Looking for something related to blogs' structure (narrative, formal, anything) I stumbled across the draft of the paper "A Brief Overview of the Linguistic Attributes of the Blogosphere" by S. Nilsson (downloaded it from the author's blog.)

This draft contains some interesting points but definitely needs editing and revising, especially method and results sections. It's hard to put them together and provide a coherent summary. Below is my attempt to do that.

Blogs are a combination of monologue and dialogue; they are self-reflecting entries and invitations for conversation. This dichotomy provides base for a hypothesis that a unique variety of language has evolved from written and oral communication to serve the purposes of this virtual environment.

Purpose of study

• to identify the structure of weblogging community;
• to determine if a separate variety of language exists in this community

Method
The study employs methods of social network analysis with a particular focus on language and oral/written discourse dichotomy.

The paper has five sections: history and explanation of weblogs; ethical issues in the online research; blog linking as social networks; linguistic analysis of the blog post; discussion of the possibility of a new variety of language found within the weblogging communities.

Blogs have its own format that makes them different from other types of Web publications; blog writing has a distinctive authoritative voice; blogs create their own community. Blogs can be compared to the Wunderkammer - "a random collection of strange, compelling objects".

Ethical issues of whether blogs are public or private domain, whether a researcher should reveal/conceal real and virtual identities are quite irrelevant to the argument. Of course, these issues are important for any research but they should be mentioned only if something unusual happened and has been resolved during the research.

Blog communities are considered in general as cohesive groups to which people feel they belong (Milroy, L. (1987) Language and Social Networks. Oxford: Basil Blackwell Ltd.) or in a way more related to blogs as "groups committed to a knowledge domain and producing information artifacts (through the software) that support shared goals (Eismann at the BlogTalk conference.)

Results

To measure the sense of community the following was accomplished:

1. Blogrolls of 75 blogs were mapped and visualized. It is not clear if all links were mapped or reciprocal links only. It is also not clear how the centrality of certain blogs was established. However, as it turned out the community of these 75 blogs of researchers interested in online environments is very dense with several cores.
2. Types of links were divided into links of solidarity (links in posts and in blogrolls) and links of information (links in posts but not in blogrolls.) Four (!) weblogs were then analyzed (no background statistics as for dates and periods of posts, length, etc. is provided.) Blogs with high rate of solidarity are in the center of the network (how these blogs were chosen anyway?)
3. Plural personal pronouns ("we / us" as opposed to "they / them") were then calculated for these four blogs. The in-group pronouns rate is lower than out-group pronouns rate. The table with the data as well as the interpretation of it are not clear.
4. The last procedure based on calculating keywords specific to this blogging group seems to be the most complete, understandable and interesting. In blogs the most frequent words are blogging jargon and academic jargon. It means that these bloggers associate themselves with blogging and academic communities quite well. Also, these bloggers use frames (meaning they place their writings within a large context and readers have to know this context to able to understand what is the conversation about.)

The last part about Blogspeak as an amalgamated oral/written variety of language derived from Netspeak (Crystal, D. (2001) Language and the Internet. Cambridge: Cambridge University Press.) discusses issues that has been discussed quite a lot in previous literature.

In 1998 and 1999 Stanfod University and The Poynter Institute conducted a study of news readers based on the eye tracking (the overview of the results can be found here.)
They used SMI EyeLink eye tracking system and special software developed by Stanford.
There were 67 participants who looked at the screen coded in different ways: by format (headlines, banners, photos, articles, briefs), by topics (crime&disaster, science&medicine, local, etc.), by events (clicks, typing, etc.) and by visits (urls, providers, etc.)
Some of the interesting findings:

1. Strong preference for text over graphics as entry points: among the first three glances readers focus on text twice as more often than on graphics (due to limitations of the study it's impossible to say why.)
2. Tendency to ignore artwork: among 504 captured glanced only 112 included artwork.
3. Banners noticed: of 97 pages that included one or more banner ad, 39 pages attracted at least one of the readers' first glances to a banner ad.

The main limitation of the study is that as it has been shown some information absorption takes place beyond the area considered within an eye fixation. It means even if one didn't fixate one's eyes on something, it's still likely that the information was absorbed.

Finally, I am catching up with literature about blogs. Academic papers are harder to find, but there is abundance of articles about blogs written by (prominent) bloggers.

One of them "What We're Doing When We Blog" is written by Meg Hourihan in 2002 (06/13/2002.)

Much water has flowed under the bridges since 2002 but the enthusiasm and emotional involvement is pretty much the same.

