Is the Lazy Sheep bookmarklet for a good idea?

sheepTagging captures the variety and commonalities in thinking about an object. When many people tag a url, it turns out that some of the tags are common, while some are unique. A lot of people, making decisions independently act in an intelligent fashion, and provide value for everyone. manages to balance the individual’s selfish motivations, and the group good, in a manner I have not seen many social systems do. It’s a delicate balance, but maintains it very well. Let us look into how it maintains that balance, and why I think that the Lazy sheep bookmarklet disturbs it.

Cognitively speaking – just what is happening at the individual level when you tag a url? I encounter an article I would like to bookmark. Semantic networks related to the article are already active in my brain. When I press “post to” it’s not much more work to note down some of the more important associations. captures a process that happens at the cognitive level anyway. And in the process of capturing the associations, it reinforces them. When I note that this article is about “tagging” and “cognition”, I am explicitly informing myself that I am currently interested in “tagging” and “cognition”. So the process helps me keeps my bookmarks, and tells me a little about myself everytime.

At the social level, the moment I post to, I can see if there are others who are interested in the same article. If so, did they use the same tags? I can see how others think similarly and differently, and also see the long tail of idiosyncratic tags.

The whole thing works because I make an independent effort to tag the article, and find out posthoc how others tagged it. That is an essential ingredient of

In the Wisdom of Crowds, James Surowiecki identifies four factors that denote a wise crowd:

  • diversity of opinion (each person should have some private information, even if it’s just an eccentric interpretation of the known facts)
  • independence (people’s opinions are not determined by the opinions of those around them)
  • decentralization (people are able to specialize and draw on local knowledge)
  • aggregation (some mechanism exists for turning private judgments into a collective decision).

The Lazy sheep bookmarklet violates two of the four principles – “diversity of opinion”, and “independence”. The bookmarklet is described as a

“a bookmarklet that auto-tags and auto-describes your bookmarks.”

It makes some sense at the individual level – I can gain from the wisdom of the others, without doing any work. But even at the individual level, there are disadvantages. First, the auto-tags might not capture my idiosyncratic associations (reducing findability when I look for the article later on). Second, it replaces the self-knowledge with social knowledge. Instead of a moment of reflection on my current interests, I simply find out how others think about the topic. Social knowledge in the context of self-knowledge is a beautiful thing, mere social knowledge just encourages the sheep mentality (which is the point of the bookmarklet I guess).

At the social level (which is what worries me more), if enough people started doing this, the value of would be diluted. We would loose some of the richness of the longtail, and just reinforce what the majority is saying. The first few people who tagged the article would set the trend – others would merely follow.

(the sheep picture is attributed to Duchamp on Flickr. Thanks for sharing)

7 thoughts on “Is the Lazy Sheep bookmarklet for a good idea?

  1. Hi Rashmi,

    I left a longer comment on my blog, but the trackback apparently didn’t stick.

    Short version: I think auto tagging can be very helpful, but it might be like using PowerPoint templates: after a while everything starts turning out the same way if you’re not careful.

  2. I replied on your site as well.

    I think the important thing here is choice. The user needs to display some choice in the use of the tags. Maybe the system can offer recommendations (of their own, the group’s, or some friend’s tags). But the tagging should not be completely automated. The user should look at the recommendations and decide which ones are relevant or not. And add more if needed.

    Mimicing is part of human nature. We learn from each other, we copy each other. Heck, we even yawn when the person next to us yawns. We unconsciously mimic each other’s mannerisms. As such, some methods of supporting mimicing are definitely part of a social system. Question is how much. IMO, completely automated tagging is not good idea (if it gets wide adoption).

  3. It depends on the audience you’re tagging for. I agree with the idea that it makes more sense to let someone tag independently, and then compare. Childhood education is based on a variant of this model, and it’s also the reason I like to see movies before having read too much about them. I always hit metacritic *hard* immediately after returning from the theater.

    In the case where you’re tagging for someone else, though, Lazy Sheep makes a little bit more sense. It helps you adapt to existing categories, and makes your bookmark a “me too!” instead of a “look at this…” My objection to Lazy Sheep is that the general userbase of is just not that interesting a group to borrow knowledge from. If my suggested tags are all going to be “blog”, “web” and “humor”, then no one’s needs are being served.

    I’d be a lot more interested in a tag suggestor that is more aware of the sorts of domain-specific language in use within particular communities of practice,

  4. Rashmi, I really like this piece. I have some problems even with the recommendations, not that they are not helpful (they are), but I often miss cross tagging with other terms I associate with the same object. The refindability of the bookmark is really tough if I did not include these tags that cross discipline boundaries. If I am tagging an object with folksonomy I will also want to tag it with tag (and its variants) as well as ontology (if it applies). If I do not do this I may have difficulty refinding the article at a later date.

    When tagging for others I am not so sure these recommended tags actually help as the bookmark already contains these tags. In narrowly distributed groups, like, the recommender systems can decrease cross-cultural and cross-discipline tagging. The recommended tagging follows the norms of that community and if the individual has interests or vocabulary terms that are relevant not germane to that small group’s ontology the recommendation system will not prompt for inclusion. In turn the recommender systems enforce or encourage a narrow focus rather than a more distributed approach, which brings with it the advantage of providing synonyms and a bubble-up thesaurus, which is what I find insanely helpful and a rare occurrence.

  5. I stumbled across this piece and found it very interesting. You seem to be suggesting that this Lazy sheep bookmarklet results in people being lazy about their own learning?

  6. Lazy sheep bookmarklet results in people being lazy about their own learning?

    Thats one way of putting it. My point is that tagging is effective since it is good at eliciting associations from people. Automating the process of tagging (while pragmatic at some level), is likely to reduce the data quality. IT definitely encourages laziness – but that would be fine (the best systems often let people be lazy and not have to do the work that the system can do for them). Tagging (at least one a type of system is not one of those cases.

Comments are closed.