Ethnographic Blogging: Reflections on a Methodological Experiment

This paper describes how a weblog was utilized as a major component in a long-term, multi-site ethnography with both “virtual” and physically situated research components. “Ethnographic blogging” describes not only the act of writing on a website and hoping that someone will read it, but the process of regularly maintaining a blog, and the modes of interaction and observation that this process gradually enables. In my own study of self-identified ‘geeks’ and ‘nerds,’ ethnographic blogging involved traversing news sites, forums, and other blogs for relevant content, leading to opportunities for dialogue with other bloggers and readers; establishing a persona online as a researcher, which has encouraged subjects to invite me to public and private discussions about their culture and identities; and bringing together online subjects from multiple physical sites, among other opportunities. My own experience of integrating a blog into ethnographic research was largely experimental, though I offer these reflections to encourage researchers to consider what alternative means of qualitative analysis online may have to offer us.

One afternoon in early 2008, I received an email with a hyperlink from Church, a regular visitor and commenter on my academic blog. The link led to an ongoing conversation begun in the comments at Hipster, Please! (hipsterplease.com), a geek music and culture blog, and continued in a forum at Game Music 4 All (gamemusic4all.com), a site dedicated to covers and remixes of video game soundtracks. I was doing a study on how terms like 'geek' and 'nerd' had been appropriated by fans, techies, and misfits as proud markers of identity, rather than insults; Church often sent me links that he thought I might find interesting. In fact, Church did this for many other bloggers as well. In the course of browsing other websites for my research, I often saw bloggers write, "Thanks, Church!" with a link back to his YouTube page. I'd seen him credited by Time writer Lev Grossman, the bloggers at TechDirt (techdirt.com), and Z., the blogger behind Hipster, Please! Just a few days before I got this particular email from Church, in fact, I'd noticed that he tipped off Boing Boing (boingboing.net) -one of the most trafficked blogs on the web -on Z.'s new tattoo, a picture of a 20-sided die of the kind traditionally used for Dungeons & Dragons. Nevertheless, on this particular afternoon, the link was a bit of a surprise for me. This was the event that led me to realize for certain that my blog was not simply a public journal, but a central component to my ethnographic study.
Normally, I assumed, Church read all kinds of geeky things in the course of his day anyway, and, being a helpful sort, he just sent along a link when he thought someone might find it interesting. This link was different, however; it was meant specifically for me. As I skimmed a lively exchange about why "nerdcore hip-hop" may seem less community-oriented than Harry Potter-themed "Wizard rock," I saw a note by Matt, another regular reader and commenter on my own blog: That's a whole other conversation entirely. May be Church needs to pull Jason from Geek Studies into this discussion and see what he found in his research. (matts 2008) I was mildly shocked. They were talking about me. In part, my shock was due to the fact that I had been thinking of this discussion as the display of their expertise as members of certain geek subcultures, upon which I would draw as a researcher. It's true that I was well into my ethnography of geek identity, but I was still getting a handle on the dynamics of nerdy music scenes.
This, however, was only part of the shock. Another part was simply that I realized that I had actually inserted myself into a real (albeit "virtual") community, and I hadn't even realized it entirely before this point. I was invited to this conversation not because I had published extensive research on nerdy music scenes (I hadn't), but because I had been writing regularly and informally at my own website, Geek Studies (geekstudies.org), for nearly a year before this conversation took place.
It's very easy for the virtual ethnographer to feel like a "lurker," even if she has announced her presence to a particular online community. Likewise, it's quite easy for a blogger to feel like the process of writing online amounts to little more than putting messages into bottles and casting them into the ocean. I knew I had readers, as they left comments on my posts. What I had not reflected on very much before was how my readers traveled elsewhere on the web, and sometimes took me with them, welcoming me to follow; how my blog wasn't just a place to air thoughts about my research, but a headquarters from which to launch other journeys into my virtual "field"; how my blog was not an island that occasionally traded links with the other island-blogs on the horizon, but part of an ongoing discussion of geek culture that could be much more personal and wide-ranging than I had allowed it to be.
