Formation of Norms in a Blog Community

Carolyn Wei, University of Washington

Blogs are often situated within a blog community of similar interests. These communities can be a useful way for readers to access a specific slice of the estimated one million blogs published in 2003 (Technorati.com, as cited in Lindahl & Blount, 2003). Online blog communities can form around numerous themes such as , blogs about Java programming, or blogs by redheads. Each of these blog communities has its own practices and behaviors, some of which are shaped by explicit community guidelines. The Redhead Blogs community, for instance, specifies types of redheads allowed in the community and warns against offensive content (Redhead Blogs, 2004). Overt guidelines may shape some of these standards; however, the formation of norms within blog communities may also be influenced by the content or values of member sites. Because blog posts are often spontaneous and informally written, norms within a blog community may be formed “bottom-up” from actual practice.

This paper reports on a study of a community of blogs devoted to knitting and compares the community’s stated normative guidelines with actual practice. Although explicit rules may help to shape a fledgling blog community, the behaviors of individual bloggers may also help to create norms. Further, actual practice may also elaborate on stated guidelines in unexpected ways. The purpose of the study is to identify common themes, behaviors, and practices with respect to content and site design within this community of knitting bloggers and to examine the formation of these norms, particularly in relation to the community’s stated rules. This paper reports the results of the content analysis of a selection of posts from a sample of members of the web ring. Some specific norms of the web ring are discussed, and potential explanations for their formation are discussed.

Online communities have sprung up around other media such as newsgroups, digital games, or chat rooms and have developed their own norms, unique to the medium and the culture of each group. For example, the fans of the band Phish developed social and behavioral norms on the Phish.net Usenet group that revealed shared values, which are also reflected in the list’s FAQ (Watson, 1997). Distinctive behavioral norms have also emerged in environments as diverse as a Usenet group for soap opera fans (Baym, 1998), a lesbian café in a MOO (Correll, 1995), a role-playing MUD (Schaap, 2002) and the WELL, an electronic messaging system for topical discussion (Rheingold, 1994).

Blog communities may also develop practices and norms informed by the blog format. The blog format allows for highly individualistic expression. For example, Rebecca Blood, a veteran blogger and writer about blogging, noted that the introduction of Blogger, a popular tool that made it easy to write freely about anything, probably influenced the growth of “short-form journal” blogs filled with quick thoughts on the writer’s daily life (Blood, 2000). Further, in her Weblog Handbook, Blood identified conventions, genres, standards, and etiquette that are often observed in blogs, but notes that “[e]ach weblogger creates a personal version of the weblog format, dictated by purpose, interest, and whim” (Blood, 2002, p. 8). These individualistic blogs will often join in communities of similar views or interests linked to each other in their sidebar navigation menus (Blood, 2002). Empirical research has also identified common practices among blogs in general. Halavais (2002) found that blogs randomly sampled from blog directories such as weblogs.com and livejournal.com most commonly discuss top news stories. Herring et al. (2004) analyzed a random sample of blogs from blo.gs and described many aspects of blogs such as author characteristics, purpose of the blog, structural elements, and temporal features.

Unlike these previous studies which discuss the practices and attributes of blogs in general, this study examines the norms of a specific community of a knitting bloggers in order to see how the community’s culture may influence the formation of norms. The Knitting Bloggers NetRing is a web ring of over 300 blogs devoted to knitting. Potential members elect to join the web ring by submitting their site for approval by the ring administrator. Upon approval, they are given a snippet of code to place on their site, which will allow visitors to travel in a linear fashion to all the member sites by clicking “previous” or “next” links. In 1996 web rings were considered a possible alternative to search engines because sites could form topic-specific rings. But the perception of their utility shifted as early as in 1999, when web rings were seen as a way to “unite passionate seekers and purveyors” and seemed to “function primarily as forums for sharing poetry, personal journals, and broken hearts” (McDermott, 1999, p. 68). They are often used to link personal web sites together, and according to Dinty W. Moore, web rings are “very grass-roots” compared with the structure imposed by corporate portals like Yahoo! (McDermott, 1999, p. 70).

