Original image

There are Two Kinds of Hashtags—Which One Do You Use Most? 

Original image

How is language evolving on the Internet? In this series on internet linguistics, Gretchen McCulloch breaks down the latest innovations in online communication.

What's the point of a hashtag? Most reference works say that it's something like "a word or phrase preceded by a hash or pound sign (#) and used to identify messages on a specific topic." And that's certainly what you'll find if you look at Twitter's trending topics, from the frivolous like #FakeDogFacts to the political like #BlackLivesMatter to the utilitarian like #followfriday.  

But what about posts like the following? Does anyone really expect to see a bad 8th-grade glasses photo under #swag? Is #cantstopcrying going to help people looking for Les Mis reviews?

Of course not. Linguist Allison Shapp did a study of over 10,000 randomly-selected tweets and found that hashtags on Twitter come in two categories. Index hashtags are our first kind, the organizational hashtags you typically think of. Shapp found that they often contain links and were more likely to be favorited and retweeted—which makes sense, since they often refer to a real-world event. The second kind of hashtags are commentary hashtags, a social kind of hashtag that's more likely to contain other people's usernames.

Shapp also found that if a tweet contained multiple hashtags, they were more likely to be indexes (which tend to be shorter), and that index hashtags were also more likely to be integrated into the rest of the tweet, whereas commentary tended to occur at the end. And the more often someone tweeted, the more likely they were to use more commentary hashtags, although surprisingly the really frequent tweeters didn't use quite as many commentary hashtags as we'd expect.

One common style of index hashtagging, which Shapp called the "context template," looks like this:

Out of context statement in prose #context #context

For example:

Crazy in Love #beyonce #superbowl #halftime

Shapp's analysis is of Twitter, but it's easily applicable to other social networking sites that use hashtags. And since Instagram and Tumblr don't have such a short character limit, people there often use both index and commentary tags at the same time.

In this instagram post, for example, #CaturdayNiteDerpOff is an index tag (because the internet is amazing) but #JustHereForTheDerps and #GonnaGetDownAndDerpy are clearly commentary.

'Hello? Is this the #CaturdayNiteDerpOff? May I join?' #JustHereForTheDerps #GonnaGetDownAndDerpy

A photo posted by Gremlin (@gremlinthecat) on

What about when people use "hashtag ___" in speech? Well, it's not going to help people search through the airwaves to find a particular spoken utterance, so spoken "hashtags" are near-universally the commentary kind. I didn't follow people around with a voice recorder to get examples, but there's an even cooler way to demonstrate it. If you search for the word hashtag itself on twitter, you find a number of people using it without any hash mark at all to clarify that they really want you to interpret their hashtagging as commentary.

Not every hashtag falls neatly into one category or the other: index and commentary hashtags are more like two ends of a hashtag continuum. Somewhere in between is the #marketing hashtag, where #brands #hashtag #random #words that are #topical but which no one is probably searching for. And Shapp points out that hashtags sometimes start as one-off commentary hashtags but get picked up by a larger group of people and become indexes, making them difficult to classify. One common example of hashtags on this boundary are meme hashtags, such as the "problems" set—#FirstWorldProblems and #90sProblems are indexes, but people also coin one-off "X problems" hashtags as commentary on any problem characteristic of a particular group.

You could call those #hashtagproblems.

Original image
iStock // Ekaterina Minaeva
Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning
May 21, 2017
Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
Name the Author Based on the Character
May 23, 2017
Original image