fugacity

I just checked. Yup, you’re right. Funny though that Pat of all people claims it’s not dead lol

Moore’s law makes no comments about the cost of each transistor in an advanced process. And believe me, they ain’t cheap. It’s not a coincidence we’re up to PLC flash… why go for 32 levels when TLC is likely already a pain?

You wanna store a few hundred bytes? Print some mechanical knobs and call it a day. You wanna make some real storage devices?

Hire top PhD:
Physicists for quantum effects used (and parasitics mitigated)
Chemical engineers for CVD and other very hard and expensive clean room processes.
Electrical engineers to design analog circuitry for charge pumps and multi-level cell readout technology, as well as digital VLSI/HDL design for digital logic including storage controllers
Mechanical engineers for packaging design and automation for your expensive and dangerous production line
Civil engineers for your fab plant, which is so large that significant infrastructure needs to be built to support your fab (e.g. TSMC in Taiwan funded/built a municipal scale desalination plant of which a significant fraction is used for semiconductor processes)

Until we have replicators as the other commentor pointed out, I’m afraid we aren’t even close yet. Fingers crossed we hit type II civ sometime but I won’t be holding my breath for it.

Maybe it’s a cartel but I don’t have my hopes up. Storage technology is only getting more complicated and the number of players is only decreasing.

In my view (and maybe I’m wrong) there’s just not that much money to be made in it, contrary to what consumers think. Why fight each other over pennies when you can both earn dollars? Maybe if China figures things out, but you can be my ass I’m not gonna trust a CCP backed storage company lol

Moore’s law hasn’t died, if you mean number of transistors per area. Linear scaling to transistor counts has.

Unless this is a matter of price collusion (which I doubt as it appears more as a supply demand issue) I don’t think this unregulated capitalism is bad. Last I checked making any kind of products involving semiconductors isn’t cheap or easy. Maybe it is once you figure out how to, but the R&D costs involved are insane.

We as consumers want prices as low as possible. Suppliers want prices as high as possible. Samsung (and the like) clearly aren’t willing to make more of a product at the price that it is currently at (which is a mistake to begin with). There are plentu of other players making ssds, and the prices are all very similar. Something tells me that they’re not gonna price things for cheaper because they can’t survive that way.

I hate Apple as it’s an anti-competitive walled-garden monopolistic closed-standard anti-repair evil trillion dollar corporation, but this isn’t true. Modern iPhones have closed the gap significantly in hardware specs (display, processor, optics, IPXX rating, and now thanks to EU even USB-C) and they’ve always been better for general use in software. That, added with the fact that flagship Android manufacturers have learned how to play the pricing games of Apple, means that Apple’s price to performance ratio is pretty competitive with Android phones these days.

Their main products are pretty good these days, as much as I hate to admit it. I’ve never even owned an Apple device, and won’t as long as I can.

Well, capitalism definitely has a role, but it’s not exactly a coincidence that China started out with cheap labor (and maintained it so). A country that manipulates its currency for specifically for export reasons is definitely also to blame. (Before you say but US also manipulates currency, the levels of currency manipulation are not comparable: if they were, BRICs would be our world reserve currency)

Anyways, new places don’t go to China for labor, they go for overall manufacturing costs.

All that said, from my (somewhat limited) experience Chinese manufacturing is sort of a niche. If you’re willing to invest all the resources into NRE and QC and not afraid of corporate espionage of your manufactured product, you can definitely save a lot of money (China really isn’t all that good for prototype or small batch manufacturing if you need a made-to-order part/product as the headache from language barrier and quality issues are greater than the cost savings). Apple clearly makes it work because they don’t care if you copy their PCBs - good luck copying their custom-designed ICs.

“Creating streets that are safe and pleasant for people outside of cars promotes alternatives to driving.” I don’t disagree with this, but the problem is that in the US there often aren’t any alternatives to cars to get around. And to be frank, I’m not gonna be walking around on the streets of LA (where I live, insert your crime-ridden US metropolitan here) unless I have good reason to. Getting hit by a car due to RTOR is the least of my worries as a pedestrian. I think a lot of change is necessary (such as locations of stores, etc) beyond safe streets to reduce the need for cars. For instance, if costs of living in the city were better, people wouldn’t need to use cars to commute. Maybe it’s a starting point to fixing our transportation issue but honestly I don’t see it.