Membership in the Knitting Bloggers’ ring has been steadily growing since it was established in early 2002, particularly with targeted magazine articles about knitters’ blogs in Interweave Knits (Stoller, 2003) and Vogue Knitting International (Petrovski, 2003) and popular press articles about the resurgent interest in an old needlecraft (see Tomlinson, 2003, for one example). With a regular influx of new members who may also be new to blogging, the development of norms within this web ring may be an especially relevant issue for maintaining the qualities and sensibilities of the ring that helped make it so popular.

Like many other online communities, the Knitting Bloggers embody a distinctive culture, where members are often passionate about knitting and are excited about getting to know like-minded people either through personal interactions or through the words on a blog. Relationships between the knitting bloggers can move fluidly online and offline. Some bloggers regularly get together in real space to participate in “stitch ‘n’ bitches” (informal knitting and gossip parties) or go to yarn stores. Group activities may also be carried out online. One interesting tradition is the “knit-along” where a group of people knit from the same pattern, such as for slippers or a sweater, and blog their progress. Often, one of the knit-along bloggers will post pictures of the participants’ finished items as they complete the knit-along. The unique culture of knitters who blog likely contributes to some of the behaviors observed in member sites.

There are also clear membership guidelines for the ring displayed on the page that one uses to join (Membership Guidelines, 2003). The normative rules are summarized below.

• Must be written in English (“Only blogs written in english [sic] will be accepted.”).
• Ring code with “previous” and “next” links must be visible (“The ring code must be visible from the page where your blog is located, not on any other pages.”)
• Must post at least once a month and preferably once a week (“To stay a member of the ring you must post regularly. For our purposes regularly will be at least once a week. If you fail to post once within a month, you will be moved back to the queue....if you continue not to post, you will be completely deleted from the ring.”).
• Must actively post about knitting (“You must have a knitting blog, where you talk about knitting. Simply being a blogger who knits isn't enough, you must actively post about knitting.”).

The rules are fairly explicit about posting frequency and about display of the ring code, but there is a lot of room for interpretation regarding permitted content on sites. Knitting is practiced by people with a range of skills and backgrounds. As such, knitting blogs may reflect a diversity of content under the broad banner of “knitting.” Bloggers may write about progress on their latest sweater project, sales at local yarn stores, favorite knitting needles, or knitting-related web links. Actual practice on these websites may supplement the stated guidelines for the Knitting Bloggers. Other practices completely unpredicted by the explicit rules may also emerge.

Method

To develop an understanding of the norms that are present in the web ring, content analysis of some features of the blogs was performed. Content analysis was chosen for this study for its utility in summarizing and describing messages. It allows for the “objective, systematic, and quantitative description of the manifest content of communication (Berelson 1952, p. 18, as cited in Neuendorf, 2002, p. 10) and for measuring characteristics in mass media (Berger, 1991, p. 25, as cited in Neuendorf, 2002, p. 10). As such, content analysis seemed well-suited for establishing baseline characteristics of the Knitting Bloggers that could be compared with the membership guidelines.

Sampling

The sample frame for this study was the list of active member sites of the Knitting Bloggers NetRing. RingSurf, the host of the code that links the knitting blogs together, maintains a list of active members and public visitor statistics for each site (RingSurf, 2003).

A stratified random sample of blogs with varying experience with the web ring was drawn. To create this sample, the sample frame was sorted by number of visitors dispatched (the number of times someone leaves a site to visit another member by using the NetRing code). The dispatched statistic is cumulative (rather than reset regularly) and reflects two factors: the length of the site’s participation in the ring and the subgroup of visitors to the site who then click on the ring code. The sample frame was then divided into groups of ten. One blog was selected from each group (stratum) with Microsoft Excel XP’s random number generator. The sample included 33 blogs, ten percent of the members of the Knitting Bloggers NetRing.

Materials

The sample of 33 blogs was captured and archived with Offline Commander, a shareware program. The purpose of archiving the sites was to ensure that each coder was evaluating the same versions since blogs are frequently updated. The sites were then exported into HTML format for the coders to evaluate. A codebook and data collection worksheets were created for the use of the coders. The codebook described in detail the items that coders were to rate.

