Controlled social metadata?

Peterson’s article, “Beneath The Metadata,” brings up some points that were brought up in other articles and countered in yet other articles.  It all seems to bring up two conflicting ideas:

  1. Tagging is, by definition, an activity mostly done by people who have little idea how subjects work.
  2. There is so much stuff to tag that, even if everyone were trained in subject analysis (for free), we wouldn’t get to it all, and we still couldn’t afford it.

Is there any way to reconcile these two statements?  I was driving along the other day when I thought about the settings on my GPS unit.  My GPS company allows input from users on map corrections (this road doesn’t exist, this road is one-way, etc.).  When I update my GPS, it also downloads a small file of “community updates”.  In my GPS settings, I can decide on the following options:

  • Include any community correction
  • Include only corrections submitted by a certain number of community members
  • Include no community corrections (only official maps)

Is this a middle ground?  We can allow tagging, allow users to approve of current tags, and lets browsers decide how popular a tag needs to be before they see it/use it in searches.

Well, some differences, from the GPS perspective:

  1. On the bright side, we don’t expect a resource to change once it’s up, unlike the road system.  On the not-so-bright side, the road system (and business openings) is being added to at a much slower pace than digital media.
  2. The GPS only allows a small number of definite operations (new road, road closed, one-way, move business, etc.)  Tagging allows only one operation, but it’s a VERY unconstrained one.  In fact, that may be an idea: faceted tags.  Have tags be in response to questions, like “when did this happen?” or “who is in this picture?” or “what activities are happening here?”
  3. Checking changes in GPS-land is a very slow process.  Someone really does have to drive out there and look.  Checking tags in digital land is much faster, but there are so many more tags to check.  Over and over!
  4. In both cases, popularity constrains moderation.  I lived in Mississippi, in a town of 25,000 people, and community changes just don’t get checked.  And businesses are off by entire blocks.  Tag popularity is constrained by the people who originally see a resource, the percent of those who care to tag it or approve of tags, and (if moderating) the expertise of the moderator.  Going back to the expert tagging example, if someone tags an old painting with “chain of office,” (and assuming anyone cares enough to upvote it) how much work would a moderator have to do to find out what a “chain of office” is, what it’s referring to in the painting, and wouldn’t that be as much work as to become an expert and tag it in the first place?  Or maybe we’d need “authenticated taggers,” like Twitter has authenticated accounts, who can tag without moderation…

But still, I like the idea of being able to personally control my consumption of tags to the same degree that others have control over the production of them.  I really would like to be able to say, “don’t even bother me unless this tag has 15 people upvote it or it’s moderated.”  Thoughts from the colleagues on this or faceted tagging?

Controlled social metadata?

4 thoughts on “Controlled social metadata?

  1. I don’t know if this really helps, but I left this comment on L’s post on tagging. It’s one way to consider the general usability of tagging:

    One premise of social tagging is related to what we call “high precision/low recall” searches. IOW, I don’t need ALL dachshund images, just a couple of them (and NO doxie images 🙂 )

    Based on this premise, we don’t need exhaustive tagging practices, rather, just “good enough”, and if we invite the general public in via crowdsourcing, then we up the “findability” ante (in a good way)

    Liked by 1 person

    1. I begin to suspect that adding more users to tagging doesn’t necessarily increase findability so much as it amplifies the pattern of findability from fewer users. In other words, the more users tag, it’s not so much that better dachshund pictures will be found, but that the good-enough pictures that were found early continue to be found more and more. I’m thinking that a Monte Carlo simulation of a large archive with tagging will prove interesting, and I haven’t found one yet. Who’s up for a research project? 😀


  2. Mapping is one of those fun things, that is always changing there is a highway near me that moved (long story) so every time going down the highway it will say “correcting route… recalculating ..” And this is a hard part to get on because it thinks the location is somewhere else. I gave up on having data on the smart phone to save money this semester, and realized how useful it can be to get directions. Luckily I found some apps that don’t require constant Internet connection, and utilize from their site “
    … OpenStreetMap is completely created by people like you, and it’s free for anyone to fix, update, download and use.” So basically similar to Wikipedia, you are hoping that everyone is putting in the correct data. Not as powerful as Garmin or Google but will hopefully get you to a city center.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s