Christina's LIS Rant

What happens when you cross the streams?
Or, norms in online communities, how journal commenting is different, and waving the flag on potential issues when aggregating web comments with journal articles.

There have been a couple of interesting discussions on friendfeed recently about commenting on journal sites vs. commenting elsewhere and about commenting anonymously, with an established online persona, or with the name on your drivers license. (I'm intentionally not linking to them, because the threads ended up in some less than entirely pleasant squabbling over a misunderstanding). I'd like to look at this for a minute.

As groups use communication technologies over time, they develop group norms.  These norms might include when and what to link to, how rowdy or polite to be, and what topics are appropriate.  Some discussion forums are supportive and helpful and warm, comfortable places to be whereas others are full of insults and ribbing and out and out flame wars - that's the norm, though, so people go there for that.  Newbies generally lurk (hang out without posting) for some period of time before commenting on posts, and then initiating threads. Lave and Wenger call this legitimate peripheral participation.  People get drawn into wikis this way (first reading, then correcting errors, then starting pages or adding new information) and blogs.  So usually, except for the first few people who have to sort of blaze a trail, people learn how to interact in a new communication forum by watching and then dipping their toes into the water.  And the first few people may be using the tool in a completely unexpected way (or at least many people aren't reporting what they're doing right now on twitter). 

You might say, that the thing with blogs, though, is that it's sort of your own little home. You can do what you want there, rearrange the furniture, and moderate the comments. You set the tone with what you post, how you say it, and what you link to.  But still, you probably act like others whose blog you read act, and you link to them and they link to you, so there is a little community and there are norms.

People also develop online personas that may or may not be anything they are like in meat space.  They might play devil's advocate or be a curmudgeon. They give this persona a name and that's how the person is known.

Ok, so fine.  But what about friendfeed?  That crosses the streams from bookmarking sites (delicious to connotea to citeulike), blogs, flickr, twitter... all places that have their own norms.  Or even, if you can't say the whole of one of these things has a norm, at least there are identifiable communities within these things that have norms (knitting bloggers, librarian bloggers, mommy bloggers, whatever).  So then, in a re-design of the site, friendfeed got rid of the logos for each of the streams (which many of us have re-added using GreaseMonkey) - even further mashing up these streams.  It turns out that instead of this causing a whole lot of confusion, hurt feelings, and people acting inappropriately (for the most part), it's caused the formation of new norms and ways of doing things - because we're all pretty adaptable, and it's basically the same people, with the same user ids, and because we know that people label delicious things differently for themselves than to share, for example.

So if it's ok for friendfeed to aggregate all of these things, then it should be just fine for journals to aggregate comments on their papers from all public sources, and provide them to article readers as commentary and context, right?  Here are some good possible outcomes:

• this could provide a filtering function - these people thought the article was good, so it's worth reading
• could make science go faster - spark new ideas, new collaborations
• correct errors faster
• provide context for the paper

However, the norms when people comment directly on the journal site are quite different. People think through their comments more. People are sometimes forced to use their real names (the names on their drivers licenses). There is probably more civility because these people might be the reviewers of your next paper! (see Nascent discussions)

So what happens if you cross the streams?  You cross the off-the-cuff, "oh, dude posted his 15th paper from his dissertation, d'ya think he'll ever do new work?" and "is this for real?" with the "I think the caption for figure 3 should state..." and "in your analysis, did you consider error stemming from ... " Ew. And some of these comments might be coming from the same people (oops!). If someone does a thoughtful review of an article on their blog they probably would want that connected to the article, but how about these other things?  It's public, therefore it's ok to use?  How do you tell which is which?

Dealing with journal articles, and commenting on them in a scholarly form has existing norms of behavior, from the offline world.  Between journal club members, you might get pretty rowdy, but you would never think of sharing that stuff with the author of the paper.  Questioning is good, but there are some comments that stay between friends.  

I'm not saying that this shouldn't be done, but it would be better if people could opt in.  Use a tag to have a comment aggregated or something. Then, when the comments are posted, mark them appropriately.  You'll still have people who aren't really involved in our online world who are completely confused by the comments, but it's sort of difficult to bring them from 0-60 with a page layout.

This is my story and I'm sticking to it, at least until someone on friendfeed tells me I'm all wrong :)