Procedures

Two coders separately assessed the sample of knitting blogs for the same variables. I was one of the coders, having been a member and observer of the web ring since 2002. The other coder was unfamiliar with the Knitting Bloggers. She did not knit but was familiar with knitting vocabulary and concepts because of previous experience with related needlecrafts. Having two coders helped to establish rating reliability. The purpose of using coders with different knitting knowledge levels and relationships with the Knitting Bloggers was to determine the effectiveness of coding blogs for norms without previous immersion in them.

The selection of variables for coding was informed by the explicit rules of the web ring as well as by my earlier observations of knitting blogs. Items included

• Presence of author’s name on the blog;
• Author’s gender;
• Blogging software;
• Web ring memberships;
• Placement and appearance of Knitting Bloggers ring code;
• Presence of knitting project lists;
• Frequency of posting;
• Number of photos; and
• Subject matter of entries.

Coders looked specifically at the homepage of each blog for the bulk of the variables. For the in-depth coding of subject matter, coders analyzed all the posts made in a specific one-month period between October and November 2003 for the entire sample of blogs. To code for subject matter, coders noted for each day whenever one or more of 13 possible topics appeared. Examples of topics include current personal knitting project, current knit-along project, yarn purchased, non-personal knitting news (such as about a new yarn store), pets, and family. Some of the topics were termed “news” because they reported an event or story, like one might share with a friend while gossiping about “what’s new” in each other’s lives. Each day might be coded for multiple topics. On a given day, a blogger may be coded as talking about a pet, her family, a knitting magazine, and a current knitting project. After coding, all the topics were tallied. A maximum score of 31 days could be earned for a given topic, assuming the blogger posted about that topic every day in the month of interest.

All raw data were entered into a Microsoft Excel XP file. Differences between the two coders were noted and resolved in one of three ways. Inter-coder differences in dichotomous variables (yes/no variables indicating the presence of a feature) were resolved by crediting the blog with that feature. For nominal variables such as the appearance of the NetRing link, differences were resolved by alternating between the two coders’ scores. Discrepancies in scalar variables such as the number of amateur knitting photos on the home page were resolved by using the mean of the two coders’ scores. The resolved data were then imported to SPSS 11.5 for statistical analysis.

Results and Discussion

To determine the reliability of the two coders’ work, percent agreement was calculated. The two coders agreed on 85.19% of the units, that is, across the 33 cases and 62 variables (33 X 62 = 2046 units). The rest of the statistics reported below are for data with inter-coder differences resolved.

Blog Demographics

The sampled blogs were a wide range of ages with first post dates ranging from June 27, 2001, to November 9, 2003. The median first post date was June 1, 2003, so most of the blogs were relatively new. In fact, about 79% of the blogs were first published in 2003. Many members of the Knitting Bloggers were therefore not only relatively new to the ring but were probably also new to blogging. They may still have been developing their own style and feeling their way through the norms and styles of the Knitting Bloggers as well as of blogging in general.

All of the sampled blog writers were female, except for one person whose gender was unknown. The authors’ genders were determined based on names, photographs, and textual references, such as to a husband. There are some male bloggers in the Knitting Bloggers’ web ring but none happened to have been sampled. The gender makeup of the Knitting Bloggers’ web ring probably reflects rather closely the genders of knitters in general, at least in North America. Knitting is stereotypically a female craft. However, one might expect a higher percentage of males in the Knitting Bloggers’ ring than in real life if only because male knitters are somewhat uncommon, and they may seek fellowship in an online venue. (There is another knitting blogger community for Men Who Knit.)

Three blog tools were commonly used by the sampled blogs. Blogger was the most popular: 45.5% of the blogs were published using Blogger, compared with 18.2% using TypePad and 18.2% using Movable Type. The popularity of the Blogger, TypePad, and Movable Type software and templates explains why some of the blogs within the web ring had a similar look and feel. As noted by Blood (2000), Blogger (and now TypePad) makes it easy even for lay bloggers to blog often and freely. Later this paper will report on the actual frequency of posting by the Knitting Bloggers.

