Jason Danforth

cautiously radical, collaboratively experimental designer

[Jason Danforth @Json_danforth]

Nothing like a vacation to finally get some work done. #drawingmachine #creativecoding #threejs

[Jason Danforth @Json_danforth]

Every time I open After Effects I feel like a brave little Hobbit about to entire a dark, foreboding forest like "Bilbo, bruh, you don't have to go in there..."

[Jason Danforth @Json_danforth]

FYI, while I repeatedly smashed the refresh button on TurboVax this am, my mom called 1-877-VAX-4NYC and was able to talk to an actual human and get me an appointment. Godspeed friends.

[Jason Danforth @Json_danforth]

Just a regular old weekday morning of calling your friends to make sure they're ok

[Jason Danforth @Json_danforth]

It is beyond heartbreaking to hear the stories from my Asian friends describing their experiences, their fear, the vulnerability of knowing that their elderly parents could be attacked at work or on their way to the super market. In any U.S. city. At any time. Just devastating...

[Jason Danforth @Json_danforth]

@baersteve better than Christmas. absolute top of my list of desired features.

[Jason Danforth @Json_danforth]

MASSIVE!

[Steve Baer @baersteve]

Finally, CPython based grasshopper components. This was so much fun to put together.
https://t.co/Iclt43FQXR

[Jason Danforth @Json_danforth]

End the filibuster. Pass the For the People Act. Be a goddamn democracy.

[Jason Danforth @Json_danforth]

windows open

[Jason Danforth @Json_danforth]

can’t wait to never rent a car again

[Jason Danforth @Json_danforth]

@EricAnastas I think this is part of the reason for comments like these (which I don't necessarily disagree with, despite still wanting to focus on how these tools can apply to the design of, rather than the business of, architecture): https://t.co/uFbXVMBlZC

[George Valdes @georgevaldes]

Had such a great time interviewing @dr_danieldavis couple months back.

💯 Firms can leverage data outside of design ops.

[Jason Danforth @Json_danforth]

If I randomly sample 100 architects tweeting about ML and the future of design, what proportion should I expect to understand the difference between a parametric model and linear regression? Assume the sampling distribution is approximately normal and observations are independent

[Jason Danforth @Json_danforth]

@EricAnastas and there isn't a ton of discussion around how the later can be implemented in a design practice. both supervised AND unsupervised learning are a tricky fit for most computational design tools workflows

[Jason Danforth @Json_danforth]

@katdov technologists want their work to have impact, but it's seen as a power-struggle instead of an opportunity for collaboration.

[Jason Danforth @Json_danforth]

@katdov shiny tech makes for great marketing pitches, but shifting the core design culture away from sketches on trace by a principal (top-down) and towards data, hypothesis testing, and validation (bottom-up) is a huge ask.

[Jason Danforth @Json_danforth]

unexpected WFH perk: the neighborhood characters. top of the list is "running boy": a kid (late teens? 20?) I see at least 1x week, dressed normally (i.e. not exercising), ALWAYS running somewhere (no evidence that he's being chased by anything other than existential dread)

[Jason Danforth @Json_danforth]

@emptybox_design I have a 35% confidence level that we'll get there!

[Jason Danforth @Json_danforth]

Feel like it's pretty commonplace for architects to use "machine learning" as an ambiguous stand-in for "sophisticated algorithm", without considering how much of our work requires certainty ("the wall goes here"), not probability ("60% likelihood the wall goes here").

[Jason Danforth @Json_danforth]

Stats "humor" aside, there isn't much discussion of the difference between deterministic algos and predictive tools, or how the later will fit into design practice.