Yannick Trapman-O'Brien
patreon
Archive Highlight; “What’s in a Name?”
🕑 Added 2022-01-31 13:00:07 +0000 UTCPart 1 in a series on the Opening Few Minutes of the Telelibrary
Hey everyone,
I want to thank you all for your patience this month with Patreon content. Unfortunately, I’ve had some family matters that have needed care and attention, and so much of this month’s content has been pushed back. I’m grateful for your understanding.
That said, I’m glad to have this chance to share with you a little side-project that has become … much larger than anticipated. In celebration of 1000 calls in the Telelibrary, I’ve been crunching numbers to look at patterns and pathways made by User activity. There’s a few findings I’m excited to share, the first of which begins, conveniently enough, at the beginning.
If you’ve never experienced The Telelibrary—how did you get here? But also, please know there are some minor spoilers ahead. You can also increase your spoils and learn more about the System in a previous Archive Highlight.
All ready? Okay, please enjoy this Month’s Archive Highlight!
~
“What’s in a Name?”
On Beginning with an Introduction - Pt. 1
Scrolling through some of the many excellent active “User Names” in the Telelibrary
- The Telelibrary is a personalized, telephone experience for an audience of one; each User charts their own path through the ever-changing and expanding catalog, and no two calls ever follow exactly the same route. It can be hard to have a conversation about the design choices or interactive interface for a piece in which Users are given (and gradually take) so much control over their experience.
However, the one thing that is standardized for every call is the opening, so as we look at the ways Users are primed to participate by navigating choices, the beginning is a great place to start.
First off, an opening theme plays [1], and as it fades out, the User receives their first prompt for input:
“Welcome to The Telelibrary. Please state your name:”
To quote my favorite R. Eric Thomas book; “so much is already happening.”
For one, the logistics of Telephone Theater mean that I begin a call without really knowing any of the following: Can they hear me? Can I hear them? Is this the person who signed up, or someone else? Did they bring someone else along? Asking someone to state their name generally lets me cover all of those bases. It’s also an easy question to begin the necessary work of training a User to respond to voice prompts. For the format to work, The Telelibrary has to feel more like a phone call than a performance one can sit back and enjoy passively.
So we’re about 43 seconds into the experience, and making good time. I make sure to affirm and acknowledge the participation; the System explains it “already knew that,” but has been instructed “it is polite to ask.” This is absolutely a case of me trying to cram a joke in when I can, but it’s also a design choice; throughout the The Telelibrary, but especially in the first 5 minutes, I am constantly trying to strengthen two impressions for a participant:
1) “I am speaking to an automated phone System”
2) “this is not a normal automated phone System”
In a way, you could restate those impressions as “I know what happens next” and “I have no idea what happens next”—which handily summarizes the balance I’m trying to strike between forming expectations that help people participate and navigate the experience and breaking those same expectations to provide surprise, wonder, and encourage creativity and innovation.
Which brings us to 50 seconds into the piece. Let’s pick up the pace some. Next, the Systems ask the participant “What would you like to be called today?”
Again, a fairly simple question, but already the User experience can begin to diverge here. To show how, I ran the numbers on the first 1000 calls to the Telelibrary, which were completed by 746 Users.
Looking at the response to this question for about 600 of these Users[2], we see that about 22% of them took the opportunity to “play” under a new or assumed name. Many, many smarter people than me have written about the power in granting participants the ability to play as a character, so I’ll spare you the at-length about liminality, anonymity, enthusiastic play, and crossing the threshold of magic circles and simply say that for a piece that begins assuring you that “this System is here for you, to use as you like” even before your call begins, allowing someone to entertain themselves by “trying on” a new or silly name is a great tool.
And the other Users? 8% of callers in the group analyzed chose to use a shortened version of their name, or common nickname (e.g. “Ben” for “Benjamin” or “K” for “Karyn”). I think of this being a middle-ground between full-contact imaginative play and simply answering the question. The User in this case is to some degree increasing their familiarity with the System, or playing a slightly different role.
