The world-changing power of tags

hkd
The Overlap
Published in
4 min readMar 2, 2016

It was May of 2005, and I was super-psyched about my brand new Palm Treo.

the freakin’ rad Treo 680

Remember those?

It was no iPhone but it was still pretty sweet. Playing around on the subway one day, I took this awesome photo:

(Terrible shot really, but remember, only 0.3 megapixels in that camera)

The poster behind my sleeping friend’s head was an advertisement for the Takeshi Murakami show “Little Boy” at the Japan Society. Giggling about my own hilariousness (sadly, something I still tend to do), I saved the photo to Flickr. At the time Flickr was a new-ish photo sharing site that was exciting because it was one of the tools that was starting to help people understand what the web was actually for. We all knew about the ways the web could speed up things we already did, like shop or transfer large files or share funny stories. Flickr was among the sites that was helping us see the things we could do that were unique, that had never really happened before, except by accident.

So, I dropped my awesome photo into Flickr and tagged it “little boy” and that’s when magic happened. All of the sudden my photo was part of a collection of photos like this:

Blob, Takeshi Murakami, Japan Society, May 2005

See! See! It’s the same giant yellow blob! Scroll up! It’s the same blob! OMG!

OK, so this may not seem exciting now, in the world of hashtags like #makedonalddrumpfagain, but think about it: a random photo of a photo of a thing (giant yellow blob, above) is instantly matched with other photos of that very same thing, just because I typed “little boy” when I uploaded the photo. It doesn’t contain any information about the show or the artist, and yet, the yellow blob appears. My tag didn’t contain any relevant descriptors at all. In fact, the tag means something completely different, and should have resulted in something like this:

google image screenshot search result for “little boy”

It seemed like magic, but it wasn’t AI, it wasn’t mind-reading, it wasn’t high-tech at all. In fact, it was so low-tech that it took about 10 years of many extremely bright people thinking and working with the web before anyone took a step back and thought, what if we just let people categorize things however they want? (note: Granted it probably required the right level of user adoption as well, so apologies to any brilliant people who did have this insight but were just ahead of their time and couldn’t make the magic happen without the crowds.)

Most people still thought about information in terms of the dewey decimal system. I was myself responsible for a variety of terrible web interfaces that required people to pick from endless lists in order to classify things (Apologies!). How could we trust ordinary people to pick the right labels? It would be chaos!

As it turned out chaos, on the web, was a good thing. Around the same time that I was giggling over the giant yellow blob, Clay Shirky gave a great talk called “Ontology is Overrated” and Thomas Vander Wal coined the term “Folksonomy” and we started to think a bit differently. We finally started to understand that maybe it was time to think about our design efforts as a way to guide behavior by applying as little structure as possible, rather than as a way to stem the tide of mayhem that was sure to surge if we didn’t make sure every term was completely planned out before hand. All of the sudden the sign of a successful product was users using it in unanticipated ways, not just when they were following along like sheep.

Fast-forward 10 years to a world of trending hashtags on twitter, and services to help you pick hashtags more likely to trend, and #blacklivesmatter or #jesuischarlie or #refugeeswelcome, or yes, #makedonalddrumpfagain among many many others. With just a few words you can connect with world-changing movements, add your voice to protests, to calls for change, or simply tune in to what millions of people have to say about a live event. It’s ridiculously simple and powerful, and likely unprecedented in human history, and many great things have and will be written about this topic. In fact, I’m assuming that there are, at this moment, dozens of Linguistics Phd candidates working on the definitive studies of the use of tags, and we’ll all be a lot smarter about them soon.

In the meantime, it’s worth a small moment of appreciation for the insight that simply stripping out tools of complexity and control (you must pick from our predetermined list of categories!!) and letting people make their own choices is one of the reasons the web is so powerful. Even though we think of it as a tool humans created and control, the web operates much more like a biological or economic system, whose complexities are too great for us to reduce to strict taxonomies or 5 year plans.

--

--