“A minute or two delay… actually doesn’t amount to very much, and that’s what a typical case would be of forcing a driver to wait an additional cycle.” You say this, and it might be the case the vast majority of the time, especially if the stoplights are separated by a large distance and there aren’t many cars, but traffic is a distributed problem and without seeing some sort of study that indicates this I don’t buy into it. During heavy traffic, if the cars from one intersection back up into a previous intersection due to reduced throughput I can’t imagine how an additional cycle is the only cost. Maybe this is just dependent on the traffic situation, because I have a natural bias to think towards traffic situations in LA (which don’t necessarily represent the rest of the US).

“The Philadelphia paper is the seminal work on all way stops being safer than signals in urban contexts.” Can you tell me who the authors of this paper are or maybe offer me a link? I would like to read it, thank you.

“Studies on roundabouts being safer are… even more conclusive and abundant. I really can’t cite just one because damn, there’s so damn many.”
Yeah so I’m pretty sure roundabouts are better in every way except for space. But if only getting more space would be easier, because surely we could just replace a lot of our roads with trains at that point right? I think roundabouts are a red herring because they literally don’t fit in most of these intersections (they don’t even have space for a left turn lane in many of the intersections I drive in). Heck, if we’re talking about space-throughput tradeoffs we could just theoretically make every single intersection a graded interchange and that would provide a huge amount of throughput (but this too is a red herring).

" All-way stops and, of course, roundabouts are both provably FAR safer often with no impact or a positive impact to overall congestion." This is a pretty big statement to make, and I was wondering if you could provide me the sources for this.

“The city values keeping more cars moving faster over both safety and financial responsibility.”
But isn’t keeping cars moving faster financially beneficial? From an energy perspective, needing to stop for every stop sign is way worse on fuel economy than going through a string of green lights and stopping every now and then. Don’t get me wrong, I think using cars as a main mode of transport is incredibly stupid, but I think there must be some tradeoff between time/money/resources wasted due to traffic and time/money/resources lost due to premature deaths or poor living quality due to (non)fatal accidents.

Let me preface that I think using vehicles as a primary source of transportation inherently scales poorly, and you can easily argue this by looking at how much a road costs versus a rail and how much mass you need to move per person on car versus train.

That being said, I really hate this article because it relies on anecdotes from various people and opinions without making any effort at citing relevant statistics. It literally cites the TOTAL number of pedestrian deaths to vehicles in 2022. I tried to find some statistics on right turn on red light, but all I could find were 20 year old or older studies, most of which actually concluded that right turn on red doesn’t really account for a large number of pedestrian injuries and deaths. Like this one, for instance, which claims that right turning on green can also result in pedestrian accidents which could result in much more severe injuries (I can see how this might be true but there’s no evidence to back this up.)

It’s interesting for me to look at this from a utilitarian perspective: Surely there is a tradeoff between the amount of time wasted due to traffic increase due to right turn on red, and the time equivalent to the amount of lives lost due to RTOR (assuming RTOR results in more deaths). This of course is an incomplete/flawed way to look at things as we don’t give highway collision motorists the death penalty for causing huge traffic blocks; iirc though it is how a lot of safety studies are done (look into how the statistical value of a human life is determined from highway transport administrations).

I would really appreciate if someone could chime in with some actual stats and numbers (though I doubt they’re readily available) about the topic, rather than some anecdotal comments. I’m not a fan of symbolic legislation that doesn’t provide real benefit (think plastic straws bullshit), and I would like to see a convincing take on whether or not this is that.

Well, the devil is in the details. People like you, who has actually figured out how to use an adblocker properly for YouTube, and me, who is willing to actually pay for YouTube premium (you’re welcome for the subsidy), surely form a small proportion of the actual number of YouTube content consumers.

Maybe I’m wrong, but my view is that the majority of users just want to watch videos without having ads and they aren’t willing to devote time (for adblockers) or money (for subscription services) and are completely ignorant that they are the product regardless. And those users act like they are entitled to content and that leaving YouTube is somehow significant to the big picture.

In a sense I agree with that piekay though. If they can’t serve me targeted ads on YouTube they lose that money trying to develop technology to track me in that regard. How much money that is I guess is hard to say, since the tracking on YouTube certainly can carry over to other parts of Alphabet.

