
HUMANS, Coffee in Tech // Invisible Machines S1E12
Invisible Machines podcast by UX Magazine
How to Vote for a Job in One Rich AI
You can also apply for a job in humans or one rich AI. You can leave your feedback not every guest is comfortable leaving feedback to our berry star us managers. At the end of the month you can we will share our profit with this organization in and the profit percentage will depend on on the votes that our guests makeOkay next we have here information about humans in one rich AI there's many many guests ask some detailed information so here you can tell you more about humans about our team about our mission and it can also tell you about hue like what does it doLike how why do we have this virtual assistant?
00:00
Transcript
Play full episode
Transcript
Episode notes
Speaker 2
Like, is it okay to use anxious to mean eager? And whether there's something wrong with that 10 items or less sign at the grocery store. Hi Steve, thanks so much for being here with me today. Thank
Speaker 1
you, Mignon. It's really a pleasure to be speaking with you. Yeah,
Speaker 2
well, I just find the work you have done fascinating and I can't wait to hear more about it. Let's just start with the usage panel, the entries in the American Heritage Dictionary that show how usage experts, what they think about common usage problems, you were in charge of that. And I just, I would love to hear at first how it all got started.
Speaker 1
Great. Well, thank you. Yeah, the usage panel has been an integral part of the American Heritage Dictionary from its conception up to the end of the publication of the dictionary in 2018. In the 1960s, when the first edition was still being prepared, it was an interesting time in the American publishing world because dictionaries were a very easy way for publishers to make money. So a lot of publishing companies would produce their own dictionary. And so one of the more well-known ones, of course, was Miriam Webster, and their third edition came out in 1961. And at the time it came out, it was seen to be unusually permissive. It really wasn't, but there were pundits who really ran with that idea. And your listenership might be interested in a book by David Skinner called The Story of Ain't, which describes that whole era and everything about the publication of Webster's Third in Great Death. Right, because the dictionary included ain't in a
Speaker 2
way that people didn't like.
Speaker 1
Yep, and that was a huge thing. And there was a publisher who saw a way to counteract that and proposed a dictionary that would be more prescriptive than descriptive. I'm sure that you're viewers are familiar with the whole prescriptive, descriptive debate. The truth of the matter is most dictionaries are actually very descriptive, and the American heritage dictionary is as well, even though it had a reputation of being prescriptive that came out from this period in the 60s, because it was originally conceived as an antidote, if you will, to Miriam Webster's permissiveness. And as part of that, the publishing plan was we're going to have this usage panel of a couple hundred linguists and people who used English in their everyday life, poets, writers, journalists, that kind of thing. And the intent was that these usage notes would help the dictionary decry all the permissive things that Miriam Webster had put out. However, as it turns out, especially the linguists who are on the panel, not to mention the actual lexicographers themselves who were brought on staff to rate and compile the dictionary throughout the 60s, being principal lexicographers, they themselves were very descriptive. And how the project ended up was very different than the way it was initially conceived by the initial publisher. There was a six-year, seven-year period where the usage panelists were being pulled three or four times a year, 20 or 30 questions at a time, starting with the letter A somewhere around 1963 and 1964, leading up till about a year before publication in 1969. And those several hundred usage notes became the initial block of notes that appeared in the dictionary. And it was very successful, and the dictionary got a lot of publicity mileage out of that and continued to do so throughout the next five editions. One thing that
Speaker 2
people often ask is, you know, who is this panel and what makes them qualified to determine what's right and what's wrong? I mean, they're experts in writing, but still, like, how were these particular experts chosen?
Speaker 1
I can't speak to the 60s because I wasn't there. But at that time, of course, the publishing world was very based in New York. So you see a lot of New York literary figures, and I would imagine that the linguists were somehow associated with people who are working on the dictionary. By the 90s, the odds, the editorial staff made a point of getting much more regional diversity so that you had people from all parts of the country, not just the coasts, and looking for different type of linguistic specialists, reaching out, branching out to different types of poetry, different mechanisms of expressing the written word, including people who wrote graphic novels like Alice in Backdill, as opposed to just people who were doing strict prose, visual artists who used words in their medium, Pulitzer Prize winners, that sort of thing, to really expand the pool.
Speaker 2
Nice. Nice. Who are some other people that were on the panel that the listeners might recognize?
Speaker 1
Oh, there were plenty. And I'm going to grab the book myself so I can write it. And your listeners, even though the dictionary ceased publication in 2018, the website still works, so listeners can go there and check it out. And then the usage panel list is included there too, and that website is Agedictory.com.
