ethical stance to take. It's the chief operational principle in such situations, and it's a principle not present in the U.S. Constitution" (p.123). I find this argument unconvincing because it assumes hierarchical relationships; suggests that conflicts necessarily have a dimension of oppression; and labels oppressors unambiguously as wrong. Do I really need to point out why these factors are problematic?

This litany of absolutist statements is generally confined to Chapter 5, the chapter on politics and ethics. It is sharply contrasted by other parts of the book, in which they give rein to postmodern uncertainty -- the starts and stops, the on-the-other-hands and the that-is-not-to-says -- and this makes for some joyless and tedious reading. This is partially a stylistic problem, I think, as the postmodernists cited by these authors (Foucault; Deleuze & Guattari) tend not to engage in these unending rounds of hedging.

Why they couldn't strike a balance between these two poles is beyond me. But then again, this book really is all about poles. In Chapter 4, they discuss "postmodern mapping," a vague term that essentially involves setting up pairs of binaries in a cartesian grid and eyeballing where different things should be situated in those binaries. As the authors explain in the concluding chapter, multiplying binaries is a way of complicating and questioning them: Western society is so used to binaries that this way is easier than trying to abolish them outright, they say. So we get poles in spades, poles that intersect other poles that intersect others in a grim recursion.

Well, I got a little carried away by the "hate" part of this "love-hate" relationship. Let me conclude with a little more "love": the book really does have some valuable insights and some valuable case studies, and it's a rare methodology book in the subfield of computers and writing. Chapters 1 and 3 are particularly valuable. I'll probably end up copying these chapters for my grad-level qualitative research class next spring.

Blogged with Flock

Reading :: Case Study Research: Design and Methods

Originally posted: Tue, 14 Sep 2004 18:49:08

Case Study Research: Design and Methods

by Robert K. Yin

This book is a classic handbook for conducting case studies. I remember reading it in grad school and again when I was preparing to conduct a study at Texas Tech; I particularly remember its impassioned defense of case studies as a legitimate and rigorous approach. So when I began preparing for my qualitative methods class (to be taught in spring 2005), I got a copy of the third edition.

In addition to an impassioned defense of the case study, Yin discusses what case studies are and are not. Case studies, he says, are not strictly qualitative; they can be quantitative as well. They are best performed in pairs or in greater sets, since this allows comparisons. They should have several types of validity. They can draw on six sources of evidence. You get the idea: the book catalogues the basic moves of conducting a case study with exhaustive taxonomies.

Yin illustrates each of these points with brief descriptions of classic or exemplary studies. None of these descriptions is especially captivating, to my mind, and neither is the book -- a workmanlike affair that makes its points, gets to where it needs to go, and gets out. Perhaps that is best for its genre -- a research handbook -- but I can see that I'll need to pair it with other, more interesting books if I use it in the class, which I probably will.

Case Study Research is not an exciting book. It does what it needs to do, though; there's a reason it's a classic.

Blogged with Flock

Reading :: Science on the Run

Originally posted: Thu, 02 Sep 2004 22:51:12

Science on the Run: Information Management and Industrial Geophysics at Schlumberger, 1920-1940

by Geoffrey C. Bowker

I have some nostalgia for this book. In 1997, while a PhD candidate at Iowa State University, I obtained a summer internship at Schlumberger Well Services in Austin. By the end of the summer, I was sure that I wanted to continue with the sort of research I had done there -- qualitative studies of knowledge work -- and I was also sure I wanted to end up in Austin. (It took a few years, but I reached both goals.)

When I returned to Ames, I ran across Bowker's book Science on the Run, an examination of how Schlumberger began and how it carved out a place as a technological innovator. At the time, I wasn't familiar with Bowker's work; I was just interested in Schlumberger and in the insights that this study of early research practice might inform my more recent study of software development practice. I quite clearly remember reading the book while waiting outside Rebecca Burnett's office to talk about the dissertation project.