The basic argument is the following: More and more articles about blogging appear these days. Most authors are in "unenviable" position of describing blogs based on observation, not experience. [I wonder what if scientists always tried to experience what they studied?] Those who are not bloggers don't get it and bloggers themselves should talk about their own evolution.

Common ground for blogs - the format (short posts, reverse chronological order, permalinks.) It shapes all social interactions associated with blogging. Bloggers use blogs to link to friends and rivals and to comment on what they're doing.

It's interesting how bloggers are contrasted with non-bloggers. "We blog ... we're united ... we use tools ... etc."

In blogs readers always deal with the newest information (sense of immediacy.) Organization of information is independent of its topic. Blogger is not what he writes about, it's how he writes ("frequently, ad nauseam, peppered with links.")

When I am reading manifestos like this one the word "graphomania" keeps coming to my mind (definition: a passion or urge to write; also called scribomania.) The question is if there were always so many people with urge to write or is it the Zeitgeist...

The title of the qualifying paper defended today in SLIS by Kathryn La Barre is long:
Faceted Maps of Knowledge and Domains: "Peeling the Onion of an Idea," an Examination of the use of Facet Analysis in Website Design.

Qualifying paper is basically a research agenda for the doctoral dissertation. Below is the short summary of the presentation.

The facet classification as an application for libraries goes somewhere back to the beginning of the XX century. The current timeline starts with the post by P. Merholz "Innovation in classification", followed by an extensive discussion that ended up in creating a faceted classification discussion list.
What it actually means is that information architects (who are web development practitioners clearly distinguishing themselves from designers and programmers) turned to the good old theory to guide their work.

There are two examples of using faceted classifications in webdesign: faceted search and faceted navigation (Images of England and CompUsa correspondingly). The former one allows setting a number of parameters for search, the latter one provides browsing by many categories.)

Research questions:
1. How has Facet Analysis been used in the architecture of websites?
2. Does the use of Facet Analysis mark a shift in the base stucture of websites?
3. Might the current use of Facet Analysis on the Web demonstrate its potential for Information Retrieval and other areas?

Faculty members made some observations that these questions are broad and it might be necessary to narrow them down as well as to operationalize what is a shift and the base structure of websites.

Methods:
- Content Analysis
- Survey
- Interview
- Usability testing
- Transaction log analysis

The theoretical framework for this research is B. Hjorland's domain-analytic approach (JASIST, 53, 4, p.257.)  It might be a valuable framework (I need to read the paper closely myself as it talks about discourse communities a lot) but Kathryn didn't give a good answer on the question "How your research benefits from this framework?"

As in many cases with web studies, it seems that this research will require a lot of efforts to connect simple statistics and practical evaluations of websites with "real" theory. At the same time, it will be very interesting to see the results of this research.

The paper "I'm Blogging This": A Closer Look at Why People Blog" submitted to the Communications of the ACM is more of exploratory nature. It is rather an essay than research. It gives some insights and generates ideas but doesn't provide any rigorous and scientifically justified results.

The paper reports results of the ethnographic study of so called "ordinary" bloggers.

The goal is to uncover range of motivations for creating and maintaining blogs and to provide suggestions for improving current blogging tools.

Sample:
23 people, 16 men and 7 women. Initial list created by Google search on Stanford University portal, then the sample is "snow-balled" by asking initial bloggers to name other bloggers.
Method:
Interviews with bloggers (no questionnaire provided), text analysis of blog posts (no coding scheme)
Findings:
Major motivations for blogging: documenting the author's life, providing commentary and opinions, expressing deeple felt emotions, working out ideas through writing, and forming communities.

These motivations (except the last one) are also right for diaries and journals, so we don't particularly get an answer to "Why people blog." Indeed, why would one bother about publishing all this if one can simply keep it on the hard drive? The paper suggests that there is a great need for and awareness of the audience among bloggers. This is a point where it seems necessary to go further.

Why are there so many people who need to be heard? Are they lonely, misunderstood, underestimated? Do they want to prove something? Or may be the intention to keep a diary is amplified (modified) by the technology itself? How did they learn about blogs? Why and how did they make the decision to blog? Did they consider other possibilities? Did they keep a diary before?

Well, these and others are the questions for my imaginary study.

As for design issues, the paper recommends the following (among others):

1. Greater ease of use, editing and management.
2. Integration with online communication and document management tools.
3. Advanced search within and across blogs, blogs categorization and organization tools.
4. Quality assurance (subjective assesments by readers).