Me screaming loud
(This comic is a mild criticism of it, it’s not an endorsement, but even this elicits in me a) AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
ADDENDUM: the worth bit re: pronoun fields above doesn’t mean they’re worthless, it just means that limiting it to just be that signal has exhausted their usefulness.
Add a pronoun field if it’s all you can do.
But it certainly beats the empty promise of a fucking pronoun field.
Is this perfect? No. It requires a full lexicon and knowledge of the specific language to work the way it’s advertised. And there are some easy concessions in the design for ease of first implementation (e.g., one pronoun). But know I do have design sketches for _all_ language bits that I’d love to support as this thing matures (including, say, using no pronouns, or supporting complex case agreement setups, or split agreement between two parts of a sentence).
I think this approach is a million times better than the pronoun field alternative. The user deserves to control their experience; a computer can now refer to them correctly, and model that experience for other people; and the user can select the way they want to be referred to with a set of properties for which the UI can make good inferences for cross-language preferences (you can select .feminine gramamtical gender and an English neopronoun, and if it’s not English it’ll fall back.)
The user preferences type isn’t a modeling of gender (gods know that there’s no good reason to model a person’s actual gender in a system), but also isn’t just a modeling of pronouns or grammatical gender. It’s a bag of declarative grammatical (morphological) properties.
It’s the Morphology type: https://developer.apple.com/documentation/foundation/morphology
Remember that Apple SDK types can expand over time — none of the axes it represents are `@frozen`; they’re merely the closure of properties from all supported languages.
The approach I’ve shipped last Monday is four years of discarded designs and iteration. It lets a localizer write a template string, and then insert user preferences into the process.
You, as a localizer or developer, can write:
"Do you want to call ^[them](…) back?”
and indicate in the … that it needs to be customized, and optionally for who. And then have the OS inject user preferences into the process.
• As a social signal that needs to be voluntarily seeked, they’re worth little.
• Computers cannot read them anyway. Computers talking to or about a person do not understand the content of pronoun fields.
• Pronouns are an anglocentric approach that does not localize — people in other languages want the way they are referred to to match, and sometimes this means matching grammatical gender and number, not just pronoun. Sometimes it means to do more.
Cabal, the open-source & p2p group chat I've worked on with friends, is in a quadratic funding pilot experiment run by Open Collective!
What that means for us, as an example: for every 10 USD donated, we get a total of 83 USD
boosts appreciated 🖤
the beat is a joke but really these aren’t separate things
and also gay
“After thinking reality was centerless and fundamentally unknown and alien, yet us aliens along with it, we tried to build new organizing principles, but they fell apart again and all we have left are the compasses we built to go explore the double-ruins of the real.”
Switchy hypnokink top, adjustably cogitohazardous, dorky Domme, senior engineer, loves Her own 💓. RESIST. (🔞)
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!