Returning to it now, I find that I'm better equipped to appreciate it. Science on the Run is a study of the copious Schlumberger archives; it attempts to answer how Schlumberger developed its applied science while fighting for market share in a relatively new market. That market was logging oil wells; the question was whether advanced and innovative methods could provide better insight than simply drilling by guesswork. And Schlumberger's answer was to form electrical currents in the shafts, record the results, and use them to describe the strata through which the shaft extended. This was a fairly straightforward method except for the fact that, strictly speaking, it didn't work.

Let's qualify that, because the really important thing about this book is its discussion of how a locally contingent method became "bootstrapped" into a universal one. The electronic logging performed by Schlumberger worked quite well in the localized environments in which it was developed. However, geological features vary quite a bit from place to place, and consequently methods were not universally applicable: methods that worked wonderfully well in the lab (in a bathtub at the Ecole des Mines in Paris) or in one field (the played-out Pechelbronn, located near the Schlumberger family home) did not work as well in different geological areas (such as the southern US in the 1920s). Rival methods, such as seismic methods, carried the day in that arena because oil in those US fields was often bound up in salt domes, located easily by seismic methods but poorly by electrical ones. Schlumberger's methods were localized and contingent -- as all are -- but they had to be made to appear universal and able to reveal nature. Bowker engagingly discusses how that work was done: by cultivating successes in post-revolutionary Russia, where western technologies and electrical methods had positive associations, and where Schlumberger could act as a broker for other methods; in Venezuela, where oil companies would construct their own road, electrical, and pipeline networks to route around the local and turn the world into a suitable laboratory; and finally back to the United States, when seismic methods had failed to work in particular terrains and electrical methods had improved enough to function in a broader range of formations. Bowker also covers the patent battles, secrecy, deception, and espionage that attends such high-stakes, cutting-edge research. In this account, nature, society, science, and politics are reunited and demonstrated as part of the same activity.

A few points that are most relevant to my current work here. I was surprised and happy to find a precedent for the way I've been using a particular term:

From the point of view of the network builders themselves -- and from the point of view of a formal description of their networks -- there was no distinction between the act of organizing humans and that of organizing nonhumans; both sets of entities had to be aligned within the same network. I call this process of alignment convergence. (p.46)

Yes, "convergence" is a good term for describing how heterogeneous elements are spliced together in often unusual juxtapositions. And examining these heterogeneous, converged networks leads Bowker to what I have elsewhere called integrated scope:

In short, we should expect to find the same regularities at each level of granularity of the network -- from the micro level of individual operators, electrodes, and drillers' mud to the macro level of the oil industry as a whole and the social organization of campsites. The strategies are the same because in network terms almost exactly the same sort of thing is being done. (p.47)

In a previous review, I've already discussed Bowker's insights into the building of networks in Venezuela: pipelines, roads, electricity grids, and air networks were built to effectively route around local conditions, obviating them in terms of Schlumberger's operation (p.77). Bowker expands on that theme here, though this chapter is largely the same as that earlier article.

One other thing. In my investigation of telecommunication in the US, I've become really interested in the notion of universal service, a concept that has shifted quite a bit and that was originally developed by AT&T to ensure their monopoly status. Similarly, following Bijker and Pinch, Bowker discusses "trajectories," "something that can only exist after the fact, as a historical construction" (p.18). For instance, in the 1930s, the inelastic demand for oil (greater supplies did not lead to greater consumption) meant that big oil companies had to coordinate "to manage prices and output," without falling afoul of the Sherman Antitrust laws. The result was an interest in conservation of "a dwindling, precious natural resource" (p.71). In today's market, with ever-rising levels of consumption, the construct is no longer needed to manage production -- but now it has taken on a life and a meaning of its own. Another example is that of the State of California, whose legislature became convinced of the efficacy of electrical logging and made it obligatory. "Schlumberger had finally succeeded in laying down a natural law, so the legislature laid down a social law to back it up" (p.103). The clarity with which Bowker discusses this point is characteristic of the entire book.