I'm sure it occurred to me before this that my website could work for me as a research tool, but this event crystallized for me the sense that blogging could even be a key component to virtual ethnography. In this paper, I will share some reflections on how blogging helped me to embed myself into communities relevant to my multi-site ethnography, and I will advance an argument for why maintaining the blog itself can be thought of as a process for online research that can achieve a broader picture of community than a more arbitrarily bounded virtual study.
Locating "Culture" in the Digital Age While I did not always expect my blog to be central to my online research, I did plan for my ethnography to have a substantial "virtual" component to supplement participant observation research conducted at physical sites. This multi-site approach was an attempt to deal with the elusive nature of my object of study, self-identified 'geeks' -a cultural group that has not been widely studied, occupies no single physical location, and remains difficult to clearly define at all (Konzack 2006). Concerns about hard-to-pin-down populations aren't new to my study, of course. The classic assumption that ethnography ought to be practiced in a single location over an extended period of time is well suited to populations that actually occupy the same physical space for extended periods of time, but not all populations and phenomena can be described as such (Hannerz 2003;Hine 2003;Marcus 1995). What about studies of foreign news correspondents on the go from place to place (Hannerz 2003), movie fans who may never meet face to face (Hills 2002), or news and support sites for people following specific current events (Hine 2003)?
As Ulf Hannerz points out, "sites of modernity" can present particular problems for the traditional ethnographer: "What do you do when 'your people' spend hours alone at a desk, perhaps concentrating on a computer screen?" (Hannerz, 2003, p. 211). Hanners thus suggests a multi-site approach suitable for approaching "some problem, some formulation of a topic, which is significantly translocal, not to be confined within some single place" (p. 206). Along these lines, George Marcus identifies an alternative mode of investigation, which "moves out from the single sites and local situations of conventional ethnographic research designs to examine the circulation of cultural meanings, objects, and identities in diffuse time-space" (Marcus 1995, p. 96).
This is not to say that traditional participant observation has no place in such ethnographies, but simply that it is sometimes necessary to question certain assumptions about effective research practice and integrate additional modes of experience into the ethnographic project. If we can understand culture as constructed through symbolic gestures, forming "webs of significance" we spin ourselves, as Clifford Geertz (1973, p. 5) suggests, we must consider other sites of symbolic expression besides those best witnessed in person. Or, as Christine Hine argues in Virtual Ethnography, "If culture and community are not self-evidently located in place, then neither is ethnography" (Hine 2003, p. 64).
My plan, then, was to get a broader view of geekdom, to "follow the metaphor," rather than select an individual group or place of focus (Marcus 1995, p. 108). The sole defining characteristic of the study was to be the use of the terminology in question itself -'geek,' 'nerd,' and the increasingly common implication I kept hearing and reading that these are somehow bonded through a 'geek culture' or 'geek community.' Some locations did come to mind for particular place-based ethnographic opportunities, such as Silicon Valley workplaces, comic book conventions, or certain notoriously geeky computer science departments (Margolis & Fisher 2002;Turkle 1984), but keeping to a narrowly placebased approach presented some practical and theoretical concerns. Would the geeks in these specific sites shed light on the phenomenon of geekdom more broadly? If I instead chose more temporary sites to get at a broader sample of self-identified geeks, such as fan and tech conventions, would a weekend be enough face time? And would the geeks who surround themselves all the time with other geeks -or the geeks who go to special fan and tech events just once or twice a year -actually be representative of self-identified geeks more generally?
Looking online, then, allowed me to tap into additional, broader bases of geeks and nerds. In many cases, the geeks attending the gaming and tech conventions represent only a fraction of such events' total target audiences, with much of the rest actually observable online. For instance, Penny Arcade (penny-arcade.com), a webcomic widely read in geek circles, has about 3.5 million readers (Geddes 2010); only about 0.5% of those (19,000) attended the Penny Arcade Expo, a major event for video gamers and Penny Arcade readers, in the largest year that I attended (Krahulik 2006).
