I wondered that, too—seems like retail shops have historically used brighter lights (halogen prior to LED?). That is, they're actually meant for use during daytime, to reinforce the existing outdoor light in large spaces.
The options do seem dizzying! I wonder: why are there more options for LEDs than there were for CFLs? Why is it more likely to confuse commercial buyers?
CFL -> LED transition? AIUI, LEDs are naturally cool and need an extra phosphor to warm them up. But it's not like warmer LEDs are way more expensive.
Trying to form a theory for the widespread rise of daylight-blue and mismatched color-temp bulbs in commercial lodging.
Feels like more than half the time I travel now, even at nice hotels, I flip a switch, yelp at the light, turn it back off again. That was less common 5+ years ago. Why?
HCI folkS: I'll be in Barcelona this Thu-Sun for CHI. Ping me if you'd like to hang!
I think an accidental tyranny of programmers has held back invention in user interfaces. I want weird, courageous, idiosyncratic, alien.
I share early reports of how coding agents are changing that, and what it might mean for people and institutions inventing interfaces today.
But malleability isn't enough. Novel interfaces take imaginative design, deep domain insight, and fluent programming. Few have all three skills, but programming can produce working software on its own—so interface invention culture is dominated by programmers.
I extend those architectural ideas to a sketch for a new system: a malleable reading environment, designed for idiosyncratic expert use.
In the demo, I show how one system can freely combine ideas from half a dozen isolated augmented reading HCI papers.
I demo this with two intersecting Obsidian WYSIWYG editor plugins for reading: one which transforms spoken text references to book deep links; and another which presents audio transcripts as playable, editable text blocks.
Both created without direct code supervision.
To get composably extensible complex UIs, you need very complex plugin API surfaces. Historically, that's meant most plugins are simple; complex plugins have been behind too high a barrier. But now coding agents can pay the costs of boilerplate and formal ceremony.
Agents can make one-off apps, but they’re usually toys, and they're still silos. If you want to change Photoshop's UI, you're not going to ask Claude to rewrite it, but with your new feature. I show how we might bring malleability to deep interfaces that experts live in all day.
⭐ New talk! Watch / read / visually skim here: andymatuschak.org/tat
Coding agents might help us finally break out of two cages: the app model, which traps computing in one-size-fits-all silos; and programming as a specialization, which has crowded out cultures of imagination and domain insight.
Something key about the "liveness" of knowing the other is there, and only being able to interact when they are…
TODO develop new political movement "against inboxes"
Underrated "feature" of AIM / IRC: you could only send people messages if they were online!
Sort of built-in backpressure. If you're busy, you won't sign on much, so you won't end up with a "big unread count". Feels so quaint now, like a little village. Say hi to your friends during passeggiata…
Yes, I think it's mostly focused on the craft and polish of the visual elements of interfaces. I don't expect that to change too much—like you, I'd like a great text on the deeper layers of the design stack. About Face focuses on a lot of the right questions, but too abstractly, I think.
Interesting to imagine structuring those crits as Figma artboard sequences, so the student could try each step themselves before revealing the author's take. (in the style of buff.ly/i7jdsOm)
I've often complained that interface design lacks great textbooks—still mostly taught through apprenticeship and trial/error. Interface Craft is shaping up to be a great resource, particularly for polish and craft. interfacecraft.dev
The step-by-step UI crit/redesigns are a really nice format.
Crepes are 4:1:2 milk:eggs:flour so they'd be in the bottom right corner I think
Oh, gosh, sorry you're dealing with that. DM me your Patreon account email address and I can fix it, I think?
Yes, yes, outstanding buff.ly/8gbKvke
Skimming through the figures, I saw… "IHOP Transgression". Wh—what is IHOP Transgression? Apparently their omelettes (and burritos??) contain pancake batter!
As for the dark abyss: wouldn't popovers go there? (1:1:1) Or choux pastry, at 2:2:1 milk:eggs:flour?
When I'm using coding agents and LLMs extensively in some tasks, I feel way more resistance to grappling with difficult, uneven problems in others. As if I get recalibrated to expect ease, effortlessness, velocity in everything. Not good!
Curious how you deal with this, if you feel it too.
Thank you!
Boston-area mutuals: I'll be giving the MIT HCI seminar this Tuesday at 4PM. I've got some open time on Wednesday morning—DM me if you'd like to chat! 👋
Every night this house is full-on blue and red inside. I like to imagine it’s like a cold-plunge sauna, where the occupants alternate between the rooms as needed to dial in their melatonin
Prototypers: do you still see a role for semi-high-fidelity modalities like Origami? Or just straight from sketchbook/mocks to vibecode?
Curious: why avoid? Is it just that "working in public" creates overhead, since it means doing more work along the way to make stuff legible to viewers without context? Or more that "saving it up" produces a stronger emotional impact?
Ah, I'm thinking not so much about the design of game UIs (though that exists) but more about the design of games in general, which requires both imaginative design work and also technical skill to experiment and iterate.
Right! Hm. I guess I'm thinking more about new representations: the non-linear timeline editor, constructive geometry in CAD, computational notebooks, bidirectionally editable audio transcripts, color grading…
Shouldn’t this be a problem for all design sciences, like architecture?