Blogged with Flock

Game :: Metroid Prime

Originally posted: Wed, 01 Sep 2004 07:38:41

Metroid Prime

Every once in a while I review a GameCube game I've been playing. Previous games have included The Legend of Zelda: The Wind Waker and Eternal Darkness. Reading the reviews led me to believe that Metroid Prime would be similar. It, the reviews said, had a similar mix between action and puzzles; it had a similarly strong plot; it had visuals and audio to rival the other games.

Well, sort of. Metroid Prime, like Zelda, is the 3D heir of a traditionally 2D franchise. But whereas Zelda takes a third-person over-the-shoulder perspective, Metroid Prime is a first-person shooter game. And what I quickly learned was that although FPSes are fine for a quick game of Quake 3, I really don't like them for extended periods. I can't figure out where my feet are supposed to be. More than that, I don't get a sense of the character's personality. Watching Link in Zelda, or any of the characters in Eternal Darkness, enhances the narrative considerably for me -- to my surprise -- because I see them as characters. If I'm watching the character, even if I control their actions, I get a sense of their reactions and I develop an understanding of them as a character; I develop affection or respect or contempt or whatever for that character. But if I see everything in the first person, there isn't any separation between that character and me; and if I'm myself, where's the story? I'm just sitting in the living room blasting things.

That's not the only obstacle to the game's narrative, of course. Other problems exist, and I don't think they are just my own idiosyncracies. Although some reviews have claimed that Metroid Prime has an exceptionally strong story, it seemed paper thin in practice. Want me to boil it down for you? You have to pick up tools, kill anything that moves, solve puzzles, and collect artifacts. Oh, the ultimate goal is to defeat space pirates. Where's the complexity? I simply didn't find the narrative interesting enough. Yes, I think we've all enjoyed fiction with paper-thin plots, but the difference is typically made up with compelling characters. Here, there literally are no other sentient characters except you and a subset of your adversaries, and none of you are in a talking mood.

The final nail in the coffin is Metroid Prime's byzantine save system. In Zelda, you could save at any time; if you saved in the middle of a battle, you'd start again near the site of the battle, otherwise you'd start exactly where you left off. In Eternal Darkness, you could save as long as no active enemies were in the room with you. But in Metroid Prime, you must seek out far-flung "save stations." What's more, you must move through the sprawling, difficult-to-navigate terrain wherever your signals lead you, even if you've already been there several times before. I'm told this design is to help you become familiar with the terrain. Well, the terrain is tedious, and so is the gameplay. I gave up a third of the way through.

Next up? Probably Second Sight.

Blogged with Flock

Reading :: Competition and Deregulation in Telecommunications

Originally posted: Thu, 26 Aug 2004 19:17:45

Competition and Deregulation in Telecommunications: The Case for a New Paradigm

by Thomas Duesterberg, Kenneth Gordon

This 95-page monograph was coauthored by two members of the Hudson Institute, a not-for-profit research organization that makes recommendations on public policy. Duesterberg served as Assistant Secretary for International Economic Policy at the US Department of Commerce during the first Bush administration and, before that, as Administrative Assistant to Sen. Dan Quayle. Gordon served as Chairman of the Massachusetts Department of Public Utilities and of the Maine Public Utilities Commission. With that kind of pedigree, it's not surprising that the book is both knowledgeable about the subject and enthusiastic about telecommunications deregulation. In fact, its central thesis appears to be that the Telecommunications Act of 1996 (TCA96), though it contains some deregulatory impulses and language, is not deregulatory enough and preserves managed competition rather than moving toward a free market. With this point in mind, the book provides a valuable critique of TCA96 paired with a really interesting historical perspective.