Following from these considerations, I thus chose to pursue a long-term, multi-method, and multi-site project. I conducted participant observation and interviews at some of those physical locations that concerned me, such as around MIT and at fan and tech conventions, but I also spent a great deal of time interacting with individuals through mediated communication. Lacking a large and truly permanent gathering space, and frequently possessing a degree of computer aptitude, the material culture we commonly associate with geeks can largely be found on blogs discussing tech policy, webcomics about gamers, YouTube videos of nerdcore performers, and so on. Virtual observation and interaction thus allowed me to analyze geeks in a more day-to-day context than conventions, and to approach interviews more on their terms rather than my own. While I did take many opportunities to observe geeks in the flesh, I don't think it was unnatural to leave many other interactions entirely online. As Hine points out, "Many inhabitants of cyberspace […] have never met face-to-face and have no intention of doing so" (Hine 2003, p. 48).
Gusterson describes this sort of physically dispersed approach as "polymorphous engagement": "interacting with informants across a number of dispersed sites, not just in local communities, and sometimes in virtual form," as well as "collecting data eclectically from a disparate array of sources in many different ways" (Gusterson 2008, p. 116). As Gusterson points out, this approach can sometimes be more effective at getting direct access to subjects who belong to professional and cultural elites. Geeks, potentially constituting technological elite, may have actually provided both more manageable and richer interaction by communicating through mediated channels other than face-to-face interaction. This also seemed like an appropriate approach for subjects who were rather likely to be shy or more comfortable through certain mediated channels of communication than others.
I suggest the term "ethnographic blogging" and not "ethnographic blog" to describe my approach because it was not the contents of the website itself that made the ethnography, but the process of maintaining the blog and the relationships (with other bloggers and readers) that resulted. Ethnographic blogging, as I will describe it here, exemplifies the aforementioned approach of polymorphous engagement. It is not simply an attempt to replace the single physical site with the single online site. Rather, it involves multiple modes of communication over time. It potentially includes not just writing on the blog itself, but the construction of an entire research network through posts on one's own blog, comments on others' blogs, emails with informants, and referrals to "real-world" (i.e., offline) sites to follow up on in person.

Participant Observation as "Surf Sampling"
Many other ethnographies with an online component have focused on particular communities that offer more obvious parallels to physical space, drawing on metaphors of spatial embodiment: chat rooms, multi-user dungeons, virtual worlds, and so on (Hine 2003;Kendall 2000;Turkle 1995). The results of such work often remain quite relevant even as technologies change; Lori Kendall's Hanging Out in the Virtual Pub (Kendall 2000), for instance, presents results that are no less observable today but for being conducted in the text-based "multi-user domains" of yesteryear. My ethnography of geek cultures, however, had a broader scope in mind, attempting to draw from many perspectives in exploring the widespread cultural shift behind the meaning of 'geek' and 'nerd.' I wanted make use not just of relatively isolated pockets of highly self-selected Internet users -mailing lists with users who voluntarily sign up, massively multiplayer games charging subscription fees, obscure chat rooms and venues requiring semi-expert knowledge to access -but also the most openly accessible and easily stumbled upon areas, scattered across the entire Web. This, of course, can be potentially overwhelming, as Matthew Hills warns: Confronting the mass of data which is available online, it becomes immediately clear that no a priori meaningful or internally coherent corpus can be identified: one can only extract artificially bounded sets of information (such as my own focus here on a temporally fixed corpus of postings) which even then may remain virtually unmanageable in terms of the sheer weight of communications traffic.
Posing seemingly intractable difficulties of selection and generisability, Internet research may presage an academic crisis in confidence, provoked by the very "massification" of Internet discussion and interpretation. (Hills 2002, p. 174) Hills's methodology, then, illustrates one way to approach to the embarrassment of riches offered by online data: to bound one's scope within a particular set of information from which to draw data. This is not necessarily only an issue related to research online, however. As indicated before, any wide-ranging multi-site ethnography committed to "following the metaphor" faces the challenge of defining where to locate the object of interest, and when to stop looking. Having already refused to declare a bounded physical site for my study, I similarly approached my online research without clear boundaries, but with a few simple starting places.
