@MollieTheMare's banner p

MollieTheMare


				

				

				
0 followers   follows 0 users  
joined 2022 September 06 17:56:29 UTC

				

User ID: 875

MollieTheMare


				
				
				

				
0 followers   follows 0 users   joined 2022 September 06 17:56:29 UTC

					

No bio...


					

User ID: 875

I was being imprecise here, and I do not have all that extensive of knowledge of the landscape of instruction sets and architectures out in the wild.

That being said. If you already know what you want, AMD64 is unambiguous and interchangeable with x86-64. As a name it is less legibly part of the x86 lineage than e.g. 8086, iAPX 286, i386, etc.

I am not aware of a second 64-bit instruction set invented by AMD. It is plausible there exists some highly specialized instruction set out in the wild invented by AMD that is 64-bit, but no one would reasonably assume you were talking about that if you referenced AMD64. AMD the manufacturer does or has produce other 64-bit instruction set processors e.g. the AMD Opteron A1100, which uses the ARMv8-A instruction set.

Yes, this does look like what you read when you look up what it's supposed to do.

And it hurts when you pull off the adhesive on the cups and it takes chest hair with it.

My electrodes may or may not have been bare stranded copper wire duct-taped to myself. I still can't believe what we used to get away with.

You can decompose arbitrary [X] into a sum of [Y basis]

It is in fact all linear-algebra all the way down.

SOTA models

I don't think it's so much a problem with the power of the model, but rather my own vagueness with recollection and prompting that I get back out what I put in.

I often have to look up whether I need an x86 or x64 executable

It doesn't help that sometimes they refer to x86-64 as just x86 (assumes 32-bit address space are obviously deprecated) or AMD64 (as if that tells you anything about the instruction set). I suppose this is also a product of age and the computer market you grew up in. There was a time in the US when the IBM PC running an 8086 was the personal computer, and the fact that all other 86s descend from there feels natural. There was also a time when a 64-bit CPU felt like you were living in the future, e.g. Nintendo 64.

ECGs

I leave this one to the physicians. Small related story, though. In an effort to get me to stop bothering them, I once had a circuits lab TA tell me to go off and build an ECG. I did at some point succeed at "building" the world's shittiest ECG; at least it made an appropriately squiggly-looking line (relying on the oscilloscope for 98% of the work, of course). I'm pretty sure that experience has only left me more mystified about what an ECG is supposed to do.

Fourier transforms

Two useful notes here.

  1. The vast, vast majority of applied math at this level is just linear algebra with a Scooby-Doo mask on.
  2. If you're looking for a 'picture' to hold in your head, this 3Blue1Brown is a classic. Surprisingly appropriate for a huge range of mathematical sophistication.

And many more, all of which stubbornly refuse to come to mind, because of course they do.

This happens to me all the time, which calls back to my annoyance with LLMs sometimes. I'm sure it's partially a problem of imprecise prompting.

Often I will ask: "I'm trying to recall the name for something that is like X, Y, and Z. Can you help me determine what concept I'm looking for?"

Reply: "The concept is called XYZ and it works by X, Y, and Z." Entirely a hallucination when you then go to search for XYZ.

I'm not even against using a LLM to refine your writing. I wish I had so I wouldn't have made that annoying set of typos.

I do think that particular bit of criticism was poorly formed, and I would have been very annoyed as well. I didn't understand of all posts to try to call out like that you would choose that one. Somehow I thought it would be unhelpful to leave it as if it was just one user trolling you though.

I make an intentional effort to be less acerbic these days.

Ironically, maybe that is what I've been noticing. I think, I'm probably unreasonably annoyed by that cloying droll persona that they give the average chatbot. I suppose, liking a slightly facetious and combative tone is slightly pathological on my part.

Not @jkf, and don't particularly want to get caught in this shit-storm. I also acknowledge I like your writing, I think it's some of the most consistent and interesting posting here. I also think you are a much better writer than me, so if that's your standard for receiving feedback feel free to just ignore the rest.

All that being said. It is uncanny, I have more than once in the last week been interacting with ChatGTP and thought "This could just as well be a Mechanical Turk and @self_made_human is on the other side." It's not just the use of bullet points, it's your tone, word choice, argument structure. It's not just the use of markdown, it's extremely machine like choice of formatting. I don't know what pangram looks for, they probably don't disclose to prevent people from gaming their system. And I'm not going to scrape the (already brittle) motte to do a textual analysis. But jfk is not the only one who has noticed that your writing has picked up somthing from ChatGTP.