That historical perspective isn't of TCA96's repercussions -- the book was written only a year later -- but of the negotiations that led to it. Duesterberg and Gordon discuss a couple of things in detail. One is access charges, which "stem directly from policies designed to keep local-exchange rates low enough to meet political needs" and "to keep local rates low -- that is, to cross-subsidize them" (p.29). This arrangement worked fairly well as long as AT&T was a monopoly: access charges were leveled on long-distance charges (a relatively small fraction of wealthy users) and transferred to the local side of the business, keeping local access cheap, and since all that money stayed in the company regardless, it was a zero-sum game. (I oversimplify here -- an economist would point out that price fluctuations have tremendous effects on demand in what are actually separate markets. See p.33.) This arrangement, of course, works only as long as both services are controlled by the same monopoly (p.31); when other long-distance companies began competing, they could provide more competitive prices, partially because they initially didn't have to pay a local subsidy (p.51). More on this in a minute.

Access charges weren't the only form of cross-subsidy. Another was "the practices of capitalizing nonrecurrent expenses and using depreciation rates that fail to recognize the real rate of decline in the economic value of assets" (p.31). This practice of underdepreciation, the authors charge, resulted in "tomorrow's ratepayers underwriting the consumption of today's ratepayers" as well as underinvestment in facilities. "If competition is introduced ... the very recovery of the as yet underdepreciated amounts may be at risk": competition upsets this logic and destabilizes this second form of cross-subsidy (p.31).

Thus, the authors conclude, "an enormous subsidy burden was building" throughout AT&T's monopoly from the 1930s-1970s, a burden that "would haunt later regulators and enormously complicate their attempts to introduce competition as it became apparent that little or even none of the industry any longer exhibited natural monopoly characteristics" (p.32).

This discussion leads to the notion of universal service, which in some ways got us into this mess. "As currently applied under governmental policy, universal service refers to a system of direct and indirect subsidies that essentially maintains basic residential service at rates deemed affordable by regulatory authorities," the authors state (p.48). But -- despite the arguments of universal service's supporters -- the concept does not date back to the early days of telecommunication. "Instead, the term universal service was coined by the visionary head of AT&T, Theodore Vail, for the altogether mundane purpose of eliminating competition in telephone service and creating a monopoly for his company" (p.48). The term was coined in 1907 (p.49) and meant the need to unify telephone services under regulated local exchange monopolies -- a very different meaning than the contemporary one (p.49). The term doesn't even appear in the Communications Act of 1934 (p.50).

The notion of universal service led to the access charge cross-subsidies of which I spoke earlier. This arrangement worked well until others moved into the long-distance market and the Department of Justice filed an antitrust suit against AT&T in 1974. At this point, Eugene Rostrow, an AT&T consultant, testified to Congress in 1975 that universal service was tied to the practices of regulated monopoly (p.51). This notion wasn't spelled out in the Communications Act of 1934:

Unable to locate either the term or the concept of universal service in the text of the 1934 act, Rostrow instead appealed to the vague language in the act's statement of purpose, much in the spirit of the 1970s that discovered large concepts in the "penumbra" of founding texts.

Though the authors don't say so explicitly, this sounds like a reference to Roe v. Wade.

The point that leaps out to me here is that we have to be very careful when talking about cultural-historical development. In this case, certainly a concept called "universal service" developed historically; but the contemporary meaning becomes read back into the history, making it seem as if the kernel of the meaning has been there all along. But that contemporary meaning could not have existed in the original -- indeed, we see an interesting example of what I've been calling splicing here, in which contemporary legal trends are read back into the documents that predate them. As the authors say with some regret, "Historical accuracy matters less, unfortunately, than the political reality of universal service as currently understood, for the concept was explicitly enshrined in the Telecommunications Act of 1996" (p.52); the notion of universal service had been stabilized, nailed down, and now it undergirds TCA96.

Blogged with Flock

Reading :: Communications Policy and the Public Interest

Originally posted: Wed, 25 Aug 2004 19:53:26

Communications Policy and the Public Interest: The Telecommunications Act of 1996

by Patricia Aufderheide

If you were to write a short sociotechnical history of the US telecommunications industry -- and alas, I plan to do just that fairly soon -- there are a few large events you would have to address. One is the establishment of the Bell monopoly, more or less enshrined in the Communications Act of 1934. Another is the 1982 Modification of Final Judgment that broke Bell into AT&T (long distance, competitive) and the Regional Bell Operating Companies (RBOCs: local area, regional monopolies). And the third is the Telecommunications Act of 1996, which (among other things) opened up local telephone service to competition and mandated that RBOCs share their network infrastructure with competitors until those competitors could build their own facilities.

