Wait until you get into developing your own PCBs and discover the BK7238 for $0.81 :)
1,245 karma · joined June 16, 2011
Whoa, you weren't kidding.
https://techcrunch.com/2026/08/10/tech-industry-is-buzzing-a...
Edit: looks like it happened a few months ago:
> The OpenClaw owner, Andrew Bird, published a now-deleted blog post about it on his company’s website on April 10, according to a copy still visible on the Internet Archive.
When they release new versions of Sonnet, no-one expects them to be better than Opus.
That's the nice thing about CAN: every node is equal, and every node can send a message whenever it wants. As far as I know, LIN doesn't do that: you have one node on a given LIN network that initiates all communication, much like I2C, and the most other nodes can do is reply.
It's fine for what it is, of course, it's just missing some things that I think make CAN really neat.
It's so much nicer than I2C in so many regards: much longer range, masterless (any node can talk directly to any other node), packet prioritization built-in, deliberately designed to be more or less impossible to wedge, possible to do IC-to-IC on a board without transceivers by using a shared open-drain bus (like I2C, but with only one wire!)... and plenty more.
Really the only downside to CAN is that the spec requires timing that's tight enough that everything has to have a crystal on board. Ah, to have an alternate reality where CAN's timing requirements were relaxed enough for that not to be a requirement and it had caught on as the IC-to-IC interconnect standard instead of I2C...
This, to me, is the more interesting bit of the article.
I don't really know what a good solution looks like, but yeah, that's annoying.
From what I could tell it looked like it was in JS code that had been JIT compiled. I haven't attempted to troubleshoot further beyond setting all my future Claude Code instances to pipe their stderr to disk so that I don't lose the stack trace next time.
Anyone in the room care to volunteer?
Source: I'm a card-carrying AES member.
You are correct that it's not solvable in a rigorous mathematical sense as stated.
Nothing to see here, move along...
I think you might have Reid Hoffman (founder of LinkedIn) mixed up with Reed Hastings (founder of Netflix). :)
I don't agree with that statement universally, but I have to say I do when it comes to this article. I came here hoping for substantive discussion from those who'd had a chance to try it out; instead what I got was a seemingly endless stream of venting. There's a place for venting - and plenty to vent about with the state of AI nowadays - but to borrow from the HN guidelines you linked, it does very little to gratify my personal intellectual curiosity.
As someone once said: any fool can do for $2 what it takes an engineer to do for $1.
The point that this announcement is trying to make is, of course, that AI has already made that particular signal approximately worthless for that purpose.
Calling over the cellular network has been prohibited since time immemorial. What the parent comment is talking about is carriers also prohibiting making calls over airplane-supplied WiFi.
You can't, for example, join a Zoom meeting, or use your phone's built-in WiFi calling ability, on a typical flight nowadays, for better or for worse.
They do, in spades: https://www.amazon.com/3FT-Micro-Data-Charge-Cable/dp/B0DDWH...
I look forward to the day when they're no longer necessary.
https://github.com/YosysHQ/nextpnr
As someone actively working on nextpnr support for a fairly new FPGA architecture, it really is amazing that we have something like that in the open source world.
YosysHQ are one of my favorite companies to exist.
What this means it that you can set up an app like Torque[0] and add widgets that show you how fast each of the motors are spinning, live, and watch what happens when e.g. the engine starts: MG1 and MG2 both torque the engine forward, MG2 just enough to stop the car from attempting to roll backward in response to MG1's torque through the planetary gearset, and then MG1 spins up with the engine and then stops torquing it once the engine reaches idle.
Battery charging while idling is similar: MG1 turns itself into a generator, fighting the engine and generating electricity in the process. The throttle opens considerably, as if you'd pressed the accelerator halfway to the floor, but MG1 and the engine work together to keep the engine's RPM around ~1,200 so you'd never know it - it's as if you're driving up a really steep hill that stops you from accelerating even though you have the gas pressed halfway down. And then MG2 torques backward to stop the car from rolling forward any more than the Prius's normal "simulate a normal gas car's tendency to roll forward when the user lets their foot off the brake" would have it do.
It was fascinating to watch, and I kind of regret not building an app similar to the parent comment's link that showed what my car was doing in real time with the gears drawn out like that.
Agentic AI has its faults, but one thing I've found it to be very good at is surfacing the "unknown unknowns": things I didn't know I should have searched for but that are directly relevant to my problem.
Back when I worked at Google there was an internal page someone put up that denoted what they called "the YX problem": the observation that the XY problem, applied to a sufficiently great extent, creates an environment where more productivity is lost convincing one's interlocutor that X is in fact the correct problem to solve than would be lost by chasing X and having to later pivot to Y if that turned out to be wrong.
It's extraordinarily aggravating when it happens. I really wish it was something we talked about more.
There are many reasons one might prefer OrcaSlicer over Bambu Studio. One might be perfectly fine using Bambu's cloud services while preferring OrcaSlicer for different reasons; this is for those people.
Others might not want to use Bambu's cloud services at all; OrcaSlicer as it currently exists is fine for them.
In the 99% of cases where this status page is used, it is in fact the truth: we'd throw it up when we had e.g. data migrations to do as part of a rollout that wouldn't allow for our normal zero-downtime deploys.
So no, lies are not always quicker than the truth.
There's plenty to criticize Instructure about; let's not go reaching for straw men in the process.
That doesn't excuse any of their other messaging though.
Back when I worked for Instructure ~10 years ago, Canvas was effectively a single, giant, monolithic multitenant app with one instance backed by several thousand app servers and ~100 separate Postgres database clusters that any app server could talk to.
Schools were grouped onto pools of app severs and Postgres database clusters more or less according to locality and cluster availability. I want to say a handful of the largest schools got their own clusters, but I'm not certain, and at any rate their clusters could certainly all talk to each other.
It was actually kind of neat from a technical perspective: any Rails model across the entire Canvas world could have a "foreign key" pointing to any other Rails model anywhere else. Among other things, this allowed for users who could administer multiple Canvas organizations, even if those organizations resided on different Postgres clusters. https://github.com/instructure/switchman is their gem that made that all work. (I put "foreign key" in quotes because the whole thing was implemented in software, not with actual database FKs, for obvious reasons.)
---
Of course, the massive downside to that sort of thing is that if you manage to pop one Canvas app server, you have the keys to the kingdom. I wonder if they'll sharpen the edges between clusters in response to this...
---
(Disclaimer: I left Instructure back in 2017; much could have changed since then, and my memory could be faulty about the specifics. Caveat emptor.)
That's just the quickest page/status update to throw up; it was a one-liner to push it live back when I was on the deploy rotation.
I'd hazard a guess they have more important things to worry about right now than exact status page messaging ;)
As a hardware engineer among other things, that was one of the first things I learned about interfacing with USB C. How do so many consumer devices keep getting this wrong in the year of our lord 2026?