yodon

Not directly related, but does anyone know why McDonald's moved away from the "number 1", "number 2", "number 3" ordering style they developed, which significantly reduced the amount of time a typical customer takes to place their order (enabling more turns at the register/ordering screen without preventing customers from making non-numeric-based orders) and moved to their current in-store signage which feels almost explicitly designed to increase order time and customer cognitive load at the register? The main screens over the register offer the eyes no structural guidance on how to scan the contents, so every word and line has to be read individually. The clutter around the words makes recognition more difficult. The sequencing of content on and off of the screens means the desired information is frequently not visible at all and the customer is forced to wait for it to rotate back into visibility.

Homepage looks amazing

> sometimes I -mixing- a show and I couldn't tell you where a specific sound is coming from

And in those situations it won't work. Is any of this really a surprise?

Think about it conceptually:

Could you watch a music video and say "that's the snare drum, that's the lead singer, keyboard, bass, that's the truck that's making the engine noise, that's the crowd that's cheering, oh and that's a jackhammer in the background"? So can AI.

Could you point out who is lead guitar and who is rhythm guitar? So can AI.

This feels like a lot of rationalization for the purpose of excusing writing exactly the sort of code that Kernighan advised against.

Advising against writing complex code is not advising against learning.

The person who solves a hard problem correctly using simple code has generally spent more time learning than the person who solves it using complex code.

On a Mac, you can add sound to the notification:

osascript -e 'display notification "Ready" with title "Claude Code" sound name "Glass"'

(you may need to first use the OSX Script Editor to manually run display notification "Ready" with title "Claude Code" sound name "Glass" to get the Mac to show a permission request dialog for the notifications)

I found those two sets of threads very demoralizing when they happened.

I like to think HN is better than the level of conspiracy theory adoption and 4chan-like haha dude died let me make jokes about it posts in those pages.

Streisand effect?

A mod changed the title to something other than the originally submitted original article title, to protect a major VC.

Not cool.

See also Spaitial[0] which announced today full 3D environment generation from a single image

[0]https://www.spaitial.ai/

> photorealistic 3D representation from a single photograph in less than a second

Anyone know of an easy way to get Claude Code CLI to notify you in slack when it's done thinking for long periods?

So tired of forgetting to repeatedly look for a response that could take anywhere from a few seconds to an hour to show up.

> Comments saying "this feels like AI" should be banned.

Strong agree.

If you can make an actually reliable AI detector, stop wasting time posting comments on forums and just monetize it to make yourself rich.

If you can't, accept that you can't, and stop wasting everyone else's time with your unvalidated guesses about whether something is AI or not.

The least valuable lowest signal comments are "this feels like AI." Worse, they never raise the quality of the discussion about the article.

It's "does anyone else hate those scroll bars" and "this site shouldn't require JavaScript" for a new generation.

> Wish the stereo images were cross-eyed mode.

Me too, but mostly because it would mean we had a way to do interstellar travel. I'm pretty sure you'd need a very long baseline between the images for the angles to work properly

Is there a link to the SRE book?

>Gradual deployments are a more reliable defense against bugs than careful programming

The challenge, as I understand it, is that the feature in question had an explicit requirement of fast, wide deployment because of the need to react in real time to changing external attacker behaviors.

Must have been an amazing effort to be involved in.

From the HN guidelines[0]:

>Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that".

[0]: https://news.ycombinator.com/newsguidelines.html

(2016)

Fast and easy to take action on that page.

Good list. I'd add Hedy Lamarr, Cliff Stoll, Steve Jackson of Steve Jackson Games, John Draper (Cap'n Crunch), and RTM (Robert T Morris), but I'm certainly not faulting the list in its current form.

Interesting! Does "write your tests first" have any parallel in Hegelian approaches?

I've looked without much success for good resources on training and preparing doe incident management (blog posts, conference talks, books, etc), and haven't found much. Are there resources you're aware of that talk through the process, team training, best practices, role allocations, internal and external communications, etc.? Post mortems are great to read, but they aren't the same as an actual pro-active training process.

