A discussion broke out recently on Friendfeed about online commentary on scientific articles. The discussions were interesting because, for the first time in recent memory, there was disagreement about something fundamental. I view this as an extremely positive sign that out community is starting to grow and incorporate people outside of our core group. In fact, if there’s no disagreement, it’s probably a sign you’re doing something wrong.
The disagreement went in two ways in the two different comment threads. In the second one, genereg promoted the idea that PLoS comments would be more abundant if there was true anonymity afforded to commenters. The other side of the argument, argued by Cameron Neylon(Friendfeed) (LinkedIn), Deepak Singh(FriendFeed) (LinkedIn), and Neil Saunders (FriendFeed) (LinkedIn), was that Real Names™ are important and desirable for online comments. The threads of so many different tangential discussions are running through here that I needed to take a second to write out the background ideologies in play.
1) Online participation depends on lowering the barriers to participation, and unlogged anonymous comments lower the bar to the absolute floor.
2) Online participation depends on a functioning ResearcherID system so commenters can actually get credit for taking the time to write a comment.(Related Blog posts from neilfws, Martin Fenner, Deepak, Peter Murray, and others on friendfeed)
3) Some people don’t want to sign their name to a critical comment, so they just refrain when they don’t have the anonymous option.
4) People comment where there’s a community in which their comment will get exposure and contribute to a discussion.
5) Comments attached to scientific articles aren’t for the purpose of stimulating discussion, but for aiding interpretation of the paper in subsequent years.
It’s apparent that there’s a little bit of talking past one another regarding these points, which is why I’ve collected and laid out the arguments here where they can each be commented upon. My initial impression is that comments as part of a discussion will mostly be made wherever there is community, which isn’t PLoS at the moment. However, comments intended to be interpretive aids for the paper will be made on the PLoS site, and registration won’t be much of a barrier for someone sufficiently motivated to write such a comment in the first place.
In the first thread, which developed more subsequent to the discussion on the second, Ian York (FriendFeed) (LinkedIn) commented that he felt it was rude to take his throwaway comment on friendfeed and re-post it on PLoS. Björn Brembs (FriendFeed) (LinkedIn), Bill Hooker (FriendFeed) (LinkedIn), and Cameron Neylon then pointed out that anything posted on a public site is pretty much free for the taking whether anyone likes it or not. The issues I’ve identified here are as follows:
1) Don’t make a public comment if you don’t want the whole world to hear it. This is especially true for comments to the media, which none of the members of this thread are (or are they?)
2) People will comment in different ways in different places, and, while linking to one from the other is always possible and always within the rights of anyone to do, it may occasionally be rude.
3) Deliberately taking someone’s comments out of context is rude.
4) You get the broadest view of reactions to an article if you don’t limit yourself to comments posted directly on the site, but rather go and find comments in the communities where they’re being made.
5) PLoS could have more comments if it went and found the comments in their communities, instead of expecting the comments to come to them.
6) It’s not technologically feasible to get permission to index every tweet or friendfeed comment made by everyone.
It’s apparent that some of these ideas are also in conflict with one another. To resolve some of these issues, perhaps it would be helpful to view them in light of yet more general issues.
1) Technology should adapt to people, not the other way around.
2) Online, your identity is your content, not your Real Name™.
3) Free and open sharing provides the greatest good for the greatest number.
I don’t believe these ideals are in conflict with one another. All we have to do to make this work is to find a way to aggregate things and re-share them such that context is preserved. As the friendfeed redesign made clear to most of us, tweets are totally different items than blog posts. I think it’s a UI issue. On the right, you’ll see a sidebar containing my most recent tweets. Would a sidebar on the article containing “aggregated commentary about this article” with short snippets and a link out sufficiently respect the need for maintaining context while also respecting that viewers of the article want to see the broadest picture of the engagement with the content as possible?
What is most fair and logical is rarely what happens. The system promoted by the guy with the greatest motivation or most money is generally the one that wins out, so perhaps that can lend some perspective to this discussion.
Thanks for the post. I would be interested to discuss the topic later, now just a quick remark: I think, according to both the Netiquette and Scientific Ethics, you should replace my name (which you have figured out from the context of other discussions) to the nickname that I decided to use in that particular online community.
Sorry, genereg, I didn’t see it as a breach of any etiquette. I intended it to be helpful.
no problem, fine, I am using a nickname as a proof of principle for this discussion.
I still don’t quite see the point you’re trying to make. Your nickname is a click away from your “real name”. Would you go into a little more detail about the point you’re trying to make, because I thought it was about anonymity, but clearly that can’t be it, if you’re linking your nickname to your blog upon which you’ve (apparently) posted your CV.
a click away is enough for not getting in Google
Um, no, it isn’t. I’ve disabled the addition of rel=”nofollow” to links in the comments.
I’m finding even this back-and-forth discussion between “generag” and “Mr. Gunn” fascinating. In a debate between pseudo-anonymous and Real Names™, I’m inclined towards people signing their real names to comments. That is, if they want them to be taken seriously. An unsigned comment or a pseudo-anonymous comment that doesn’t give an easy way to track it back to a real person devalues the comment in the framework of scientific debate.
Dear Peter,
As you know, comments in the internet are different from the standard scientific debates. We still have to learn how this works. What we know for sure is how discussions in the Internet work in general. What we know from the internet forums and related experiences, is that the most democratic and open discussions happen when people have the option not to disclose their names. This of cause requires some moderation, but that is a minor issue. This is how internet works traditionally. In many forums and large internet communities such as the LiveJournal, there are options for both anonymous and non-anonymous comments. People choose to disclose or not to disclose their names, but that is always optional. These are the internet traditions. The so-called social networks, where everything is non-anonymous is a new flow in the internet history — several projects have tried to use social networks for scientific collaborations, but up to now this is mostly in the test stage, and it is not clear whether scientists will like this or not. We have already witnessed several unsuccessful projects of this type, but this does not mean that one day one of them will work. The FriendFeed room, where this discussion has started is probably the best place of this type.