Facilities are what this fight is all about, of course. Bell established its monopoly in part by building the infrastructure when building infrastructure was relatively cheap. Imagine having to build your own facilities to compete in the local phone market: arranging for right of way, erecting telephone poles, laying cables, setting up a local loop, wiring jacks into people's houses, hiring workers to troubleshoot the physical network (which constantly decays and must be repaired). Ideally competing companies can do all of these -- and indeed they're beginning to do so -- but they couldn't be expected to do all that at once, and even if they did, consumers might have a hard time trusting them. But for a long time the question was moot, because Bell would let nothing -- nothing -- attach to its network. In fact, in the 1950s Bell sued a company that manufactured the "Hush-a-Phone" -- a small plastic cup that you attached to the phone mouthpiece so that you could speak softly (pp.21-22). Bell claimed that even this plastic attachment threatened the integrity of its network; the repercussions of this foolish claim led eventually to our present state, in which RBOCs must let companies attach to their networks. In a way, Bell was right: Hush-a-Phone contaminated and changed the nature of the entire network.

Bigger changes were in the offing, though:

Digital processing has changed the very characteristics of communications networks. Rapidly evolving computing that is based on distributed processing has made it possible to decentralize networks. Many of the decisions once made in large centralized switches are now made at intermediate stops or even within the consumer's telephone. Along with increased flexibility and the potential to reconfigure the very shape of networks and subnetworks, decentralized data processing has dramatically increased the amount of intelligence -- or the ability to respond to input and take action -- in communications networks. This innovation provides a fundamental challenge to the notion of common carriage, or the restriction of network providers to transmission alone, because the clear lines between content and conduit have become muddied. Networks themselves have information, or content, built into them. (pp.10-11)

That's the present state. And it's what allows the snarl of telecommunications infrastructures -- their name is Legion -- to work together coherently and appear to provide unified, uninterrupted service. Not by themselves, of course, but through the negotiated and legislated agreement known as the Telecommunications Act of 1996. The Act is examined and dissected in this book in terms of its history, development, negotiations, and repercussions. Aufderheide is not a sociotechnical theorist, at least not in the terms of the sociology of knowledge, but she does a good job of examining the workings of the Act and providing us with enough information to begin such an analysis.

For instance, she dissects the notion of universal service, a chimeric term that has been reinterpreted throughout the history of US telecommunications. The 1934 Act, for instance, stated its overarching purpose as "to make available, so far as possible to all the people of the United States, a rapid, efficient, nation-wide, and world-wide wire and radio communications service with adequate facilities and reasonable charges" (qtd. on p.18). AT&T raised long-distance prices to pay for investment in universal service -- "an important claim of social benefit, which could discourage adoption of a competitive model of service provision" (p.18). Of course, once a competitive model was introduced, AT&T's strength became its weakness. "In 1969, MCI won the right to attach to the AT&T network in order to offer what then were merely private network services for corporations" (p.22), and that attachment was free of universal service requirements. "The decision fundamentally challenged the old logic of cross-subsidy. AT&T charged the upstarts with 'cream skimming' -- taking the high-dollar clients and leaving AT&T with the large, expensive network to service" (p.22). Eventually, after the MFJ, universal service became a contribution from long distance service to local service (p.57).