Would you consider a lightweight app that just does notification?

Installing and running Whisper locally just to get notifications seems like more resource consumption than some of us are looking for.

Some will want one, some will want both.

This is likely a separate tool, but I suspect it is trivial to build given the framework you've set up...

What I most want is something that notices if Claude has been thinking for more than say 30 seconds, and if so sends me a slack message when the output is finished and ready to read.

Given that Claude is often thinking for 30 minutes, I context switch onto something else, and wish there were a way to get a notification so it isn't sitting idle waiting for me for ten minutes (or an hour, as sometimes happens)

I very much appreciate the politeness and care the people who responded to my comment are bringing to their responses, but the issue isn't the keyboard layouts, it's the mindset that somehow the apparently terrible UX of the QMK and ZMK software is not just acceptable but beneficial.

I've been using programmable columnar split keyboards with modifier keys for decades, and chording keyboards before that. None of those things are at all new, nor are they particularly difficult. What seems to have been added in recent years is this weird keyboard-hipster-macho mentality that seems to have overtaken the community.

If it takes more than a PhD (which I have) and decades of experience with programmable remapable keyboards (which I also have) to use your keyboard, you're doing it wrong. If, as a professional software product engineer (which realistically most users of this type of keyboard probably are) you can't see that, you're probably way too far down a weird keyboard-hipster well and would probably do well to pull out and spend some time refreshing yourself with a reminder of how empowering good UX actually is. Bad software UX isn't actually "power user" stuff, and pretending bad software UX is an actually good thing is simply denial of how bad one's software UX apparently is.

It seems highly ironic that a community focused on keyboard productivity would fall into this particular hipster macho mindset, but for whatever reason it seems to have taken hold like wildfire. More power to you all, I guess. Definitely keep up the polite and welcoming aspects of the community, and perhaps one day some branch of the community will wake up to the fact that keyboards are a UX affordance, that the keyboard community is deeply passionate about their user experience, and that a good software UX actually matters and would in fact be a good thing and not a bad thing.

I feel like there should be a sign on the home page saying "you have to be at least this (arrow pointing at a height) cool to buy this keyboard"

If you don't already know how this kind of keyboard works, we don't care about you and won't bother explaining it to you because you're obviously not worth selling to if you don't already know how a programmable 42 key keyboard works.

You have to pick keycaps, and switches, and maybe buy extra keycaps for some reason. We're not going to tell you why extra keycaps are important or useful, but you should probably buy them anyway for some reason.

I'm pretty sure they would have sold me at least one keyboard, maybe several, if they'd bothered to put even 5 minutes thought into non-keyboard-hipster customers, but I'm clearly not cool enough with my multiple kinesis keyboards, chording keyboards, and mechanical keyboards.

I'm not a keyboard hipster, I'm just a guy who had RSI and doesn't want it again. People like me do actually buy keyboards.

I get that it's fully programmable, but can someone explain how you type numbers and the symbols that are on number keys on this keyboard? I didn't see any keycaps for them, and couldn't find any docs on where the symbols live.

EDIT ADDED: I'm guessing maybe there is a control that causes other symbols to become visible on the keycaps, replacing the default A-Z symbols, and they never show those alternate symbols in the photos because we're supposed to know it does that.

The name "Hedley Lamarr" was, obviously, a play on Hedy Lamarr, who sued the filmmakers for using her name.

> For the layman: These plots show a sudden burst of acceleration on the object, and the push points mostly in one steady direction within the orbital plane. A normal comet warms up and ramps its outgassing gradually, and breaking apart would look chaotic with the direction of thrust changing randomly.

> Here we see one short, powerful jet, like a sealed pocket finally venting, so the speed jumps sharply while the path stays in the same plane; it accelerates but it surprisingly doesn’t change direction.

> It resembles what a single thruster would do (what an incredible prospect!), but it can technically still be explained by a natural rupture of volatile ices in an ancient space rock.

> This incident has been resolved.