Norm: Posting in English

In accordance with one of the membership rules of the Knitting Bloggers’ ring, all sampled blogs were written in English. Compliance with this rule may reflect strict enforcement by the ring administrator and natural attraction to this web ring by English speakers given that the ring is hosted in English. The existence of a common language in the Knitting Bloggers’ ring helps the community to thrive since members can easily read each other’s blogs and link to one another. The English rule was originally put in place to ensure that websites were not “kiddie porn” masquerading as non-English knitting blogs (Submit a Site, n.d.). Actual practice by member blogs makes this rule support reading and commenting within the community in addition to protecting the web ring from pornography.

Norm: Displaying Web Ring Code

To belong to the Knitting Bloggers, members must display a snippet of code that allows visitors to travel around the web ring by clicking “previous” and “next” links. In every sampled blog, the code was indeed visible. The intent of the normative rule of displaying the ring code was to make it easier for people to surf the ring by easily finding the previous/next links. In practice, there was much variation in where the code was displayed, making it potentially difficult for visitors to “hunt and peck” around sites to find the linking code. Almost half of the blogs displayed the ring code on the top left hand side, but a third displayed it in the top right hand side, and about 5% displayed it in the top center of the page.

The task of finding ring code on the blogs is further complicated by the different visual appearances of the link. About a third of the sites used only textual links to the ring, about 40% used one of two variations of a drawing of knitting hands (see Figure 1), and about 20% used a photograph of a knitted swatch (see Figure 2). There were also a few sites with broken image links. The different link styles reflect the lack of guidelines about the appearance of the images and suggest that individual choice in display may overrule any standard look for the ring. The membership rules say to display the code with previous/next links but they do not forbid the modification or personalization of the code.

Figure 1. One variation of the knitting hands image web ring link.

Figure 2. Knitted swatch image web ring link.

Norm: Posting Regularly

The Knitting Bloggers’ web ring requires members to post at least once a month and preferably once a week. In the 31-day period that was sampled in October and November 2003, the sampled blogs posted an average of 13.63 entries (SD = 9.59) over an average of 11.65 days (SD = 7.22). Collectively, the blogs were satisfying one of the guidelines of the web ring, to post once a week. However, there is much range in the frequency of posting. One blogger posted only one day out of the sampled period while the most prolific blogger posted on 25 days. The stated rules of the web ring are that blogs will put back in the queue (i.e., they need to gain approval from the administrator to become active ring members again) if authors do not post at least once within a month. However, this temporary deactivation is done manually by the ring administrator and it may be time-consuming for that person to monitor adherence to this rule on hundreds of blogs. The result may be that, in practice, members of the Knitting Bloggers are allowed to flexibly interpret the rule regarding frequency of posting. They may blog when they feel like it rather than because they are compelled by a rule.

Norm: Posting about Knitting

One of the explicit rules of the web ring is to actively post about knitting. The sampled posts were coded to see whether they touched on any of 13 topics, 7 related to knitting and 6 related to other crafts, family, work, pets, or other. Table 1 displays the average number of days that the coded topics were blogged about. The larger the mean, the more days that the bloggers on average wrote about the topic. Knitting topics are highlighted in blue.

Table 1. Average number of days where the following topics were blogged

Topic	Mean days (SD)
1. Current personal knitting project	6.24 (4.55)
2. Personal news not related to knitting	3.89 (3.71)
3. Other personal knitting news	1.92 (1.71)
4. Other news (not covered by other topics)	1.52 (2.09)
5. Family news	1.14 (1.93)
6. Pet news	1.12 (4.09)
7. Work or school news	.92 (1.39)
8. Other crafting news	.70 (1.32)
9. Knit magazine or patterns	.67 (1.27)
10. Current knit-along project	.61 (1.90)
11. Yarn	.56 (.77)
12. Non-personal knit news	.27 (.50)
13. Past knitting experience	.18 (.43)

Current personal knitting projects were by far the most common posting topic. Other personal knitting news was ranked third. After that, other knitting topics dramatically dropped off in frequency. Varieties of non-knitting news were quite common in the blogs. Personal news (not related to knitting) was the second most common posting topic. Ostensibly, the Knitting Bloggers’ web ring is for bloggers who “actively post about knitting.” However, news from other parts of people’s lives seem to be woven in. Given that many of these blogs appeared to be catch-all personal journals, as defined by Blood (2002), with personal, family, and pet news, these knitting blogs may very well be the writers’ only blog. That may be why so many of the bloggers posted about other things in addition to knitting. The blogs may represent more than just their knitting. Or knitting may reveal much more about a person than an ability to manipulate sticks and string.

