throwaway219450

Ah this is also a fun default feature of Powershell, except it works on mouse click.

I think it's more about inertia than anything else. I know the dishwasher will take less than 5 minutes to unload, but I put it off anyway. Problem is the AI can take action immediately, and that's very alluring. I'm sure there's research on exactly where that breakpoint lies. My current goal is to try and spend 15-30 minutes attempting to solve a task when I feel myself reaching for the AI.

One of three things tends to happen. Often it's something that is resolved in a matter of minutes, in which case it was laziness. Or, it's something that will take more time, say up to an hour, but the act of starting has conferred a sense of ownership that I don't want to give up to an LLM. Else, it looks like a much more complex problem and worth re-assessing, and potentially brainstorming with a model.

There are several keyboard shortcuts that you would never find out from watching an agent, but are mindblowing to new Linux users. I suck at remembering vim keybindings, but I have ingrained ctrl+a ctrl+e for jumping to the start/end of a string (which also works all over OS X). I had been a developer for an embarassing amount of time before I discovered those.

There are other terminal specific shortcuts for removing last word (ctrl+w) but they don't seem to be as portable.

I use history | grep quite a lot, because of the non-fuzziness with default ctrl+r. If I know exactly what the substring is, then that's great, but sometimes I also want to see the sequence around the command too (e.g. two step processes like eval ssh agent + ssh add key)

I would say 300 is more than slightly longer. Varies by institution and style, but 100k words, 4-5 collated papers, 100-150 pages maximum are quite common.

I would be concerned as an examiner if this came across my desk, more so if I read the colophon where the author comments that they intended to write their own typesetting system, before reading Knuth and wisely concluding that they were unlikely to do anything better than TeX. Top tier yak shaving there.

Always loved Iain M Banks' Culture series and it's very sad that he's not with us any more. If I could pick any fictional civ to live in, it'd be up there.

The current discussion around alignment seems quite relevant. Culture Minds are so intellectually superior to the humanoid population that it's not even a debate, but their goal appears to be a content and healthy civilization with a strong focus on individualism. Somewhere I think Banks implies that Minds see it as a sort of badge of honor that their citizens like them. They certainly have no need for subservience.

Source interview: https://www.youtube.com/watch?v=TTppvBU2rU4

I would suggest watching the Animatrix - particularly The Second Renaissance (part 1 & 2) which tells the story of how humanity ended up at war with the machines. It's only about 20 minutes total, but it's really well done.

ALDI/Lidl copy every big brand under the sun. They've been litigated against, but they get away with it way more often than they lose. Many cases fail to prove that consumers were misled, and (I find this incredible) one of the strongest defenses is that store has a strong reputation for brand imitation, so there is in fact no confusion. People go to Lidl to buy Neos, knowing full well they're not Oreos. It's also surprising that other supermarkets get away with it when they sell the copycat product next to the original.

If you could afford lawyers, you could definitely get away with that game, but it's easier for IKEA to stomp you.

If you were a superintelligence why would you farm humans and waste resources on all the excess... material... that isn't required for thought? And unless the machine's goal is to specifically abuse human consciousness, why wouldn't they bio-engineer their own grey matter?

There are other sci fi stories which use humans for distributed computing and they don’t realize, but I don’t want to spoil by naming as it’s something of a revelation.

(I do realize the Matrix is a fantastic movie, but entertain the thought? At any rate, the idea that robots only run on solar and wouldn’t just use nuclear is far more stupid.)

I consider myself a decent amateur baker and I've never noticed a huge difference in bread flours. I buy from a local mill because they sell interesting grains and I want to support them. Otherwise I get King Arthur, or Bob's Red Mill if I need something a bit more specific. KA is particularly good because their recipes are usually reliable and they're designed around their flours. The selection of baking ingredients is usually much wider in the average US supermarket compared to the average European one in my experience.

Some thrift stores will take them, if you have a good one nearby (I’ve been lucky here, but they’re rare). You can also put a note on your local freebie groups and someone might bite. In some cities you can leave them bagged outside your house with a note and they’ll be gone in hours. On some tech forum classifieds, people would give away cables and junk fans for the cost of the postage.

I thought the same thing, and then wondered if this must be an enterprise problem, because what home user adds loads of icons - so many that IO becomes a concern - and also wants them to be randomly assigned to other family members?

But then it occurred to me that XP was of an era when you would absolutely spend an afternoon downloading some megapack of 10000 cool user icons.

Bootstrap was/is popular for the same reason. One framework and your site looks decent on almost any device. The downside is a bit like LLM work in that it all ends up looking like a Bootstrap site unless you’re a good designer.

IANAL. In states with multi or all-party consent, there is usually some wording like "reasonable expectation" of privacy, which you don't have in public by default. The store could argue you have no expectation of privacy there, so they can record audio. There might be other legal codes, but I think this is how smart doorbell manufacturers are allowed to operate in those states.

> (c) For the purposes of this section, “confidential communication” means any communication carried on in circumstances as may reasonably indicate that any party to the communication desires it to be confined to the parties thereto, but excludes a communication made in a public gathering or in any legislative, judicial, executive, or administrative proceeding open to the public, or in any other circumstance in which the parties to the communication may reasonably expect that the communication may be overheard or recorded.

Penal Code § 632(c)

> Today, legitimate businesses are very careful about how they use bulk e-mail

I don’t think this is remotely true. Sure, they’re legally obliged to let you unsubscribe, and sure, it’s not dick pills, but every US company will immediately send you a newsletter when you purchase something, review requests and, if they/you use Shop for checkout, expect an abandoned cart reminder.