To provide an origin point for my virtual wanderings, I selected specific websites to follow regularly during the period of my field research. These included sites such as Boing Boing (boingboing.net, over 2 million readers; Schofield 2008), Slashdot (slashdot.org, "hundreds of thousands of readers"; About Slashdot 2010), Gizmodo (gizmodo.com, 5 million readers; Kim 2008), and Penny Arcade (penny-arcade.com, 3.5 million readers; Geddes 2010). I picked these because they were among the most popular websites that overtly identified themselves and their audiences as geeky, and because they represented a range of stereotypically/traditionally geeky interests (comics, computing, games, gadgets, etc.). I chose these suspecting that some might be more fruitful than others, and knowing that any might point me to other sites worth following longer-term. And indeed, I discovered a number of other websites that I had not known about before. I added some of these to my regular roster of sites to check as appropriate, like Hipster, Please! (hipsterplease.com) and Time's Nerd World (now re-branded as Technland, techland.time.com). Others, I simply bookmarked for later retrieval, like the occasional single post about geek identity on someone's personal blog.
Rather than limit myself to focusing deeply and exclusively on an arbitrarily limited subset of Internet locales, I attempted to experience the online materials that geeks might encounter in a way more or less like how they would traverse them: that is, by straying from my origin point, following promising hyperlinks, scanning thousands of headlines, reading some pages in depth, and taking opportunities to contribute to discussions myself by leaving comments or writing posts with links back from my own blog. The resulting methodological approach is a snowball sample of websites and events located through a kind of cyberspatial flaneurie, aptly described by a term coined by my colleague Paul Falzone (personal communication, 2008): "surf sampling." It may seem directionless and potentially overwhelming, though it represents a means of engaging with sites and individuals from a media user's perspective rather that a researcher's systematically limiting gaze. After all, common features of many blogs imply that this seemingly aimless wandering is a typical mode of traversing the web.
As an example of how the format of blogs already encourages this kind of behavior, consider the "link post," a feature of many blogs' regular updates. Link posts collect hyperlinks to various sources such as news sites, YouTube videos, comic strips, free music downloads, and funny or insightful pieces on other websites. Many highly-trafficked blogs, such as Metafilter (metafilter.com), are practically little more than collections of links to other blogs that audiences may find interesting (and a forum to discuss those links), while others regularly blend linking with commentary and additional content, like Boing Boing (boingboing.net). For instance, Z., the blogger at Hipster, Please! (hipsterplease.com), commonly puts up posts with "Nerd News in Brief," alerting readers to geeky musicians' upcoming tour plans, new track or album releases, music videos hosted online, interesting stories sent along by readers, and so on. In the course of my research, such link posts frequently led me from one site I knew well to other sites I had never heard of at all, and which I might never have returned to again if they seemed to post very infrequently or appeared otherwise unlikely to offer many observable expressions of geek culture and identity. The process was not unlike the anthropologist's task of navigating a community of individuals in search of useful conversation and observation.
Moreover, one of the major uses of social networking software such as Facebook and Twitter is to share links with other people, with Facebook even surpassing email in frequency of use for this function (Carson & Angelova 2009). And so I turned to other media over time to direct my search for geek culture, receiving links from friends and informants through email and Twitter. Sometimes referrals from the Web led me to realworld events, as well, such as when I found out about a local "hardware hackers" gathering from a blog I started reading regularly partway through my research process.
