Comps readings this week Hertzum, M. (2002). The importance of trust in software engineers’ assessment and choice of information sources. Information and Organization, 12, 1-18.
Not a hundred percent sure this lives up to its title. This was a study of about 11 people on a software project - and based on conversations I've heard on software engineering where I work (with people who do know about this sort of thing), this was software development and not truly engineering. Anyway, this article dug into the trustworthiness to add that it's also important that the information seeker has enough to determine if a source is trustworthy. This study reports the results of observation of 16 meetings, 11 interviews, and reviews of project documentation. Cost is also not really very important at all when compared to quality as a measure.

Bates, M. J. (1990). Where should the person stop and the information search interface start? Information Processing & Management, 26, 575-591 (looks like also available here: http://www.gseis.ucla.edu/faculty/bates/searchinterface.pdf, but I have a photocopy of the journal pages.
I am such a big Bates fan girl. She starts with

Much of the advanced research and development of automated information retrieval systems to date has been done with the implicit or explicit goal of eventually automating every part of the process... An unspoken assumption seems to be that if a part of the the information search process is not automated, it is only because we have not yet figured out how to automate it... The implicit assumption in much information retrieval (IR) system design is that the system (and behind that, the system designer) knows best... the system controls the pace and direction of the search... but not all searchers want that kind of response from an information system.

Eureka - exploratory search, anomalous states of knowledge, hcir, human information interaction... yes, of course! I think this is why engineers *love* Engineering Village's facets - it gives them a view into how the system interprets their query, teaches them what it knows, and gives them power to control it.

Do systems provide any more support for strategy than they did? Any more strategic advice? I'm thinking no.


Bates, M. (1989) The design of browsing and berrypicking techniques for the online search interface. Online Review, 13, 407-423.
At some point, I looked, and I had been assigned this article in like 5 different classes. Not only do I have a copy in every binder, but we actually had this journal in a bound volume in our collection. I was so close to cutting the article out with an exacto knife before the volume went in recycling, but I chickened out. This article pulls together evidence from a bunch of places to show that information behavior is frequently an evolving process, with learning, refocusing, and chainging of strategies along the way. Different strategies might be following citations backward or forward, running through a journal that appears to have good stuff, browsing a place on the shelf, looking for more from a particular author, etc. Apparently, way back, Garfield had a hard time convincing librarians and others that people wanted to follow citations - librarians thought subject access was all people needed. She argues that systems should enable searchers to do all of these different types of searches.
Digital libraries should allow readers to jump back and forth between the article text and the references, see the section headings in advance and be able to jump directly to them (like to methods or conclusion), see what cites the article, browse journal table of contents, browse classification schemes and jump up, down, and across the hierarchy,
It's interesting to think of what journal platforms do this stuff well (and others that don't do this stuff at all). She also talks about trying to make an equivalent for flipping through a book and reading random passages to see if you like the author's style - google books allows this, but proprietary ebook systems typically do not. She also says that systems should allow the user to take notes, highlight, and clip interesting pieces to save for later and use off line (or outside of the source)... some do this, but not as well. The articles on digital libraries by Soergel still suggest these things years later, and yet we don't always have them.

Started re-reading Rogers, E. M. (2003). Diffusion of innovations. 5th ed. New York: Simon & Schuster.
Excellent book and Rogers' stuff is so readable. Quite long, though. I think I said this on this blog before, but the current edition is really on crappy paper with a very thin cover. Like newsprint inside. I bought mine in probably 2006 and kept it out of sunlight, but it's still discoloring with age. I hope it doesn't be come brittle.... when I actually *buy* a new book, I expect it to last! Anyway, like other books, this will get its own post.

Labels: comps

  ¶ 12:07 PM| |cites (technorati) |