The next 23% of Users fall into a group I call “accidental names.” These are people who answered conversationally when asked “what would you like to be called today?” The System, dutifully literal and seemingly ignorant to all context, cheerfully assigns them their chosen name.
One of the many “games” of The Telelibrary is the way Users can toy with the limited, occasionally Bedelia-esque way in which the System attempts to understand human communication and commands. But I think this joke also serves to underscore for Users that what they say does, in fact, matter; the pattern of the System listening closer than anticipated is one of the more interesting and generative threads in the piece, so I’m always happy when I get to start it early.
And the remaining 46% of Users—whose chosen User Name is an exact match for the first name they provided on signing up for The Telelibrary—does that mean they are missing out?
I don’t believe so. For one, The Telelibrary is an experience where a massive number of interchangeable User actions combine into an infinite number of combinations. That being said, many of the themes we’ve evoked already can pop up in any number of different interactions — and I absolutely find the opportunity to sneak in my favorite jokes, bits, and motifs throughout the 50(ish) minutes of the experience.
What’s more, through the simple act of taking a name, even their own, the User gives me something to work with. It’s a fairly common sales trick to say a customer’s name frequently in order to build rapport and gain their trust, and one with some decent science behind it. It also gives me the chance to repeatedly use direct address, which helps discourage passivity (“Hey—I’m talking to YOU.”) It’s also interesting to look at this practice and remember that the System I’m portraying is some kind of virtual/digital interface (and/or AI, and/or algorithm, and/or robot-phone-neural-network-chatbot —see future postings for more discussing on the “soft worldbuilding” at work in The Telelibrary). In his book “A Human’s Guide to Machine Intelligence,” Kartik Hosanagar discusses many of the “trust-inducing interfaces” employed by the designers of our increasingly intelligent technologies. A key finding repeated again and again is that allowing Users some measure of control and customization has a dramatic effect on their willingness to cooperate with an interface “thinking” for itself. In the case of algorithms, Hosanagar cites research from Wharton explaining that
“... that trust in algorithms is insensitive to the amount of control offered to the user. As long as users have some control—however minimal—trust is significantly enhanced.”
All this talk of names and User control brings us to the next part of The Telelibrary’s opening:
“What would you like to call me?”
Which is precisely where we'll pick up next month.
And that’s all for the January Highlight! Good word, team: we made it about 75 seconds into The Telelibrary. Jokes aside, I think it's worth (over)indulging a bit on the opening of this piece, both because it's a fairly spoiler-free way of discussing some of the larger themes of the work, and because more than any other single part of The Telelibrary, it has been consciously, exhaustingly drafted, re-drafted, reworked and (over)designed. I can't think of a better case study of the way I work and the way I think about performance right now than to tackle this intro, and I'm excited to be sharing it with all of you.
More to come next month!
Until then — thanks again for your support.
- Yannick
[1] “An” because it is possible to unlock non-standard opening music—though we won’t get into that today.
[2] Why 612 Users out of 746? There's a variety of factors contributing to that gap. First, a number of Users actually came so early that they predate the System & User name prompts. Beyond them, a larger number of Users came before I began taking detailed notes on calls. My documentation for this piece has changed a few times, so when doing analysis like this, I try to look at a group of Users for whom I have comparable notes.
Comments
Yannick Trapman-O'Brien
A great question - I'm very interested in having other people be able to use/explore this data, but because much of it is personal information/personal adjacent, I think where I am currently at is I would want someone to reach out to me and explain what data they want and what they are looking for (even if "what they are looking for" is just to poke around). "Cleaning" this data would definitely be a big task, but I'd be excited to do it, and to learn what other applications this data may have—especially when we get to next month and talk about what people name the System. It's the same stance I've taken with "IF WE WIN" for now - on the website, I invite people to reach out if they have an interest in using the data collected of spending plans. I've also slowly started emailing professors in Philly who I think may be interested in that data-set. Very open to recommendations of folks I should reach out to!
Jacob Ford
Would you consider releasing any of this data as parsable text/CSVs/public docs?
Noah Thompson
Pure elegance in design and execution - memorable from the very first minute.
