Lynn Cherny, (2005) Gakking Memes: LiveJournal �Conversation�
Comments by EugeneMedynskiy
I found interesting your discussion of memes on LiveJournal
. There's a paper from the HP Information Dynamics Lab that discusses the spread of memes in the blogosphere: http://www.hpl.hp.com/research/idl/papers/blogs/∞
. They identify common patterns of meme infection, and I wonder if these are the same on LiveJournal
. Especially since LiveJournal
users are likely to cite their source for a meme, this would probably be an easier analysis than what they dealt with.
Your mention of how 'substantive conversations' migrate and mutate is also thought-provoking. I wonder if it's possible to track topic spread via, e.g. word counts or something of the sort. This would probably be quite expensive to compute (since any friend or co-community member of a person who made a topic-based post could potentially pick it up), but there seem to be small-scale things one can do to get a flavor for this activity. One would be to track how cross-posted entries are appropriated by the different communities they are posted to. Another would be to track how active users who are members of a whole lot of communities polinate topics across communities.
Finally, just briefly, I wonder how your description of LiveJournal
communication as often being 'purely phatic' can be considered into its design. Xanga, for example, allows a person to give 'eProps' rather than shout-outs. I wonder what other non-textual signalling mechanisms could be or have been employed for this.
Comments by TomErickson
An interesting and thought provoking article. A few questions/comments...
I wasn't convinced, at least by the data you reported, that the the problem with the friends list is that it is "unfortunately named." It seems to me (without any data to cloud my vision ;-), that users of various social networking sites are pretty saavy about what being in a 'friends list' means. I'd guess that had the friends list been renamed the "people I regularly read" list, that most if not all of heartache would still swirl around inclusion/exclusion, and that the concerns about being "boring/offensive", and the tactic of including people you don't regularly read on your "people I regularly read" list would still be used. If using the F-word really is contributing, I'd expect to see discussions having to do more with personal relationships (like/dislike, angry, hurt, ignored) than with author-writer roles. Anyway, I'd be interested to hear about what leads you to believe that it is the use of the "friend" construct that is causing problems...
You quote someone as saying that "...commenting behavior provides a far better glimpse of social networking at work than the static friends list attribute." While I agree that this is interesting, it strikes me that the 'friends list' minus the 'set of those commented on' is interesting as well. That is, to the extent that this similar to the set of 'pity friends', *that* provides an interesting index of social ties...
I was interested in your discussion of the role of memes in conversation. For some time, I've been pondering the notion of there being mini-genres (or perhaps baktinian speech genre), where there is a very simple interactive structure (Q&A, limerick, top-10-reasons list) which an author can deploy to evoke interaction, and in which readers can participate -- successful mini-genres tend to have (a) a widely understood interactive structure (usually one that has a obvious termination point), (b) have length and/or other constraints that keep individual responses relatively light weight, and (c) within their constraints allow participants to display cleverness, perspicuity, etc. Another way of putting this is that they have a game-like character consisting of a combination of implicit rules that participants follow (or creatively stretch) and the ability of participants to perform at varying levels. ...It seems to me that memes, as you are talking about them in this context, have some of these characteristics (although I am a bit unclear about where you draw the line between a conversational meme and a conversational theme). ((The observation that LJ participants are tending to attribute memes, but not the more abstract conversational topics strikes me as quite interesting...))
Comments by CliffLampe
It's interesting how prevalent memes are in conversation forums of all stripe. Slashdot has had pervasive memes since it's early days, and Fark, Kuro5hin
and other online discussion spaces have similar phenomena. As Tom argues above, meme's might play a role in spurring online conversation, though rather than "mini-genres" I'd think of them as boundary objects. Being "in" on the meme helps you to establish a group identity and feel like a more central participant. People who aren't in on the meme are on the periphery of the group. To some extent, this is the extended version of how Netspeak and netiquette were first established.
Memes also tend to have a short shelf life. Using a meme that has been dropped by the community is a clear signal of peripheralness, and may hinder future interactions. It would be really interesting to talk more about the constructive and destructive traits of memes in online conversations.
So what is it about some topics that they become memes while others die on the vine? My guess is there is a "diffusion of innovation" story where participants central to the network (lots of friends, highly regarded, high reputation) pick up a meme and other emulate their use of it. I'm sure there's more work on how memes start and end in online groups, so any pointers would be great.
The observation in this paper that really caught my attention and continues to bother me is the "everyone is talking about X" phenomenon. I wonder about the use of "everyone", and I think it's being used several different ways, perhaps all at once. I realize this isn't even close to the main point of the paper, but the relationship between local observation/perception and inference about global behavior is maybe even more susceptible to distortion online than offline.
The first thing I think is happening is hyperbole; "everyone" is hyperbole for many people. Secondly, I think there is an implicit qualification in there, something to the effect of "everyone [who is important to me] is discussing X" or "everyone [relevant to this conversation] is discussing X". Most difficult to overcome, however, is the inability to distinguish local phenomena from global ones. A great example is that people who are doing about as well financially as their neighbors tend to think of themselves as "normal", especially with respect to the neighbors, the only group of people that matters to them.
So to bring this back to Lynn's paper, how are LJ users or other bloggers (or web users, etc.) to know that a meme that has grown popular in their social network, but that it's completely unknown elsewhere? How do people in a niche know they're in a niche, especially when that niche is so pervasive that it's hard to see outside of it? Part of the reason it is hard to see out of it is that evidence that confirms their belief (e.g. another person mentioning a meme) weighs much more heavily than evidence that disconfirms it (many people NOT mentioning that meme). I think sometimes as people who work on social interaction through computers we try very hard to bring people together around shared topics, and the side effect is that we make everyone feel like a king of his or her own conversational fiefdom. I'm not sure what the answer is, since we take it almost as a first principle that it is good to bring like people together. But there also needs to be some way to demonstrate how large or small topic-sharing clusters are, and to add a bit of global perspective.
Comments by JudithDonath
Really interesting paper. It is more an anthropological description of the LiveJournal
phenomenon than an analysis of it, though the discussion of the issue of �friends� and what it means to �defriend� someone, and the effects of having chosen that loaded word rather than some other for what could be called �feeds� is good.
I was particularly interested in the discussion of memes and how they move through the LiveJournal
environment. I think this is a fascinating way to look at the deeper social structure of the users, for it is not only about who connects to whom, but about the way status is embodied in an information society.
[Please insert additional comments in this section followed by your WikiName