Archive Highlight; "A Rose by Any Other Name" (Patreon)

This is Part 2 in a series on the Opening Few Minutes of the Telelibrary. If you haven't read Part 1, or you need a refresher, it's recommended you begin at the beginning. Otherwise, you're welcome to join me here - 75 seconds into a Telelibrary call.

"What would you like to call me?"

There are countless stories of humans and our capacity to pack-bond with anything—technology very much included. However, regardless of whether a character is portrayed as being non-human, non-sentient, or human, I have found that allowing a participant to name a character tends to hasten the development of comfort or intimacy, and to create a small but important sense of ownership. In much of my work, I try to intensify “one-on-one” performance by evoking the idea of “one-on-none” performance: a piece where you interact with a responsive, engaging presence, but are also the only person there. This sense of ownership thus works to not just increase your engagement, but to enhance the sense that you yourself are the author of your own experience, and thus reflected in everything you encounter.

The name you give to the System also becomes the most commonly used, all-purpose command in the piece, as Users ask questions by calling for the System by name (to answer the obvious question: yes, Users call me “Alexa” sometimes—sometimes even when they named me something else). As a result of all of these things, the System name can inspire many of the things Users will end up projecting on their auditory guide through The Telelibrary: personality, attitude, intention, and more—all of which I actively work to pick up on and amplify.

Users often speak to me about “their Telelibrary,” and have mused at length about the existence of endless variations of the System (such as in this excerpt of a User-submitted reflection on the piece—User Submitted Poem #2):

As a performer, asking for a name is one of those small choices I love deeply in interactive performance: a small, easily accomplished investment of energy and creativity for a participant, which grows quickly in meaning and value. It becomes a shared language between us, and a measuring stick for how their attention and engagement is waxing and waning — helpful in any interactive work, absolutely vital in The Telelibrary, where I have so few tools for "checking in" with the audience.

But what happens when we zoom, out, and look at these names from a larger perspective? More specifically, what if anything can we learn from looking at the names given to the System in the first 1000 calls?

Scrolling through an early list of names given to the Telelibrary System

One major take-away is that even a simple question like “what are the most popular names for the Telelibrary System?” turns out to be more complicated than anticipated. After reviewing the data a few ways, I think the best and most representative sample to look at for that question is all currently active Usernames: every time a User accesses The Telelibrary, the User name and System name they choose are updated in the notes on their User ID. When we look at those names, we find this:

For the record, 11th place is a tie between “TheLibrarian” and “IDon’tKnow,” which amuses me greatly.

As fun as the popularity contest is, I do think when it comes to describing trends, we need to go deeper. You’ll notice that “TheLibrarian” is quite similar to “Librarian.” So if our question goes beyond popularity to ask “what kinds of names do Users give to the System,” we can start to make groupings. For this, I looked at the names Users gave the System on their first visits2, and then started to cluster names that were functionally and thematically similar.

Witness the power of Friendship, Libraries, variations on the name of the piece, and  monosyllabic male names.

I don’t claim to know the “why?” behind any of these, but I do have some best guesses. The variations on Telelibrary and Library seem fairly straightforward: Users are applying deferring to what little information they already have. The fixation on “Friend” is more curious, especially considering that every recorded riff on the name of the piece combined is still only just as frequent. Certainly some of the presumption of friendliness comes from the way I play the character, though at this point I have only spoken 39 words. Instead, I sense that this impulse to frame the technology as a Friend falls somewhere between being optimistic and prescriptive.

As for the male names? If you combine all the single-syllable-conventionally male names in the top 10 (with “Sam” being admittedly far more androgynous), they add up to a frequency of 63 - 2nd place, with “Friend” and “Telelibrary” derivatives tied for 1st. Why do these names appear so often in the first 1000 calls? Beyond shrugging and offering “the patriarchy?,” I don’t feel prepared to theorize on the forces driving these selections (though I would confidently guess that the primary language/region of my audience is a big factor). Instead, I’m left to wonder - what do we think these names have in common? A presumed friendliness, or proximity to libraries? Or is it just their brevity? Here, after all, are the 10 most commonly appearing “conventionally human” names:

Is there something generally familiar, or easy, or trustworthy about these names for my participants? And if so, how could we use that information?

But that’s a question for another time. For now, even though we’ve only covered the first 100 seconds of The Telelibrary, I think it’s safe to say “I’ve learned a lot today.”

~

That’s all for the February Highlight. I’m going to bounce to other projects for the next month or so of Archive Highlights, but if you’ve got further questions about the Telelibrary you’d like to see me explore (including those within the first 5 minutes), then feel free to email me, or drop them in the comments below!

Until then — thanks again for your support.

- Yannick


~~~

METHODOLOGY

In which I pretend I know what I’m doing here, but you don't need to read any of it unless data ruffles your jimmies

I mentioned that there are multiple ways of looking at this data. This centers almost entirely on the existence of returning Users. For example, if we simply look at every visit to the Telelibrary (or rather, the 902 sessions out of 1000 in which a System name was recorded) we see a frequency chart which looks like this:

Because we aren’t controlling for returning Users, and since the most recurring Users have attended the Telelibrary as many as 17 times, they can easily generate a trend almost single-handedly (Ewe and Computer appear as outliers here, but they aren’t alone. Taken all together, the top 5 most frequently attending Users kept their System name consistent 76.33% of the time, and these 5 Users alone represent 6% of all calls).

There are two ways I found to control for the influence of returning Users, so that their decisions were not overcounted. The first was to look at the “Current” User data - which is to say, every User ID was analyzed by their most recent visit. I keep this current for many reasons, including the personalized notifications Users occasionally receive.

You’ll notice in 739 “active” Users, there were only 382 Unique names. I find this to be the best dataset for “most popular” name, because it essentially allows returning Users to settle in on their preference over time.

However, the next way to look at this data was to remove all but the first call from every User - you could call this a “first impressions” analysis. That looks much more like this

… which is to say, almost entirely identical. There are some slightly more pronounced differences as you move further through the list, but it shouldn't surprise me too much that the difference is slight. Only 110 of the 739 initial Users of the piece have returned for a second session.  How do these returning Users differ from one-time callers? Does their behavior change gradually, with familiarity, or are they different from the moment they call? In what ways does the Telelibrary change for their 2nd call? Or their 5th? Or their 17th? Great questions for another time.

A final note: you may have noticed that none of these datasets track 1000 names for 1000 calls - even when I said every visit in the initial 1000 calls, I still only tracked 902. This is because the opening 5 minutes changed drastically over the course of the first month and a half of sessions, as did my documentation. In brief - I didn’t always offer a chance to name the System, and even when I did, I didn’t originally track that data.  

Creators—myself included—it's so easy to start tracking data. And nearly impossible to get at the data you never collected.