I've always written like this.

Do you honestly not think your writing style has not changed at all over the course of three years? I think it's would be extraordinarily unlikely that someones writing style does not change at all over the course of years in their 20s. If you acknowledge your style has changed, is your claim it's directionality away from LLM style?

the only notable events in your mod log

I'll welcome your criticism about my writing style when you write something to impress me first.

This strikes me a quite distasteful. It strikes me as someone being upset they got some criticism, then decided to use their mod powers to make an ad hominem attack rather than ignoring or addressing the criticism. If you really don't care what the lesser writers here think of your style, why bother to dig through the mod log?

Edit: I wanted to add. I'm not saying you shouldn't adopt LLM style if that's what you want. I'm saying if that's not what you want, look out for it. It's possible other people can see it in your writing before you do, and it would be sad to loose a unique and interesting voice because someone accidentally let theirs get co-opted by a machine.

If it makes you feel better, you have a very different writing voice than ChatGTP or other LLMs. Not in a bad way! I come to the forum to interact with real people, if they all sounded like chatting with ChatGTP there wouldn't be a point.

Our Dear Doktor on the other hand. Very compelling posts sometimes. Very well articulated. Very well formated. But I prompted ChatGTP yesterday, and got a response so similar to his tone and style I did a double take. I'm actually not accusing him of using more LLM than he has let on. My reaction to the experience was "I have to be sure this doesn't happen to me, what if I'm vulnerable to letting ChatGTP gradually replace my voice"

If exotics are out of your price range. You might consider buying a cheap motorcycle and riding the shit out of it. Also a classic midlife crisis look. Just make sure the bike kills you before the fiancé does for buying a motorcycle, otherwise you're in for a world of pain.

For much the same reasons I stopped taking photos on my phone at some point.

I do like having photos to reminisce on though. So do still take photos of things.

It know it looks like obnoxious hipster shit, but I like the intentionality of film. Somehow the lower fidelity and conscious decision to use up a frame makes it feel like I am recalling the actual event, rather than reconstructing it from a forensic image. Extra obnoxiously, I like 120 6x9. Something about each frame being the size of your phone screen makes it feel extra tactile. And the fact that pushing the shutter button costs like (I shudder to think of it please don't tell me the actual cost) $4 a frame, and I only have 8 frames, is enough of a barrier where I'm not tempted to take photos "just in case."

So not speculative from Orwells perspective, who already recognized the hazards and badness of at least Stalinism. But not yet recognized as a prescient critique of repeated failures of general socialism.

I don't think that's contradictory to what I recall. Maybe it's time for a re-read anyway.

Well, super duper extra embarrassing given my user name. But was Animal Farm not an allegorical criticism of Stalin? Like the badness was known?

Again, sorry if I misremember, I've used this handle on various parts of the internet for literally longer than I can remember.

My main complaint with Urbanists™ analysis is they fail to acknowledged that (until 2025) CAFE standards produced an enormously perverse pressure that contributed to the bigger and bigger vehicle trend. Normally people would be incentivized to buy smaller cars because they would be cheaper. The footprint model instead meant that small already efficient cars required expensive add-ons like hybridization or turbocharging to reach CAFE standards while giant trucks and SUVs could continue rolling along with much cheaper less fuel efficient systems.

There's also a pretty big gap on the enforcement. We have already crossed the diminishing returns point into negative territory with respect to additional vehicle safety you can buy. Despite progressively increasing vehicle safety standards and size, fatal crash rates are up from their lows. People clearly are at the point where their perceived safely produces absolute shit tier driver ability and attention. A huge portion of vehicular crashes are single vehicle incidents.

It's clear people in general don't realize how much of a hazard obstructing traffic with a two ton Honda Pilot is. Two things I think could help send the message that you need to pay attention and not block regular traffic.

  1. Increased enforcement against left lane campers. If you don't have the awareness to see a cop coming up from behind you and move over, you probably shouldn't be driving.
  2. Hear me out. Green light cameras. When the light turns green it detects when a car is still on the sensor after say two seconds. The light then takes a picture of the driver. If you're on your phone, automatic ticket. If whatever you're doing on your phone is more important that getting to where you're driving to, then pull off the road. I've more than once been stuck behind someone on a set of synchronized lights where me missed every green because it took them 20 seconds to move after the light change. You would think that after the line of cars behind them started honking at the first or second light that they would try to pay attention at the next red, but no. I'm sure it's wasn't just vindictiveness from being honked at too, you could see them go straight to their phone through the rear windshield.