Knitting Bloggers’ membership guidelines do not quantify “active posting,” but when combined with the guidelines on posting frequency, a reasonable estimate is once a month. According to the web ring’s rules, blogs that infrequently post about knitting are probably candidates for removal from the ring, yet they are allowed to remain. It may be that the Knitting Bloggers’ normative rules may be loosely enforced. Or possibly some knitters may go through “slow periods” where they do not knit and therefore may not be inclined to blog about knitting.

Norms: Common Knitting Blog Features

Besides norms about posting and ring code links, other common features were observed in the sampled knitting blogs. Over 90% of the blogs supported commenting, an impressive number given that commenting is not offered by the most popular blog software, Blogger. Blogger users must utilize a third-party tool or service to enable comments on their sites. The commenting helps foster community since fellow Knitting Bloggers as well as any other visitor can leave praise for a current knitting project, ask questions about knitting technique, or post any other kind of comment.

Other features found on the knitting blogs are not basic blog features in the same way that comments or archives have become. On the knitting blogs, these features seem more for amusement and to show belonging in the community, such as the Weather Pixie. These features are described in Table 2.

Table 2. Common blog features

Feature	Description of Feature	Percent of Blogs with Feature
Blog Buttons	Personalized buttons that other sites can download to use on their own sites to link to your blog.	36.4%
Weather Pixie	A cartoon whose clothing and background update to reflect local weather.	21.2%
Tag-Board	A light-weight discussion board.	18.2%
Percentage Bars	Flash plug-in that allows percentage updates, for example, to display progress knitting a hat.	18.2%
JavaScript Referrer System	Displays the referral sources to the site within the last 24 hours.	12.1%

Some features on the knitting blogs are unique to the knitting genre. One convention is to track progress on knitting projects or to show off completed works. Over 57% of the bloggers tracked their “Works in Progress” or the items that they are currently knitting. These were directly listed or were linked on the home page. Over half of the blogs also kept lists or photo galleries of their completed works. Another popular practice was to keep a separate list of links to favorite knitting blogs in a side bar, without commingling them with links to other kinds of sites: over 60% of sites linked to other knitting blogs in this manner, suggesting Knitting Bloggers spend time visiting other blogs in their community. These simple normative practices help knitting blogs develop a distinctive flavor, one that allows visitors to see that the blogs are knitting blogs even if the posts are not exclusively about knitting.

Features such as commenting and lists of links to favorite knitting blogs support the Knitting Bloggers’ community, perhaps demonstrating the “tightknit” nature of the web ring. These features allow knitting bloggers to develop stronger connections to one another through links and dialogue in comments. Interestingly, the Knitting Bloggers’ web ring may be becoming too diverse and new specialty knitting blog web rings are emerging, such as for Men Who Knit, knitters with cats (Knitting Kitty), and gay knitters (QueerKnit). About half of the sampled knitting blogs belonged to more than one web ring. Some of the popular rings were Fiber Arts Bloggers (for a variety of textile crafts) (27.3%) and QueerKnit (15.2%). Membership in multiple web rings leads to a question for future research: how do blogs adhere to norms and rules for multiple blog communities? Granted, there is much overlap of interest and probably practice between these knitting blog communities, but when differences exist between them, how does a blogger resolve them?

Coding Challenges

Coding web pages for content can be problematic because of broken links. In some cases, archived blog pages were no longer available so the posts on those pages could not be analyzed. Furthermore, it was discovered that prior immersion in Knitting Bloggers was extremely useful in certain situations. For example, it may be difficult to distinguish between a personal knitting project and a group knit-along project without prior knowledge. Much of the writing in a blog is in shorthand, and it can be difficult to detect the whole story behind the blog by looking only at a month-long block of posts. For example, a blog author might fully describe embarking on a knit-along glove project and refer to it subsequently without mentioning the knit-along. Without prior knowledge, a coder may not be able to tell the difference.

