
Hypermedia, JSON Schema, HTTP, accessibility, security. Just an @awwright, didn't you hear I come in six packs?
Arizona, USA
Joined September 2018
- Tweets2.3K
- Following64
- Followers94
- Likes173
Austin_Wright retweeted
Replying to @awwscript @tmikov
If an API has particular requirements, it should use HTTP features, or be described in the ToS. Saying "you must use this particular software" is a roundabout way of saying "We have opaque terms of service, changed at any time, without notice or consent."
Category-theory foundational Functional Programming is actually unrelated to state or pure functions; variables and mutation are perfectly compatible with it.
Functional Programming means describing your data & I/O so well that it only represents one possible program.
I wish more large open source projects would have a single point of support and an easy way to pay for a support contract... I just want someone to listen to my problems and at least pretend they're important, instead of evaluating 20 support channels for which is the easiest.
Austin_Wright retweeted
Would papers get rejected if they used legal code with variable names instead of Greek laden pseudo code?
"Don't follow the machine-readable (i.e. HTML) instructions, follow the English prose"
I don't like these gimmicks
Yoooo I just discovered an ancient macOS UI, if you go to Reminders -> select a reminder item -> ⓘ -> images -> Add Image...
Austin_Wright retweeted
to commemorate alan dye moving from apple to meta, here's one of his best quotes
It would look similar to a shell script. All commands would be variables that could be redefined, and would implicitly create targets, that in turn depend on the files named in the invocation. (It would know which commands consume and create which files.)
As usual Apple started a bad trend so I'm announcing a new versioning system: vibe.breaking.patch
- Increment vibe when marketing team says to
- Increment breaking whenever you remove a guarantee
- Increment patch whenever you add a feature or fix a bug
- No resetting to 0
I have yet to see a good reason for a tokenizer that's not solved with an improved parser. Unless your language isn't context-free for some inexplicable reason, and even then that's usually solvable with only minor adaptations.
Don't make me tap the sign...
┌─────────┐
│ precision ≠ accuracy │
└──┰──────┘
(\__/) ┃
(•ㅅ•) ┃
/ づ






















