Don't glosso me over

It’s been a few weeks since I joined Aella’s new social media thingy, glosso. One part of the site, Ranko, quizzes you on the traits of your friends.

Currently, there are 86 adjectives. The more ratings you make, the more adjectives get unlocked for you to be rated on. Eventually, after clicking and clicking and clicking for a short eternity of friend-roasting and eye-glazing (or do I have that backwards?) you get to see a list of the traits that describe you best.

Here’s my top ten:

The averages reflect the rating scale; a +1.0 means “slightly agree” and a -2.5 means halway between “disagree” and “strongly disagree”. The column with the arrows tells if there’s a correlation between how long someone has known you, and how they rate you. Similarly for the gender column, but no such correlations are seen here.

This list is obviously correct. But wait: it’s sorted by raw score, so it doesn’t show how I compare to others.

Thankfully, we can also sort by percentile of glosso users:

Actually the top twelve now, because I’m sensitive and would be offended if you didn’t have proof of how smart and wise I am.

Dour and intense? Utter bullshit. A strongly-worded e-mail is on its way to those responsible.

After you make several hundred ratings, even more analyses are unlocked for your… enjoyment.

Dour power 🌸

I like getting feedback, so I can become better. But what do these results have to say to me? How much should I trust them?

Well, I could ask how Aella designed this thing – but she’s an expert in online surveys, and I am not. She has access to the source code, and I do not. So I won’t be doing a deep dive into methods, today. But one challenge comes to mind: differences in rating strategy. For example, when I’m asked to rate if someone is disgusting, do I disagree, because I’m not easily disgusted? Or do I base my rating on how disgusted I think some random person would be, if they’d seen and smelled what I have?

Aesthetic properties like “disgusting” live in the subject, not the object. “Tasty” does not belong to the cookie when there isn’t a cookie-lover to enjoy it. And a cookie-hater would be right to believe that many humans are cookie-lovers.

Will some adjectives be more of a Keynesian beauty contest than others? I don’t see any instructions on Ranko like “rate based on your personal taste, not on how you expect others to rate, even if you think they have much stronger opinions than you”. Or the opposite. So I expect a bit of inconsistency.

But onto a much more important issue: I disagree with some of the ratings of me! Consider:

• +0.84 offendable: How dare they slightly agree I’m offendable!

• +0.78 insecure: Or am I?

• +0.82 neurotic: AAAAAAA

When I first saw these scores I was confused. How could my friends misjudge my inner world like this? I’ve been pretty neurotic in the past – which is why it’s clear to me that I’m not, anymore.¹ And “offendable” is just… wrong?

“Insecure” isn’t far from the mark. Some of my social priors are weak. They’ve grown stronger over the years, but not as fast as the anxiety has faded.

There’s an inner path from uncertainty to unpleasantness that I rarely follow anymore. The two are not identical, but to an outside observer, it can be easy to mistake them – to lump together “insecure” and “neurotic” by not looking closely enough.

So, behaviourists: please step away from the adjectives.

1. There are contexts where negativity rears its head, like if I’m in caffeine or sugar withdrawal – but that’s not a generalized personality trait: it’s a mistake. As long as I maintain good habits, there is no trait.