This conflict in part drove the move toward deregulation, an attempt to move from the monopoly model to a competitive one. But how to maintain the commitment to universal service? One deregulative idea was for the government itself to offer "phone stamps," modeled on food stamps, directly to disadvantaged consumers. Imagine. Another, more decentralized version was described in Peter Huber's 1987 study The Geodesic Network: In a decentralized network, "each supplier can independently provide end-to-end service, and yet suppliers as a group can offer universal access. by providing horizontal links among their networks" (qtd. on p.29). Eventually, universal service was operationalized in the Telecommunications Act of 1996 as subsidies provided by providers to K-12 schools, libraries, and similar nonprofit parties at discounts of 20% to 90%. The discounts were set based on "poverty level, as measured by school lunch eligibility" (p.100) -- a wonderful illustration of how different sociotechnical networks can be spliced together.

Another interesting case is that of local service. In the MFJ, the RBOCs had been granted limited monopolies in local service. But with the Telecommunications Act of 1996, competitive local exchange carriers (CLECs) were allowed to compete. Ideally, as I mentioned earlier, these companies would build their own infrastructure; realistically, the price of entry was far too steep. So the Telecommunications Act mandated that in the short run, RBOCs must provide their services at a discount to the CLECs, which could then repackage that service and resell it to consumers at competitive prices. Eventually, the Act envisioned the CLECs bootstrapping themselves: building their own physical facilities, providing hardware-to-hardware or facilities-based competition (p.44). The Act "leaves to the FCC and to states (in an unclear formulation) the determination of what price is fair for resale"; mandates number portability; disallows mandated access numbers; and sets up payment arrangements for calls that traverse multiple competitors' networks (pp.63-64). Of course, once again the competitors cream-skimmed: By 1998, CLECs controlled 1% of local lines, but those lines represented 3% of local phone revenues, because the CLECs aggressively pursued business rather than residential lines (p.84).

I enjoy reading histories because they give interesting insights into how our world became the way it is, and this one was in general engagingly written. Aufderheide injects occasional flashes of humor that keeps the book lively. More importantly, the book keeps track of stakeholders, their influences, and their negotiations as the book moves inexorably to the passing of the Act and its repercussions.

>

Blogged with Flock

Reading :: The Wisdom of Crowds

Originally posted: Tue, 17 Aug 2004 19:24:20

The Wisdom of Crowds

by James Surowiecki

I picked up this book partially because it touches on some themes that are common in the books I've been reading lately -- distributed cognition, the sociology of scientific knowledge, etc. -- and partially because Dan Drezner recommended it. (My wife tells me it was also discussed on NPR.) And indeed it does provide some interesting insight into the issue of group decision-making. But not quite in the vein I expected.

To understand why, here's the money quote -- which I lifted from the book's website but which echoes the message of the book:

Crowds are best when there's a right answer to a problem or a question. (I call these "cognition" problems.) If you have, for instance, a factual question, the best way to get a consistently good answer is to ask a group.

This focus on problems with "right answers" is what separates Surowiecki's book from folks like Latour and Machiavelli, and even distributed cognitionists such as Hutchins. Like Latour and Machiavelli, Surowiecki argues that "crowds" can often make better decisions than individuals, even if the individuals are not terribly well informed or intelligent; as long as their input is aggregated well, decision-making turns out to be far more accurate even than the experts'.

Now, to my mind, this thesis is severely flawed in that it is explicated by reference to problems whose answers can be retrospectively verified, typically with quantitative measures. Examples include guessing a heifer's weight, the coordinates of a lost submarine, or a political candidate's margin of victory. Surowiecki explicitly disallows the question of deliberation (though he sneaks it in through the back door) -- the focus is on aggregations of guesses, the closing in on an objective solution, not the forging of open-ended political settlements. "Choosing candidates and making policy in a democracy are not, in that sense, cognition problems and so we should not expect them to yield themselves to the wisdom of the crowd" (p.270). It's not surprising to find that Surowiecki was led to research this book through his examination of market behavior. But this is an entirely different kind of "crowd wisdom" than, say, Machiavelli's discussion of republican deliberation. In Machiavelli's view, a "crowd" in a republic will almost always make better decisions than a prince, not because more people are working on a cognitive problem, but because they introduce deliberation and moderation -- exactly what Surowiecki disallows. Even Hutchins' discussion of distributed cognition is less closed and less goal-oriented than Surowiecki's.