In practice, this kind of sampling process essentially assumes occasional serendipity, much like aimless Web surfing for personal entertainment. Consider, for instance, a simple chain of referrals: I visited The Morning News (themorningnews.org) every morning for interesting links and current events. I learned of the site from someone I met at a comic book creators' event. While The Morning News sometimes features links to stories so esoteric they might be considered geeky (e.g., humorous information graphics), it is not branded as a "geek" site per se. On one morning, however, The Morning News offered a link to an article on "The Scientific Laws of Romance" at 10 Zen Monkeys (10zenmonkeys.com) -a title that seemed potentially relevant to my research, offered by a site I had never seen before. I clicked the link, skimmed the article, and quickly decided that it didn't seem worth bookmarking for my research. (I did sometimes find myself thinking unexpectedly about sites I had dismissed days earlier, leading me to dig hopefully through my browser history, but that is a risk of approaching Web research as Web surfing.) At the end of the "Scientific Laws" post, however, I saw a link to an article that did sound interesting based on its title. Under a "See Also" heading -offering links to other stories of similar content on 10 Zen Monkeys -was the link: "Girls are Geeks, Too" (Sirius 2007). I clicked on this and found a discussion of the roots of nerdity between a sociologist and the writers of She's Such a Geek! (Newitz & Anders 2006), a collection of essays about women's perspectives on geekdom. One of the editors of She's Such a Geek! noted, "We originally wanted to call it 'Female Nerds' and people complained. They felt like nerd was too negative, and that geek had been re-claimed as a badge of pride -kind of like 'queer'" (Sirius 2007). I copied this segment and added a reference to it to a post on my own blog, "Geeks vs. Nerds" (Tocci 2007), where I had been building an ongoing collection of examples showing how people distinguished between these terms. This is a fairly typical example of my online research process: wander online, follow links, hope to find something useful, and produce content in response. Much of the content I produced was in the form of comments on others' blogs, but as this example indicates, posting things to my own blog became a key component to my research process.

Establishing a Persona and Attracting Informants
Though I selected specific blogs to act as my "starting points on the Internet," the most relevant location for me turned out to be my own blog. Gradually, I came to see my own site as a way to "ground" my studies in a virtual place, if not in a physical one. Through my blog, I came into regular contact with a number of active and self-identified geeks and nerds online, contributed to various virtual communities, and attracted feedback from strangers which sees frequent citation throughout my ethnography. I have developed relationships with informants, and have come to regard some as friends, even as I have yet to ever meet most of them in person. In a sense, I did operate within a community of geeks in conducting my research, specific people I connected with regularly, though it was beyond anything I would have planned.
The sort of integration into online communities that I eventually achieved with Geek Studies was not something that happened overnight. This resulted from a long-term commitment to maintaining online interactions. Recall that I hadn't really started blogging with this expectation in mind, but it's rather clear in retrospect that this is how I found myself with informants I never asked for. In the early months of writing on Geek Studies, the site served mostly as a personal tool -a way to keep track of websites worth citing later, and to get my thoughts down in writing. More obviously social functions emerged very gradually, such as getting occasional feedback on my thoughts from friends and past interviewees. It was still nearly a year before I had established enough of an online "presence" in geek communities online to be invited to such conversations as the one described at the start of this article, and longer still before I got into even deeper interactions. Eventually, Matt, Church, and Z. let me in on their daily emails for link sharing, friendly banter, and long-winded intellectual conversations (which occasionally led Matt to curse me for giving him too many interesting things to think about when he was supposed to be getting real work done). While I never started Geek Studies with the intention to muscle my way into anybody's personal emails, encountering "natives" willing to serve as informants, seemingly "deus ex machina, like a fairy godmother," is precisely the sort of happy turn of fortune that ethnographers rely upon (Rock 2001, p. 34).
It's worth reflecting on why my informants sought me out and let me into their group, without ever even meeting me face to face. The easiest and most honest answer, I think, is that they agreed with the version (or vision) of geek culture I described, and were interested in seeing me succeed. That's not to say that we always see eye to eye; Matt, for instance, once enshrined a days-long debate between our two blogs (on the risks/benefits of broadening the market appeal of stereotypically geeky media) as "The Geek Culture Debate" (Sweeney 2008). But our disagreements have been genial, the links they have sent me have portrayed geeks both positively and negatively, and, of course, they were not the sole source of data for my ethnography so much as helpful guides who pointed me in new directions to explore on my own.
Establishing a research persona online and inserting oneself into ongoing discourses is not just a matter of writing posts on a blog, though of course that is a major part of it. In my case, it probably helped that I was capable of communicating in the terms of the populations I studied. To those concerned about the relative depth of a virtual ethnography compared to a long-term, place-based ethnography, it's worth noting that online research does require some understanding from the ethnographer of local cultural codes and contexts to process (mediated) interpersonal interactions. Despite the absence of certain verbal and physical cues in online conversation, alternative methods of communicating tone have made their way into instant messaging, email, and text messaging. Notably, simple emoticons -such as the Western :-) and the Asian ^_^ -have become practically commonplace in text-based communication online. Geeks have also developed a variety of online text-based slang terms and syntaxes, which include the use of numbers for letters (commonly known as "leetspeak" or "l33t"); deliberate misspellings (e.g., "pwn" for "own," a verb used by hackers and gamers to indicate dominance); HTMLstyle bracketed comments to indicate the speaker's intent (e.g., "<sarcasm>Oh, I think that's a great idea</sarcasm>"); and, borrowing from programming languages or text-based gaming, a slash preceding parenthetical comments or the description of physical actions speakers might make or pretend to make if they were present (e.g., "/waves at visitor").