In addition to the internet traditions of anonymity, in Science, there is a well-established tradition that the peer-review is anonymous. It just cannot be conducted the other way. There are no alternative solutions to this currently. I am not sure whether you are aware of the idea of online commenting, but basically one of the ideas (the one which is behind the journal discussed in our original thread) is that the major part of discussion starts _after_ the article has been published. In this case, some elements of the anonymous peer-review have to be somehow substituted by the mechanisms suited for the online commenting. It is very difficult to propose something, which would completely substitute for the optional possibility of anonymous comments. There are several problems associated with the requirement to register in order to comment, ranging from simple technical difficulties to more complex sociological problems, some of which were mentioned above. Hope, this clarifies a little bit the discussion. I will be very busy in the next days and will not be able to answer soon.
— Best regards
I am also finding this discussion about online discussions among scientists interesting.
I tend towards wanting participants of online discussions not to be completely anonymous. In terms of scientific papers, the published ‘comment and reply’ format certainly includes the commenter’s full contact information. And, at least in my field, there are several instances of heated and constructive debate within this format. I don’t know of any published ‘comment and reply’ that has an anonymous commenter (again, within my field).
Maybe I haven’t experienced enough online discussion as others, but the reasons that genereg argues for maintaining the ability to be anonymous (namely, that it results in the “most open and democratic discussions”) might be true for a lot of online forums, but I’m not convinced that’s the case for science. My experience has shown the opposite … the venues that allow full anonymity (i.e., no nickname, no pseudonym, no linkage to other content/comments) result in a ‘wild west’ situation requiring conscientious and time-consuming moderation. But, again … my experience is perhaps limited.
But, at the same time, genereg brings up how peer review is sometimes (or can be) anonymous — I haven’t really thought of how that might fit in. As others have said, that is during the review process though, not comments that come after publication. As I noted above, why would commenters need to be anonymous? If they saw the presentation of that same paper at a meeting and wanted to make a public comment, it would be … well public. If they don’t want to make a public comment, then they just tell their colleagues why they think it’s wrong or whatever. Most of us like to have our ducks in a row, so to speak, before publicly commenting on such things. To me, it should be the same online (especially online in a more ‘official’ venue like a journal comment thread).
Anyway … very interesting and important discussion to be having since we are in the midst of shaping all of this.
note: I’m jumping in here and have not followed every specific comment from each person that led to this post and am mostly interested in the generalities anyway — I apologize if my lack of context is obvious; feel free to delete my comment if it adds nothing to discussion (but I simply have no inclination/time to dig back into the lineage, sorry).
Your comments are entirely appropriate and welcome, Brian, or should I say, clasticdetritus.
Ultimately, different sites will adopt different “cultures” and implement a corresponding comment policy. Anonymous and non-anonymous comments will lead to different types of conversations and both may be valuable. Look at 4chan (a site for the group Anonymous), and their “anonymous only” policy (http://en.wikipedia.org/wiki/Anonymous_(group)). The community is extremely offensive, but is by far the most innovative and dynamic culture on the net. They have produced some of the most interesting memes on the internet (lol cats for example). This is a good example of an extreme case of internet anonymity.
However, it seems like having names or identities attached to comments in the scientific community will be the way to go most of the time. Why no adopt a technology like Open ID for the science community? Users do not need to keep registering at each site. They get one log in that works for multiple sites and webmasters in the community would adopt the technology. This would also allow for some interesting aggregation (mashup) applications.
Thanks for the comment, Webster. I believe there actually has been some discussion about using OpenID as a contributor ID. I still think it’s a good idea, mostly because the arguments against it cite the distributed nature of OpenID as a negative, whereas I view that as a strong positive.
The cultural fermentation that goes on at 4chan is absolutely awesome. The only way they’d be in danger is is the recording industry has its way and makes us all have persistent identities attached to every packet we send. Probably the best way to define my position is to say that I support only voluntary use of real names. Absent extrordinarily intrusive authentication systems, your identity is your content, not whatever you type in the name field, anyways.
The issue that is being discussed goes further than just science. Some governments like China and commercial corporations try to to limit the freedom of information flow in many ways including the internet control. Science is just a small piece of the society.
“When researchers consider the issue of the anonymity of research participants, concern may be most likely to focus on how it can be maintained, particularly when under pressure from authorities to divulge identities” (Grinyer A. (2001) Ethical dilemmas in non-clinical health research, Nursing Ethics, 8 (2), pp.123-132.).
There is no reason to think that adding comments to papers and other formal forms of science discourse should be any different or end up behaving any differently than the way we do it right here on this site.
Anonymity and real names both have their place and their effects. In practice, I think starting without restrictions, that is, allowing anonymous interactions and no moderation and some some spam blockage, and then tightening as needed is the best way to go. Each service, each community needs to get a feel for what works.
My personal bias is that real names are the most useful way to go, especially in science discourse. The comments above, made in response to this post, are indeed peer review, which in no way “needs” to be anonymous. The baggage of print publishing has set up expectations of how to conduct science discourse. I think they are outdated and invalid in the new Social Web.
I blab on about it here:
http://www.molecularist.com/lifeblog/2009/05/talk-the-future-of-science-publishing.html
But, not to be missed (and I’ve been meaning to post about it), from PhD Comics:
http://www.phdcomics.com/comics.php?f=1178
Pingback: Science in the Open » Blog Archive » It’s not easy being clear…