This book is valuable for discussing market economics and probability, as well as for providing an opening for discussions of distributed cognition. But given my particular research interests, I don't anticipate cracking it often.

Blogged with Flock

(Books for this political season)

Originally posted: Fri, 13 Aug 2004 09:06:13

As we gear up for the Republican convention and a fall campaign that will quite possibly be close and brutal, I've assembled a "top ten" list of books for this political season. These are in no particular order, and they include plenty of books that don't seem obvious. They also exclude some obvious picks, such as Machiavelli's The Prince and Sun Tzu's The Art of War. And no, they don't include Joe Trippi's recent book or Extreme Democracy, neither of which I've been able to read yet. But these selected books all have direct bearing on aspects of this year's presidential election.

So here they are:

1. Kaufer and Butler. Rhetoric and the Arts of Design.

Kaufer and Butler's discussion of rhetoric as a design art is quite valuable, especially when applied to the Lincoln-Douglas debates. The authors provide a really useful taxonomy for structurally analyzing those debates -- and I find myself applying it to the current candidates' speeches as well. Although Douglas was eventually declared the winner of his debates with Lincoln, Kaufer and Butler successfully argue that he overrelied on tactics, underinvested in a persistent rhetorical strategy, and underestimated the impact of print media on the campaign. Look for parallels in the upcoming debates.

2. Machiavelli. The Discourses.

While Kaufer and Butler provide a close analysis, Machiavelli's analysis of the Roman republic has a much larger sweep. Machiavelli, an ardent republican (small R, of course), discusses in detail what makes a republic work and what kills it. This book has plenty of advice that is pertinent to this political season. President Bush could have benefited from his advice not to trust exiles, for instance. Most interesting to me is Machiavelli's contention that republics almost always make better decisions than princes.

3. Latour. Pandora's Hope.

Latour was heavily influenced by Machiavelli, so it's no surprise that he draws a similar conclusion. In two consecutive chapters in Pandora's Hope, he analyzes the debate between Socrates and Callicles from The Gorgias. Too often, he says, we want to take sides in the debate: Might (Callicles) or Right (Socrates), power or knowledge, earth or heaven. One leads to tyranny, the other leads to the philosopher-kings lauded in Plato's Republic. But even as Socrates and Callicles were hashing out this choice, the democratic Athenians were enacting their own solution: negotiating, haggling, bargaining, and compromising to yield a workable solution. The debate between Might and Right, Latour charges, is at its core antidemocratic -- and foolish in its tunnel vision.

4. Surowiecki. The Wisdom of Crowds.

Latour and Machiavelli agree that crowds make better decisions than princes or kings. So does James Surowiecki, in this book -- which I'm still finishing up. I picked up Surowiecki's book after reading the recommendation on Dan Drezner's site, but my wife also heard a blurb about it on NPR. Like Latour and Machiavelli, Surowiecki argues that "crowds" can often make better decisions than individuals, even if the individuals are not terribly well informed or intelligent; as long as their input is aggregated well, decision-making turns out to be far more accurate even than the experts'.

Now, to my mind, this thesis is severely flawed in that it is explicated by reference to problems whose answers can be retrospectively verified, typically with quantitative measures. Examples include guessing a heifer's weight, the coordinates of a lost submarine, or a political candidate's margin of victory. Surowiecki explicitly disallows the question of deliberation (though he sneaks it in through the back door) -- the focus is on aggregations of guesses, the closing in on an objective solution, not the forging of open-ended political settlements. So the real value of The Wisdom of Crowds this political season, I think, will be in the examination of polls and voter data.

