I see this as more of a cute historical artifact than anything. There was a time when models/harnesses behaved like this, but we are well past it for frontier (or not so frontier) models.
As someone raised as a light packer, I've come to the opposite conclusion a little: I'd prefer packing "medium" to have a lot of flexibility to be dressed well for whatever may come up. A small half-full duffel bag isn't that much better than a full one, so if I'm on a short trip I've found that I'll often appreciate having the extra options.
That being said, I literally do not own a roller bag, and am quite happy taking the stairs.
I think you say “as good as they can get with [the] hard manual work that goes into compiling them” as a way to respect the challenge of the BLS and say they’re doing the best with their resources, but it’s worth pointing out they could be better with more resources.
DOGE hurt them pretty hard: "The Bureau of Labor Statistics posted a notice on Wednesday stating that it stopped collecting data in three not-so-small cities (Lincoln, Nebraska; Buffalo, New York; and Provo, Utah) and increased “imputations” for certain items"
FWIW if you listen to Signals and Threads (the Jane Street podcast) they talk about this.
They have dedicated UX designers who do relatively little code but focus entirely on making the UX as optimal for the traders as possible and one of them was on an episode.
They talked about how most of the normal UX rules for SW go out the window because quants and traders have such different needs than "normal software".
This matters. Beyond performance (a universal good), the definition of "good" UX is strictly dependent on the user. Fintech platforms for traders who open a multi-window thick client application(s) and spend hours in it every day have little in common with consumer mobile apps designed to hold your attention for maybe a minute. It's mind-bogglingly lazy to choose something like MUI (Material UI) as if it were some kind of reasonable foundation irrespective of context.
Is there any professional software that wouldn't benefit from high information density? Most software design is optimized for the non-users who have to approve it.
You can have high information density and still not have words running into each-other. Taking all of the space out from around text and information graphics makes them less legible, even for experienced "power users".
I think the goal here is more an aesthetic of being for "serious traders only" rather than a UI optimized for speed and clarity. The UI designers probably have little choice but to match the preferences of the user group though, even when those preferences are empirically detrimental.
Yeah safety critical software needs to avoid high UI density for the interface itself. For diagnostics and monitoring or general stats you can still do high density but the actual "action space" needs to be kept as low density as possible and preferably with physical separation in the UI (i.e. via physical buttons and displays instead of just a big touch screen or monitor+Mouse and kbd).
Higher information density is ideal but when people are working in less than ideal conditions (stress, lack of sleep, alarms going off, etc) you need to prioritise information clarity and clarity of actions over raw density.
"non-users" are also big targets for software. If you run a software business, chances are you want people who've never used your software to start using your software. In that case, discoverability trumps high information density.
It only adds discoverability for the things that fit the screen. In other words, you have to make it simple. I.e. the lazy way to make something "user-friendly", do as little as possible and force the user to adapt to your ways instead.
It has to be the right information, filtered and presented in such a way that supports the task. Any kind of human in the loop UI has to manage user cognitive loading. Task saturation is real.
this - most of the internal software at my job that's actually important is very dense, with keyboard navigation and shortcuts. The ones that somebody makes to sell themselves to upper management while angling for a promotion look very pretty, and never get used.
And yet, we make such beautiful and vacant software ourselves for our clients, because again, its b2b, so you aren't selling to the people who actually have to use the software. Im glad at least we arent forced that often to use specific internal tools, the better ones just get adopted organically
This makes me think about the bizarre UI regressions in macOS in the last few years. I’ve got this giant, beautiful display and somehow it becomes less rather than more useful over time due to the software it’s presenting.
I'm on the verge of swapping my macbook for a framework laptop. I never really bought into the whole apple ecosystem (I've got a pixel phone), and I find myself using the mac as a less flexible unix box. I'm mostly in the terminal, vscode, or firefox, and using rectangle and homebrew to make it more linux-y to interact with. I'd swapped to the mac because it was a more reliable hardware setup than any linux laptops at the time, but I'm starting to look for other options.
i think wayland set linux desktop back 10 years. it’s finally coming up to parity with X11. but basic stuff like “get the latest release of of consumer GUI app $foo” is still weirdly mired in odd political plays between flat pack people, distros, and the actual devs. at least audio finally seems to be solved with pipe wire!
Yeah, that was the main attraction of a mac for me. It Just Works, but enough stuff is packaged for it that it can drive like a linux box for a lot of coding tasks. I'm not going back to windows anytime soon, either.
I remember getting in to all kinds of driver rabbit holes every other time I tried to run desktop linux, but a bunch of people I was talking to on here a few weeks ago were saying that's all fixed now.
A lot is said in this thread about user preferences for info density, but it is really a lot more than preference. there's also a big component of it that is user productivity focused. Info-dense UIs simply require a lot less scrolling around and pagination to look at relevant data. For example the ability to look at multiple charts side by side instead of having to scroll one chart after another into position enables users to spot differences that they wouldn't otherwise see.
In the year of our Lord 2026 when you can be fired for not churning enough tokens, you can believe any corpo-post corpo speak has been checked by their LLMs for "compassionate voicing" and "low risk wording"
Distinguished version: write the outline of the slide deck for the talk you plan to give at conferences about it, without having shipped anything or even written code yet.
These takes Arne necessarily incompatible. It can be a great tool for helping you do this kind of big picture design, but still need you as a guide and taste-maker to get to a good end result.
It’s worrisome, but it’s also worth pointing out that no one else has succeeded at it, or even gotten close. It’s easy to forget that there are not good well-integrated person AI solutions yet, it’s just chat bots. I think it’s just harder than a lot of people think it is.
but Apple isn't known to make grand promises and then not keep them, is it..? usually they just deliver what they say they will
yet i've been reading about "well they promised AI Siri two years ago and Siri still can't set an alarm right" in every thread even remotely related to the topic
i don't remember reading this much about anything else. it seems to have soured people quite a bit, at least in my internet bubble
reply