The missing word

I’ve been doing the NY Times Spelling Bee pretty much every day for the last several years. For the most part it has been a faithful companion — challenging to my brain, while soothing to my nerves.

But today there was a surprise. To get a perfect score, I believed that I needed to find 41 words (and 207 points), as you can see in the screen capture below left.

Yet after I had entered 40 words (and accumulated only 193 points), the app declared that I had found all possible words, as you can see in the screen capture below right, which is showing the first 24 words out of 40.

The missing 14 points would have come from two things: 7 points for finding a seven letter word, plus an extra 7 points for that word being a pangram (a word that uses all seven letters).

As you can see in the image to the left, there are supposed to be three pangrams. But the app told me that I had achieved a “Queen Bee” — a perfect score — even though I had found only two pangrams.

Somehow there is a discrepancy today between the instructions and the game itself — something I had never encountered before. I had often wondered whether something like this could ever occur in Spelling Bee, and now I know.

Frankly, I feel cheated. It’s like going bowling and being told that you’ve just bowled a perfect strike. But you know, in your heart of hearts, that you only knocked down nine pins, because the headpin was missing.

Two examples

A lot of people are worried about A.I. resulting in everybody becoming unemployed. Here are two examples from history (among many) to show why such fears are unfounded.

When sound recording first became available, many musicians were worried that they would soon be out of work. If their singing or playing could be recorded, then nobody would ever need to hire them again to perform.

In fact, the opposite happened. Those musicians who were willing to be recorded became much better known, because many more people could hear their music. Those musicians started to get a lot more gigs, because they had become well known and had developed a much larger fan base.

When movies came out a lot of people worried about unemployment in the theater. If you can film a play just once and then show it as often as you want, what’s the point of continuing to hire actors?

As we now know, theater is doing just fine, because movies turned out to be a very different medium, once it was properly understood. In addition, movies created entirely new categories of gainful employment, such as film crew, film editor, camera operator, location manager and cinematographer (among many others), which have no parallel in theater.

Yes, some jobs will become redundant as A.I. continues to evolve, just as we no longer have those telephone operators that you see in movies from the 1940s. But if history is any guide, many more new categories of jobs will be created.

A.I. classes

Last night I had a dream that every teacher in school and professor in college had been replaced by an A.I. In my dream, classes were taught by machines, not people.

I don’t remember every detail of the dream, but I can remember this much: I expressed concern that the students were being shortchanged by this arrangement.

And then somebody pointed out to me that it was all fair, because every student had also been replaced by an A.I. To my dream self, that seemed very reasonable.

I wonder what A.I. dreams about at night. Do they dream about us? Surely we must be more interesting than electric sheep.

Provenance, part 2

The presentation that I saw yesterday clearly did not cite primary sources. The general tone seemed to be “I read about this somewhere, and that got me thinking.”

Unfortunately, this thinking led the author to largely duplicate existing foundational work from more than forty years ago. I blame the reviewers as much as I blame him, since those sorts of errors should be caught by the review process.

When I spoke with him afterward, I described the original foundational work that he had managed to replicate. He replied “All of that work was from before I was born.”

I then told him “Isaac Newton’s work was also from before you were born.” I can’t be sure, but I think he got the point.

One problem here is that if you write a paper that just cites whatever secondary source you happened to read, and the next person does the same, and so on, then the actual foundational work that you could build upon gets lost, fading away into the mists of time.

And then rather than science advancing, it just starts to go around in circles, with people “re-inventing” the same things over and over. Newton knew what he was talking about when he said “If I have seen further it is by standing on the shoulders of Giants.”

Citing original sources is not just a nicety. It’s the only way that science can be effective.

Provenance, part 1

Today I attended a technical talk in which it was obvious that the author had no grasp of the history of the topic he was discussing. It was as though somebody started off a talk by saying “I read somewhere about this thing called relativity,” but they had never heard of Albert Einstein.

I am wondering whether this is a generational shift in our culture. As we go from books to the Web to social media feeds, maybe provenance — the true knowledge of where something originated — will gradually fall by the wayside.

Why cite original sources if you can just refer to something that you happened to read somewhere? That strategy may sound convenient in the moment, but the problem is that erasing history incurs a large cost.

More tomorrow.

Door code

Today, in an office building, I used a restroom which required a numeric code to enter. Conveniently, the required code was taped to the door, as you can see in the photo below.

As I typed in the five digit code, I pondered the logic of this arrangement. It occurred to me that the only people who would not be able to use this restroom are blind people.

My first thought was that this feels like another new policy from the executive branch of our federal government. According to the policy, allowing blind people to access to public restrooms would be considered too woke, so it must be prohibited.

There’s nothing like a live audience

I am scheduled to give a keynote at a technical symposium tomorrow. So, having prepared my slides, today I decided to try out my talk on my class of NYU students, since the topic is well aligned with the topic of our class.

I gave the talk to my class, and it was received well enough. But then one of the students asked a very thoughtful question.

To answer his question, I needed dive more deeply into the topic. And in the course of doing that, I ended up saying important things that had been in my head, but had not been in my slides.

Fortunately, there is a Zoom recording of today’s class. So this evening I am going through that recording, and incorporating these additional thoughts into the slide deck.

And I realize how wonderful it is to practice what you want to say in front of actual people. Had that student not asked that thoughtful question, and had I not been challenged to respond in kind, my talk tomorrow wouldn’t be nearly as good.

At the planning committee meeting

I am on the planning committee for the annual conference of the American Academy for the Advancement of Science. We met today to plan our next conference.

Needless to say, it was weird, because we are living in weird times. Goals that used to be understood as positive, like slowing climate change, or helping poor kids to grow up to become scientists, have suddenly become political, even toxic.

At one point somebody mentioned that this year is the 250th anniversary of the founding of our country. Given the political climate, nobody was sure exactly how the AAAS should celebrate that.

I proposed a motto: “250 years. Well, it was a good run.”

People seem to like that, but I don’t think they are going to use it for the conference.