5. Zuboff and Maxmin. The Support Economy.

Voters have changed considerably over the years, of course, and Zuboff and Maxmin suggest that this can be explained partially through a paradigm shift in the economy. Citizens, they say, are looking for more individualized choices. That's true in the market, their primary focus, but they suggest that the shift can also be seen in how citizens have left large political organizations in favor of smaller a la carte movements. Whereas our parents might have been willing to subsume their political will to an organization, letting (for instance) a union leader speak for them, we are more apt to want to represent ourselves. Witness the success of Howard Dean's online fundraising -- and, I would argue, its swift collapse. And notice also the proliferation of blogs representing constituencies that can't be catalogued using the traditional left/right dichotomy. This phenomenon led to bloggers at this year's conventions as well as the rapid fact-checking of older media.

6. Bakhtin - The Dialogic Imagination.

Of course, the explosion of political blogs reminds us of something we too often forget, which is that the Republican and Democratic parties are themselves fragile and contentious coalitions. Discussing what the "typical" Democrat or Republican thinks leads to straw person arguments because the discussion rests on the notion that each party has core principles to which all members subscribe -- typically something you can put on a bumper sticker. Bakhtin's most read book provides a valuable corrective to that notion by reminding us that different dialogues are always going on, combining and coexisting perhaps, but not necessarily synthesizing in the sense that dialectic would have us imagine. What a timely message as both candidates begin their attempts to chip away at the constituencies of the other party.

7. Wertsch. Voices of Collective Remembering.

Bakhtin, of course, labored in obscurity in the Soviet Union -- served often by his ability to hide his un-Soviet theorizing in Marxist language. He invoked the dialectic as a sort of trope for dialogism. That protective reframing was, of course, widespread practice in the Soviet Union. In Voices of Collective Remembering, Wertsch helps us to examine the false histories, the double consciousness, and the clandestine discussions that would occur in the USSR. Furthermore, he suggests that in some cases these harsh conditions actually resulted in more critical examinations of history than here in the US. This book reminds us of what is at stake in telling history, and suggests to me that the current wild diversity in the blogosphere could help us to become more critical consumers of news and history.

8. Jungck. Future Workshops.

As we consider new computer-mediated opportunities to expand our political participation (what some are calling "extreme democracy"), we might recall an older, more time-consuming aproach meant to route around accreted political-bureaucratic structures. Robert Jungck's approach, future workshops, involved recruiting interested citizens for intensive brainstorming, problem solving, and consensus building. The point was to avoid the groupthink and blind allegiance that characterized the Nazi regime as well as the mindset of the early Cold War. The H-bomb horrified Jungck; he believed that if he could get ordinary citizens to invest themselves in decision-making, real change could occur. Future workshops later became integral to the participatory design approach, in which workers strove for political and technological change in their workplaces.

9. Latour. War of the Worlds.

And that brings us to this slim pamphlet, the second entry of Latour's on this list. Based on lectures Latour has given over the last several years, and framed by the events of 9/11, War of the Worlds suggests

that these events should spur a reexamination of how we think about other cultures. For too long, Latour says, we have approached other cultures through either ethnocentrism or multiculturalism -- two sides of the same coin, he argues. Our mistake, he says, is to posit that there are several cultures but only one nature. So we may "respect" other points of view, but deep down we believe there's only one correct one -- ours, since we are expert at apprehending nature as it really is. Nonsense, Latour says. There are just as many natures as cultures: in a later book, he uses the term "natureculture" to drive home the point that any separation is artificial. And if that's so, we should earnestly negotiate settlements with others to agree on a world we can coinhabit. That doesn't mean giving up on our cherished values and beliefs, but it does involve understanding that theirs may be just as cherished and valued.

10. Law. Aircraft Stories.

Finally, along the same lines, we get to a book to which I gave an uncharitable review a while back. I don't regret that review -- but let's give credit where it is due. Politics become instantiated in the material environment, and not just in the obvious ways discussed by so many cultural theorists. In his study of how aircraft were designed and built for the British military, Law demonstrates quite well that a variety of constraints guided this complex work. Among those constraints were the representations of the Soviet Union, its strategies and tactics, its technologies and pilots. These were just as important and concrete as wind shear and other "natural," "objective" constraints. Examining Law's argument here helps us to reexamine the supposed split between politics and technology, politics and nature, politics and work.

Blogged with Flock