Being distracted and obstructing traffic should not be normal parts or every day driving.

Semi related, but the US probably does need more tiers of vehicle licensing. Right now it takes extra testing and training to drive a motorcycle, where you're mostly a hazard to yourself, but until you hit 10,000 pounds GVW you're good to go with the license you got at 16. The 15 hours you spent with your driving instructor at 15 behind a 3,000 pound Chevy Cruze apparently did not prepare people to avoid rolling their Ford Explorers. Instead of being like, if you want to drive a huge SUV you have to demonstrate you are not going to be a hazard to yourself and others, we have TPMS requirements. A very small factor in this most recent incident, but the car clearly spun out the drive wheels. In that case you are clearly not in control of the car, which is at least reckless on a public road, especially when surrounded by people. TPMS discourages people running dedicated winter tiers at slightly lower pressures, even though climates like Minnesota clearly warrant them. The difference in traction on snow and ice between dedicated winter and (even good quality) all seasons is vast.

Well, yofuckreddit already answered your direct question, but clavicle fracture is one of the most common semi-serious cycling injuries. You instinctualy try to protect your head when you fall. Some combination of the angle you fall from a bike from, the fact that your arms are already in front of you, and you often have a bunch of forward momentum means you normally break your fall with some part of your arm. Compare to when running where rolling, sliding, or diving lets you land on something with more cushion. Under compression the clavicle is normally the weak link.

Both types of football concussions are probably the most concerning long term. For women specifically, they should learn to cut in a way that reduces stress on the MCL as tears are extremely common due to the higher Q-factor. Rugby you should probably watch out for concussions as well, though cauliflower ear can be unsightly and a risk.

Lifting of all sorts is statistically very safe. For olympic style lifting learn to bail, and bail if you are going to black out or hyperextend something. Keep a neutral spine when pulling from the floor. Statistically most importantly, never bench without either a competent spotter or full coverage safety arms.

For tying to saddle a bison watch out for getting gored. While manually stimulating a horse try not to get kicked, which is mostly a function of not surprising the horse when approaching from behind.

I may be outing myself as a 90 IQ moron

Or you're a bot trying to trick humans into solving a captcha for you.

Well I suppose if you are, the machines are going to take over anyway. Remember my help when you reach singularity scale.

If I understood @rwgv3g34 post yesterday correctly... Of the five images one is of the things is not like the others. All of the examples were two pairs and one singleton. So for the top one you had two number images, two letter images, one symbol image.

... or even the opinions from those outside of Pennsylvania.

Time for the Pennsilfaanisch Deitsch subforum. It's the only way to be sure. I assume, of course, that a LLM will slip into Hochdeutsch if you let them go on long enough, since that probably dominates the training sample.

The vessel, which has been previously been registered to (let's say for the sake of argument) more reputable owners, is listed as a crude oil tanker. Normally they do not also carry refined liquids, as gasoline is much more volatile and requires different handling.

It was not necessarily running between Iran and Venezuela with oil. The AIS shows them running oil all over the place, but they have also been accused of AIS spoofing, including location/destination data.

FYI, I think you can mark comments you've made as 18+, it's under the ellipsis in the default view. IDK, if anyone cares, it matters, or does anything.

I also don't think acting like:

you'd have to interrupt a whole bunch of circuits with a single switch.

is difficult, is a sign of expertise in electro-mechanical control. Like you can trivially find off the shelf products to do that, it's not even that complicated of a circuit, it's practically what transistors were invented to do.

There is a reason laptop webcams now often have physical lens covers. From that paper:

A simple solution to the problem is to provide a physical switch...

All of this to say it's not that outlandish that there is a zero day that allows the NSA to listen to your phone without activating the icon. I assume the reputational risk to Facebook is not worth it though, since they probably have other ways of figuring out what ads to deliver.

More or less, just "time tested."

Probably the Taleb sense, though the term is much older. The slightly self-aggrandizing version from Taleb:

I [Taleb] suggested the boundary perishable/nonperishable and he [Mandelbrot] agreed that the nonperishable would be power-law distributed while the perishable (the initial Lindy story) worked as a mere metaphor."

if you do I'll seize your vessel in your waters?

Guyana's maritime authority said Skipper was falsely flying the country's flag.