Searching on Google will clear up a number of points of confusion, but some sites are so steeped in jargon and in-jokes that the tourist may need a second window open as a glossary. Fortunately, this is where my "native" background as something of a computer geek (and my typing skills) served me well.
All of that said, I now understand writing my own posts to have been just the first step of ethnographic blogging. Writing things that others wanted to read led some of those people to link to my site from their own sites, which in turn allowed some of those sites' readers to visit through that link. Others found my site from search engines, like Google. A very small portion of these visitors left comments on my blog, or emailed me personally. As a lone blogger, there's not a whole lot one can do to "optimize" your site so that it will show up on search engines -being linked to by other well-trafficked sites is the surest method of this -but there are things a blogger can do to increase the likelihood of receiving comments and being quoted elsewhere. I was hoping to encourage such responses myself, so Geek Studies allows for wholly anonymous comments (save for IP address, I believe, though I honestly don't know what I'd be able do them with even if I were to look at them). I also specify that all of my posts are available under a Creative Commons license, to encourage distribution and quoting for non-commercial purposes.
While it's nice to get the immediate feedback of comments, posts by other that linked to my site were even more useful to me, as they introduced me to those bloggers' audiences and encouraged longer-form discussion. This is how I "met" my aforementioned informants, Z., Matt, and Church. On Hipster, Please!, Z. included a link in "Nerd News in Brief" to a post I had written on Geek Studies, calling my site his "new favorite blog." I found out about it because I use Google Analytics (analytics.google.com) to see how users are finding my site, and noticed that some visitors had arrived at Geek Studies from Z.'s link. I began following Hipster, Please! regularly, commenting occasionally, and listening whenever I had the chance to Z.'s eclectic geek music podcast, "Radio Free Hipster." Eventually I had full discussions with other blogs, going back and forth post-by-post, including with Z.'s friend Matt. And, one day, someone just decided to invite me to join their regular email exchanges with funny links the occasional mp3 music file. Now we all follow one another on Twitter as well.
Following the linkages between one's posts not only helps a blogger to become "established," but also helps to show how posts can have a long-term life of their own, potentially leading the cyberspatial flaneur in new directions. I made some open-ended musings about geeks and sexism over a few posts, for instance, and people still return to comment years afterward. Seeing comments on other sites linking to my post, such as on Metafilter (metafilter.com; see velvet winter 2009), means that my original post worked as something like a catalyst for an asynchronous focus group of sorts -an unfocus group, perhaps? -scattered over multiple sites and months of discourse. Even if no one were to come and comment on my original post, I can still see others discussing the same topic on another site that links to mine. Plus, that Metafilter post linked to two other sites discussing geeks, feminism, and misogyny, leading me to even more reading material. Suddenly I found myself getting additional traffic from a Geek Feminism Wiki (geekfeminism.wikia.com, which I'd previously had no idea existed) because someone saw my post linked on Metafilter, leading me to even more information.
My blog also represents an odd nexus linking the dispersed sites of my various methods. I managed to reconnect with a former interviewee via email, for instance, when he saw that we were both commenting on another blog of mutual interest, and he followed my name back to Geek Studies. On another occasion, I wondered aloud in a post whether the "coolest" person from one geeky event I attended could ever be found in the same room with the "nerdiest" person from another event I attended on the same day. I don't know about the same room, but one person from each ended up commenting on Geek Studies within a few days. These examples offer a sense of how a blog isn't just a website, but a way of maintaining a consistent research persona that carries between different sites onand offline.