Perhaps YouTube gets all their content for free, but it certainly isn’t free to transcode video, host it reliably, and distribute it while moderating it (given how bad Twitter is right now I’m sure they have a decent number of measures in place, even if they aren’t even “good” at it). And if it was remotely easy, believe me, there would be a lot of competition in this space.

Yes, I make Alphabet x dollars richer (or really, I make YouTube operate at a slightly lesser cost) every month by paying a subscription. And actually, I’m okay with it. A tiny cut of it goes to content creators and I get a nice piece of tech. And I support the branch of Alphabet that has technology that I think is incredibly useful and beneficial. If there’s a content creator that I like especially then I’ll support them directly.

The reality of it is that things cannot be free. Or at least it seems that way, because we have not been able to provide a free video hosting service that doesn’t take advantage of its content creators or consumers.

If you read around you’ll find (perhaps surprisingly to you) that YouTube operates at a loss. So in response to your points:

1. You can pay to get the ads removed. They make less money off of you when they can’t serve you ads, and I’m sure they’re trying to operate at less of a loss.

2. Alphabet is a public company, and it must release certain information about YouTube. Anyways, I’m pretty sure they aren’t using the money to directly line the shareholder profits. The reality of it is that it’s probably just another arm that Alphabet uses as part of its monopolistic tech deathgrip, so it’s not gonna be a straightforward computation. Maybe Disney could be used as a metaphor here?

If you don’t wanna pay to support that, I don’t exactly blame you. But practically, I don’t really agree/expect that YouTube should serve you content (or even more so, people with aggressive adblockers) without you giving something in return. Either you eat ads, you pay for a subscription, or you become the product (unfortunately this last point might be true irregardless).

They’re definitely still tracking their premium users, I agree. But my counterpoint is, what business, online or not, doesn’t track me? If I go out and buy something at a retail store I’m gonna bet my ass I’m being tracked. If I don’t want to be tracked, then I should be making sure information I consider to be sensitive is not being exposed. If there is no reasonable expectation to privacy in the public, then I think it’s fit that there’s no reasonable expectation to privacy when I’m surfing the internet.

Just like a few of the other posts, I honestly don’t get it. If they can’t sell your data and can’t serve you ads, then why would they want to spend money serving you for free? There’s so many people complaining how YouTube has a monopoly and how it’s not even that hard to run, but I seriously doubt these people. Transcoding video and distributing it worldwide while having automated moderation is not easy or cheap. If there were serious contenders in the space people would have moved on, and I don’t think it’s just the network effect that keeps YouTube as a dominant player here.

People despise ads, but then they want content for free. They use adblockers to bypass a primary revenue source for a website, then go all surprised Pikachu face when that website doesn’t welcome them. And then they get upset that they don’t want to be the product despite not willing to be a source of ad revenue. I’m willing to pay for YouTube premium (and other subscription models to get rid of ads), but a lot of people aren’t. And honestly, I really would rather those people simply leave the site. It would lower operating costs for YouTube (I don’t expect my subscription fees to go down but maybe their engineers will have more free time to work on features besides adblocker-blocking), and more people on different sites would lead to more competition.

If you aren’t willing to eat ads, and you aren’t willing to be the product, and you aren’t willing to pay a subscription, then why do you think you’re entitled to content?

I don’t know if I would see it as a pure money grab. Pretty sure game consoles, just like inkjet printers and the like are sold with zero or near zero profit (or even at a loss). The benefit the console manufacturer gains from the platform lock-in far outweighs whatever greed they might have trying to reap gains from the hardware. 10 year old hardware is roughly 30x slower in FLOPs, so we might be looking at a desire for better games or easier software development - I for sure would not envy the developer needing to target 10 year old hardware, though it’s not exactly unheard of.

That’s mostly correct but I don’t think it’s entirely accurate. Distillation is useless at the azeotropic point but ternary mixtures are used to break the azeotrope. Once you move past the azeotrope you can continue distillation to high purity. You could also do pressure swing distillation but my guess (even though I’m not exactly a chemical engineer doing unit operations for a living) is that it wouldn’t be economical. Of course, getting “100%” pure anything is really a different story…