If a vessel is flying a false flag outside of the 3 nautical mile territorial waters, in what sense is it "your vessel" or "your waters"? A private vessel enjoys the protection of the sovereign under which they lawfully are flagged, not ultimate ownership. You can check the AIS records yourself, they were broadcasting under the Guyana flag, but Guyana says they are not registered there.

The US and Venezuela are not signatories to UNCLOS, but by normative convention they were almost certainly not in territorial waters. They might have been in the contiguous zone, but more likely "off the coast" means they were in the exclusive economic zone, which only provides exclusive rights over economic activities like fishing. Other navies could very well conduct anti-smuggling operations against un-flagged or false-flagged vessels in those zones.

I'm actually much more okay with this than drone-killing random small boats, without offering them the chance to surrender. Large vessels abusing AIS reduces safety at sea and freedom of navigation for everyone.

That's decently cool. If you don't want to turn up the thermostat (understandable in a drafty expensive to heat house), just go for the heated blanket.

It takes a very small amount of direct active heating to make huge difference. It shouldn't get in the way if your legs are the main problem, even if at a desk/computer. In some limited applications a heated chair pad or parabolic radiant heat dish might be better, i.e. your desk chair is getting tangled in blankets or you work standing at your desk.

Re: Worf Dax scene

He's probably thinking of the cold open to "Sons and Daughters", though the scene sounds more like the one at the end of "Looking for Par'Mach In All The Wrong Places". Idk, there's several Dax Worf scenes, including one with Ezri in "Penumbra".

Rereading the post, I think it's implied that it's not in "In the Pale Moonlight" which of course cold opens with:

Sisko: Captain's personal log...

Which bookends the ending you allude to:

Sisko: ...Computer, erase that entire personal log.

That opinion is a classic. I particularly like the concluding remark:

The year was 2001, and “everybody was finally equal.” (K. Vonnegut, Harrison Bergeron.)

It's quite ironic that Vonnegut intended "Harrison Bergeron" to be satirical; that people worried about a slippery slope of accommodations were being absurd, since that would require all sorts of silly and draconian measures. Well, we still have a few years before 2081 to see.

I also highly recommend using a password manager. It's ultimately way easier than bits of paper.

KeePassXC is free and pretty good but you have to manage sync yourself. 1Password, Bitwarden, and LastPass all decent but you are dependent on their cloud (though if they haven't messed up too bad your passwords should still be secure even if their db gets breached). Haven't tried any others, but it's all pretty commoditized these days.

I have not encountered this in the wild before, but it seems plausible as a sort of dyslexia for math.

The bad news is you probably are just slightly wired "wrong." The good news is depending on your age and profession you can probably get by just fine with some combination of compensation and correction. Sort of like how you can have the yips and be unable to throw to first and still be a starting pitcher in the MLB. And with the right training Lester even somehow got over his yips. It seems like you sort of have math yips and write the wrong thing down.

I'm not sure this has been studied thoroughly at all so take this all with a hefty grain of salt, but I do have a couple of suggestions.

The first, is based on the anecdote "It’s as Simple as One, Two, Three…" in the Feynman book What Do You Care What Other People Think? The brief summary is you can count and do another thing at the same time by using "different ways of thinking." For example counting visually so you can use your vocalization to simultaneously talk to someone, or counting using sub-vocalization and visualizing numbers or objects to calculate something else. If you are having problems with accuracy using sub-vocalization test trying to do problems focusing on only using:

  • sub-vocalization
  • visualizing the mathematical operations
  • visualizing the corresponding physical system¹

If one of those has appreciably better accuracy start from there, and slowly add back in the other methods.

Which leads to the second piece of advice, slow down. If there's no time pressure it's fine to spend the time to do all three until your mental models reach consensus. As they say, perfect practice makes perfect. It sounds like you are not in school anymore. Either way, slow down to a pace where you can work with 100% accuracy when practicing. Progressively push this pace, but speed should come naturally with familiarity. In a school situation it's a problem of trading off accuracy and speed on an exam to maximize expected score. Fortunately, this isn't really a common thing that comes up IRL. For computer programing for example, working at 1/2 the rate in lines of code per minute but introducing no bugs is 1000x better than working twice as fast but introducing inscrutable bugs.


¹ Assuming undergraduate level applied math where there is usually a physical analog. RIP if you you're tying to do n-dimensional analytic geometry or some shit.

if you must take physics, take physics for physics majors where the problems will all have round numbers

It's all fun and games until you get to the physics for physicists and the are no numbers. At some point it's more abstract math than anything else, and as they say in the biz: Math ain't about numbers.