There are, of course, potential disadvantages to relying on the vagaries Internet traffic to define an online population. Admittedly, people commenting on and linking to Geek Studies don't necessarily offer a representative "sample" of nerds on the Internet -the ones who speak up are self-selected, just like anyone else who peeks out from behind the veil of anonymity online. The ones who link to my posts may be the ones most inclined to agree with me. Even so, I do occasionally find Geek Studies entangled in networks of linkage with other sites that don't necessarily reach the same conclusions, and I have had more than a couple commenters post to disagree with my opinions and dispute my observations.
I do at least try to keep track of traffic and incoming links using Google Analytics (analytics.google.com) to get a sense of the range of people who have stumbled upon the site through actively searching, by chance, or by accident. The resulting group is a mix of media fans, tech nerds, feminists, and seemingly accidental passers-by. In June of 2009, for instance, Geek Studies had 2,851 visits by 2,491 "absolute unique visitors" (i.e., different IP addresses from which visitors arrived). Of these, about 20% looked at more than one page on the site, and 86% were first-time visits (so, about 400 repeat or regular readers). That was a busy month, considering I hadn't been doing much blogging -most months hover around 1,200 unique visitors, with occasional spikes when Geek Studies receives a link from an influential blog. In June, much of that traffic came from the aforementioned Metafilter link.
By the same token, I notice that a hefty portion of Geek Studies traffic comes from people who are simply unskilled at finding pornographic websites. I once went through the keywords that visitors had used to find Geek Studies through search engines, tallying up how many were looking for "nerd girl sex," "young girls sex," "girls on girls sex," or other terms that seemed not specifically related to the content I write about. I estimated that such traffic may have accounted for as much as 40% of Geek Studies' lifetime traffic. I was surprised to see that not all of these visitors leave the site immediately. In June, for instance, 10 people came in search specifically for "nerdy sex"; most left immediately, but one read multiple pages, resulting in an average of about five and a half minutes spent on the site between them. ( It is not a large audience by professional blogging standards, but neither is it a small group from the perspective of a researcher looking to establish an "online persona." The base of commenters and regular readers of Geek Studies is indeed "self-selected," then, but the general openness of websites, fickleness of search engines, and dispersion of commentary between linked sites have offered me a useful way of traversing the web by using my own blog as a sort of "headquarters" for my official persona on the net. Sometimes, the population I studied has come to me: visiting the site, they comment or pass through silently, though still leaving behind some tidbit about where they came from and what they were looking for. Other times, I might not have gotten visitors directly, but I have discovered new sites and discourses, whether by chains of links or by those informants who know where to find me thanks to the blog that introduced me to them.

Conclusion
I hope that this paper continues to challenge the lingering notion that "interactions" not situated in physical space offer no useful glimpse into the lived experience of a culture.
Granted, even I originally understood the online component of my research to be a supplement to somewhat more traditional participant observation study conducted in physical locations, including conferences and conventions that were openly marketing themselves as "for geeks" or "nerdy." Ultimately, however, the online portion of my research was no less important than the research conducted in person; half or more of the quotes, anecdotes, and examples used throughout the completed ethnography were from online sources or from contacts I would not have made if not for my online presence. I have envisioned my own blog not so much as the end product of the ethnography itself (as the ethnographic film is), but as an ongoing product, and blogging itself as a process that has expanded the scope of my ethnographic method. The blog is a tool that has opened up new directions for my research that a more traditionally bounded participant observation study -or even a more bounded virtual ethnography -might not have allowed. I could imagine an ethnographic blog that takes an ethnographic film-like approach, but I leave the theorizing (or utility) of such a project for other researchers to consider.
As for my own experience, I struggle to describe ethnographic blogging in terms analogous to existing ethnographic concepts: Is it like the field notes, but made visible to subjects, peers, and strangers alike? Is it better described as an extension of my own persona as a researcher, a mediated proxy of the researcher him or herself? Or is it simply a networking tool, encouraging willing subjects (and those who could point me to subjects) to contact me? It has served all of these purposes for me, and more. I offer this paper now because blogging enabled such novel and surprising modes of interaction in my own ethnography that I suspect it may be best understood as part of a new way of thinking about multi-sited research altogether.