PR pieces and software companies don’t write tutorials to be helpful, they are advertising to you. If the LLM can do it for cheap, they really don’t care.

Do you mean 375 F? 60/40 is liquid at that temp (374 F). 375 C is way overkill for lead unless your iron is terrible.

My advice is to try letting the agent fill in the gaps. You can probably architect better than it can. Write your class outlines, explicitly define the public facing bits and what you want APIs to look like. Write the key integration tests that you know ought to pass.

The real advantage is that agents routinely write code without any silly copy/paste mistakes like accidentally accessing x twice on a coordinate operation instead of x and y. You can add some comments for what the function should do, throw in some real/pseudo code and let the LLM figure it out.

You can express a circuit as a graph, and many schematic formats are plaintext. Same goes for the Gerbers which are an ASCII format used to describe the masks that are used to define the PCB traces. The models are trained on a lot of academic information about how circuits work, most component datasheets are public and they've sucked up all sorts of niche greybeard advice from internet forums.

A huge advantage of electrical design is that the connectivity is testable with Design and Electrical Rule Checks (DRC/ERC). I suspect you could even tell the models to run physics checks on the traces that are important for things like crosstalk.

There's some cool research that looks at how strongly the weights are aligned through training vs adherence to the system prompt. Like when you know a model is lying through censorship: https://arxiv.org/html/2603.05494v2

Presumably if negative guidance is in the system prompt, there's a good chance that the model would happily comply if it wasn't there.

For stuff like this, definitely. It’s already exciting. Claude et al are surprisingly competent with real hardware. I’ve also had good success debugging basic issues with micros - not via the debugger, but giving datasheet and written analysis. I’m sure it could do things like capture from a connected oscilloscope (Salaeae have an MCP integration for their logic tools). Also for SPICE which I never bother with because the tools always felt stuck in the 90s, but having some serious automated analog review would be nice. I’m really excited to play with peripherals like the RP2040 or Beaglebone real time logic blocks, or improving/adding real USB support instead of USB serial.

For circuits I would trust a netlist or DRC/ERC review, checking a design against reference circuit values, probably not placement as that tends to be guided straight from a datasheet where it makes a difference and I like the artistic side of it.

The most interesting open source example of this I can think of is Ben Krasnow's (and friends') temperature reporting magnetic stir bar:

https://www.youtube.com/watch?v=nh1SIfipQs4

The temp sensor and transceiver PCB are cast into urethane and the coil gets installed into the hotplate. In theory lets you do all kinds of interesting stuff like monitoring when you can't get a wired probe into the vessel, or to drive a control loop that actually monitors the liquid and not the plate temperature.

You can stream Big Picture with tools like Sunshine/Moonlight or their forks. Works amazingly well over a LAN if you have a server somewhere and stream to something connected to the TV. I have a headless server and an NVIDIA GPU so not sure what the sibling comment is referring to about compatibility.

Easy enough, since this is how celestial navigation works, especially as we have very accurate clocks. You can get latitude from image declination and longitude from time, provided your clock is set from a good reference. In theory you only need one image pointed straight up to estimate declination. You’ll probably have an uncertainty on the order of miles.

I want to believe this, but every time I try Gemini coding assistance within Colab it's utterly dire. Code gen in a cell is OK, but things fall apart when you try to get into a feedback loop. The system prompt/harness fails to inform the agent about what it can and can't do, or does and doesn't have access to. It will confidently tell you it's done a thing, and then you ask, it admits can't actually do that but will happily try and fail again. Very frustrating, because I really like Colab as a platform for little reproducible experiments that may or may not require CUDA.

It’s curious that this doesn’t exist really. All the components are there, location via GPS, the tracking mechanics are not particularly difficult once you know the orientation, and there are good open plate solvers.

Maybe good sensors suddenly became cheaper? Willingness to put a real computer in the control system? For a long time even potato quality CCDs were expensive and handset electronics have been awful.

There is https://observatorycontrolsystem.github.io/ on the scheduler side

> explicit quality gates, content-addressed provenance, frozen empirical contracts, synthetic qualification

A fellow Opus 5 user I see

It depends very much on the furniture, I think citation needed on exactly which items you’re talking about. Name and shame!

The big wardrobes? They are the worst combination of impossibly heavy, awkward to disassemble and expensive to repurchase - best bet is to sell to the next occupant. On the other hand I have a solid table and chairs and they flat pack and reassemble perfectly. Same with bed frames. Shelving units go straight in the van. I wish the office desktops had metal threaded inserts for the legs instead of nylon, but otherwise also have moved several times and no real issue.

> For those of you who look down on IKEA furniture, I'd like to hear about what alternatives you think need mentioning. Especially for young people or people of modest means.

I look in antique/thrift/charity/freecycle/secondhand stores or groups. Where I live you can pick up a solid hardwood dresser for a few hundred bucks, because as you say - nobody wants their parents’ heirlooms. And lots of things can be had for free if you can hire a van to collect.

I still use chat copy/paste for this. Can't argue too much when the results work. The tried and tested alternative for niche Linux issues is browsing the internet and checking different options until something works anyway.

I do recommend checking methods against another provider. Occasionally Claude will suggest X, and Sol is like WTF are you insane? Clear often and update the prompt with your latest findings, because there are so many opportunities for the agent to find red herrings that they'll fixate on.

As with all LLM work, if you know some sysadmin, the process goes a lot smoother because you can say specifically "Write me a systemd unit file that does X" instead of a vague "I want to automate this, how do I do it" and not have a gut feel for which approaches are jank.