The overuse of the term “AI” to market technology products has been out of control for some time. Educational technologies are no different. More and more I've been seeing “AI” products in edtech that are little more than slick visualizations wrapped around basic arithmetic.
Things that are not, by themselves or by default, “AI”:
simple and obvious things expressed as percentages, e.g. percentage of students participating in some activity
any graph and visualization that is not a line or bar chart
huge dashboards full of numbers
huge dashboards full of numbers with fancy labels of the form “Engagement Score” or “[StupidTradmarkedName] Score”™
Make it stop. Seriously, machine learning, deep learning and everything that might legitimately be called “AI” are interesting and awesome and powerful, problematic and potentially biased and also full of possibility. All of that is worth talking about and fair game to market as some branch of artificial intelligence; but selling elementary math and week 1 of Intro to Statistics as AI is just ridiculous.
Constraints expose the workings of technology. Errors and failures are invitations for critical assessment. Even if the technology does not work perfectly, the ways in which it failed, the exposing of invisible constraints, can prove successful.
I have a perverse love of technological constraint. Constraint can give rise to innovation, inspiration, and, in the aesthetic of minimalist computing more generally, elegant solutions. But there are plenty of times where technology has constraints that we can't see. A lot of edtech is like this, from the user side, because the marketing is almost always about automating and making things easier. Edtech marketers constantly hide the constraints of their products from users.
This is a mistake and a missed learning opportunity.
photo: students today, serious about capitalist outcomes
Interrupting my otherwise pleasant pre-New Years holiday, I made the mistake of reading this piece, headlined “More info is available about which college majors pay off, but students aren’t using it”. The gist is that data tying specific majors to earning potential is now available but (sacre bleu!) students aren't using this data in order to select majors as much as the people making said data think they should.
I have a lot of objections to this piece, including the way that it jams together quotations in ways that demolish all nuance. But the biggest problem is in the way it makes you think that it somehow makes sense for students to see education solely as a pathway to a job and, further, that that job should be measured on the singular metric of salary.
I was intrigued by this recent post by Tim Denning where he connects minimalism and a quiet ego. I don't buy that connection, as it seems like an extrovert's misunderstanding of introversion. (I would recommend reading, as complement, Susan Cain's Quiet), but it did get me thinking about how and if “minimalism” translates to quiet. Further, it made me think about how much that metaphor of loudness translates to technology. Some technology seems to yell, other technologies just whisper urgently in your ear, and some others sit quietly until called upon.
I have used Gradescope on a number of occasions and it remains one of the best pieces of edtech I have come across. It is one of the very few tools that has saved me huge amounts of time. The auto-grading is truly revolutionary for teaching at scale, at least with the kinds of teaching I tend to do. But the best feature of Gradescope is perhaps one of the least flashy parts. It simply has a really good grading interface. Here's why:
Amidst the signs of the season are various news pieces both reporting consumerism and, inevitably, questioning the nature of consumerism. This piece from Salon (“People didn’t used to be “consumers.” What happened?”, citing research about how simply using the word “consumer” to describe ourselves can change the way we act, made me think about the pressure on educational institutions to deliver products for their “consumers”— er, I mean “students”, and “growth” in curriculum or enrollment or other measures.
This 2017 piece about teaching computing to students in Finland without the use of ipads and other tech toys came to me from Pocket's algorithms today (they know me too well...)[^1]. The subtitle (“Students can learn the basics with a set of knitting needles”) is evocative, but the whole article is worth a read because it highlights an important principle in education and technology: “knowing how to use something isn’t the same as understanding how it works.” And, more pragmatically:
The Finns are pretty bemused by Americans’ preoccupation with whether to put iPads in every classroom. If a tablet would enhance learning, great. If it wouldn’t, skip it. Move on. The whole thing is a little tilting-at-windmills, anyway.
Some problems of digital tools we take for granted in teaching emerge as side effects of usage; others are due to mis-using the tool. Auto-grading and quiz (in)capabilities of modern LMS-es fall into both categories: side effects and encouraging bad test design habits.