Conclusions

This study of the Knitting Bloggers’ web ring suggests that while rules may help guide the development of the blog community, actual practices within the blog also shape some of the norms of the community. Results suggest that while many blogs nominally adhere to the guidelines for the web ring, there may also be much variation in how the blogs behave, for example, with the different appearances of the web ring code and the range of topics that might be blogged about. The baseline measures generated in the content analysis also showed that the knitting blogs shared features that are not dictated by explicit guidelines, yet help to characterize a knitting blog, such as the “Works in Progress” lists. The Knitting Bloggers’ community suggests that community designers, such as the administrators of web rings, can create member guidelines, but actual practice by members will also shape these communities. As blog communities emerge, understanding how norms form may help community designers create an environment that satisfies members’ values and cultures.

References

Baym, N. K. (1998). The emergence of on-line community. In S. G. Jones (Ed.), CyberSociety 2.0: Revisiting Computer-Mediated Communication and Community (pp. 35-68). Thousand Oaks, CA: Sage Publications.

Berelson, B. (1952). Content analysis in communication research. New York: Hafner.

Berger, A. A. (1991). Media research techniques. Newbury Park, CA: Sage.

Blood, R. (2000, September 7). Weblogs: A history and perspective. Retrieved March 4, 2004, from http://www.rebeccablood.net/essays/weblog_history.html.

Blood, R. (2002). The weblog handbook: Practical advice on creating and maintaining your blog. Cambridge, MA: Perseus Pub.

Correll, S. (1995). The ethnography of an electronic bar: The Lesbian Cafe. Journal of Contemporary Ethnography, 24 (3), 270-298.

Halavais, A. (2002). Blogs and the "Social Weather". In Internet research 3.0. Retrieved March 1, 2004, from http://www.aoir.org/members/papers3/halavais-ir3b.pdf.

Herring, S. C., Scheidt, L. A., Bonus, S., & Wright, E. (2004). Bridging the gap: A genre analysis of weblogs. In Proceedings of the 37th Annual Hawaii International Conference on System Sciences (pp. 101-111). Retrieved March 7, 2004, from IEEE Xplore.

Lindahl, C., & Blount, E. (2003). Weblogs: Simplifying web publishing. Computer, 36 (11), 114-16. Retrieved November 26, 2003, from IEEE Xplore.

McDermott, I. E. (1999). Running rings around the web. Searcher, 7 (4), 67-70. Retrieved March 7, 2004, from ProQuest Research Library.

Membership Guidelines for the Knitting Bloggers NetRing. (2003). Retrieved November 26, 2003, from http://www.ringsurf.com/netring?action=addform&id=&ring=knittingblogs.

Petrovski, L. (2003, Holiday). Blog on, needleworkers! Vogue Knitting International, 13.

Redhead Blogs... A webring for redheads! (2004). Retrieved March 7, 2004, from http://www.2girlsdesign.com/redheads/

Rheingold, H. (1994). The virtual community. New York: HarperPerennial.

RingSurf. (2003, November 26). Knitting bloggers stats. Retrieved November 26, 2003, from http://www.ringsurf.com/netring?ring=knittingblogs;action=stats;id=.

Schaap, F. (2002). The words that took us there: Ethnography in a virtual reality. Amsterdam: Aksant Academic Publishers.

Stoller, D. (2003, Spring). Around the world in 80 clicks: Weblogs -- a global knitting show & tell. Interweave Knits, 8 (1), 6.

Submit a site to Knitting Bloggers Net Ring. (n.d.). Retrieved March 7, 2004, http://yivich.tripod.com/ring.htm.

Tomlinson, S. (2003, November 13). Knitting factories: “Stitch 'n' Bitch” sessions create tightknit groups. Boston Globe, p. 10. Retrieved November 26, 2003, from ProQuest National Newspapers.

Watson, N. (1997). Why we argue about virtual community: A case study of the Phish.net fan community. In S. G. Jones (Ed.), Virtual Culture: Identity and Communication in Cybersociety (pp. 102-132). London: Sage.

TrackBack