This thread collects names in various formats. This data will be used to ensure that the Archivist can be as structurally agnostic as possible when parsing names.

You can help by contributing names, with as much contextual detail and annotation as possible (such as whether a portion is a patronym, whether the cultural norm is to have mononyms, etc.).

Please be mindful of people's privacy—names of public figures and historical figures preferred.

Probably just have to create a font variable in the CSS and add a lil Javascript. Hardest part will be finding a good font.

Boutta be like…

enthusiastically talking to wall

duck enthusiastically yet silently dancing

baby reading voraciously

a line of rubber ducks captioned “we need to talk about the ducks”

Can probably use Speechify and ideally have a toggle to choose between different Englishes for narrator. No need to present voices as gendered; have a selection page with previews and identifiers for user to evaluate and choose their preference.

Serif, slab serif, sans serif, Google Fonts.

Initial pages can be more decorative and include the full five-color palette:

Once the Archivist begins asking questions, the page design should be streamlined to just a background color and a contrasting text color to minimize distractions.

There should be at minimum a day mode palette and a night mode palette.

Files should output with chosen color scheme as accent color.

Some questions might be a bit more opaque in a story format and agnostic inquiry mode than people expect. They might not know how to answer something so open-ended.

There should be hints that are in a closed drawer that you can open to, basically, read documentation and intent for each question, as well as some common cases, presented in a way that makes it clear the examples aren’t exhaustive.

The default version should be semantically formatted so that screen readers can do TTS, and a human-read audio version should also be available for accessibility, ideally with a speed toggle setting that persists between questions.

To keep file size of the actual generator small, audio should not be embedded in the page but linked to and played on interaction. People who want to have an offline version should be able to download a package that includes all the audio files.