Speaker 2
Yeah, I still go there probably at least once a week. Yeah, I still
Speaker 1
go there once a week. Just looking at the list down here. There were close to 200 linguists and writers, writers ranging from Pulitzer Prize winners like Susan Lori Parks. We have Robert Reich, who was once the US Secretary of Labor. People like David Skinner, the author I mentioned of the story of eight linguistic professors like Deborah Tan and other writers like Amy Tan and Faye Weldon. There's a full list in the book or on the website. It's a really nice variety of people and opinions who have the authority to speak on a language. Nice.
Speaker 2
And Andy Holland Beck asked me if there were any sticklers who always chose the most prescriptive way or who always voted against changes or did anyone, you know, quit in protest over the results. Like, were there any people who were like that?
Speaker 1
During my 10 year, no, I was there for 22 years, executive editor as seven years, but worked on the with the panel results for all 22 years. People were very generally good natured about it. They, in fact, a lot of them found it to be fun. Like, sometime in the beginning when we were still using paper ballots, some people would always return it right away and then you'd get the next clump before the deadline. Starting in the mid-auds, we went to an electronic format, which was a lot easier to collate the information and everything. People usually filled it out pretty enthusiastically. And yes, some people were stricter and others were less so, but sometimes you'll find quotations from certain people anonymized saying why they may or may not have liked certain things. Also, on the on the on the website, A H dictionary dot com at the bottom, you'll find a link to the dictionary's Tumblr site that was given about eight to 10 years ago. Within that, there are a lot of essays that that I and some of the other editors wrote. And there were a few that talked about the panelists. There was one of the last panelists that was still around from the very beginning. I'm going to get his name right. I think it was Howard. I'm sorry, it was not Howard. It was William, William Zinser, who was one of the original panelists. One of his relatives wrote in with some of his papers. And there's a really nice article about some of his viewpoints in the 60s versus later on. There aren't, you know, there's only a couple dozen Tumblr entries, but I think some of your die hard fans would really enjoy going through that and seeing, you know, some of the, the essays we have about it.
Speaker 2
I'll put it in the show notes so people can find it. Another question someone had is, did the editors ever overrule the panel?
Speaker 1
The short answer is no, but it also depends on how you define overrule. Usually we use all the material except when we realized when getting the results back that there was a problem with the way that we worded the question where there's this ambiguity. And, you know, if it was like a yes, no question, you know, would you use this in your edited prose? It was hard to determine whether or not they were going after the thing we were testing or if there was a different problem with the sentence that we hadn't realized. In those cases, we would ask an analogous question the following year, you know, fixing it so there would be no ambiguity. Aside from that, because we were, we were using often percentages or a yes, no thing, we could almost always find a way to include the information. We needed in the definition. And if there was some outcry against it, that would be covered in the note. We as the editorial staff didn't really have a need to say, well, the use panel felt this, but we're like, do what you want. You know, we could, again, we could let the note speak for itself, right? And as a general rule, the panel was pretty perceptive of being able to distinguish, you know, if you say this thing, people will say, you know, you're not going to be able to do it. And then, you know, if you're saying, people will misconstrue you, it's wrong because if you use this, people won't know what you're saying. And on the other end, yeah, go ahead. The ship is sailed. And then, of course, everything in
Speaker 2
between. Right. Right. Well, that reminds me of one question. So I've been doing usage polls on Facebook for a month or so now. And one problem I've really struggled with is how to define the circumstances under which people think a usage is acceptable. And, you know, and like, how did you phrase it? How did the dictionary define what acceptable usage you were trying to pull the panel about?
Speaker 1
Right. In the questionnaire that we said to the usage panel, unless there were specific things to a particular question that we were going for, the general rule was, would you use this word of phrase in edited prose? So it was less about, you know, are you going to use this in casual speech with your friend, you know, in front of the grocery store versus you are, you know, working at a newspaper. Is this usage going to make the cut if one of your writer uses an article? The, for your listenership who has a print copy of the American Heritage Dictionary. If you go to the front matter of the dictionary, there are some essays and instructive matter that talk about the methodology is it were that is sometimes of interest. As a general rule, though, what we were looking for is whether or not it would be used in what we called edited prose occasionally. We would ask questions that were a little more fine grade. And it's like, you know, would you use this in casual speech, which he uses in formal speech, but generally we were going for editing prose.
Speaker 2
That's great.
Robb and Josh welcome Anna Yankovska and Helen Peklo, who helped bring an idea Robb had about companies and their surrounding communities to life. HUMANS, Coffee in Tech is a cafe on the first floor of OneReach.ai headquarters in Kyiv, Ukraine. Team members get their daily caffeine fix here, but it's also open to the broader community, and has become a dependable bright spot in people's lives on the front lines of an ongoing war. HUMANS is unique in that there is no hierarchy among the baristas. Managerial duties and other tedious tasks are handled by Hugh, an intelligent digital worker created using the OneReach.ai platform. This is an exciting episode that explores the litany of ways a modest coffee shop is pushing the boundaries of technology and organizational structures.