xenadu02

When you have a bunch of stores you do sign distribution deals with a nationwide distributor. If you buy enough fuel you might even be able to sign with an oil company that has refining capacity or their own distribution arm.

So Costco is definitely not paying the same price as your local convenience store. But they aren't getting a 35% discount either.

As many others have noted: Costco does not do loss-leaders. They're likely selling at cost or a very small profit above cost.

When we bought a house in 2021 I had 9.6kW of solar installed (and 20kWh of battery) and bought our first electric car. Since replaced it and have two electric cars now. At times like this I'm thankful for that. I make my own driving energy on my roof, covering over 80% of all driving and home electric usage.

The amyloid treatments slow cognitive decline by 35% or so for 18 months. Some papers argue (and I believe) the effect is small enough many patients don't notice or seem to benefit.

We shall see if yet another generation of amyloid-targeted treatment does any better. The amyloid-hypothesis boosters seem to think so - we had a thread about this not that long ago. But I remain skeptical.

There have always been opportunities for those who seek out the less traveled paths.

If all your peers are only accepting what LLMs tell them then you should do the opposite. Spend some of your time writing code by hand, especially when working at a new layer of the system or on new things you've never looked at before.

While the agents churn out more CRUD crap in the background spend your attention learning how everything works.

I do way more experiments with way more things by taking advantage of LLMS. When I have them do research I use that as a starting point for my own reading (but don't blindly trust them).

I find I am learning more faster.

Change is an opportunity. Use it to your advantage. Let other people while and moan while you get things done.

I assume the parent comment meant extract as in vampiric. Collecting rents merely for owning assets rather than doing something useful.

China certainly allows you to get rich from starting a company so long as you're well-connected politically and/or obey the party. There is plenty of benefit to be had in a way absolutely not possible for someone in the USSR.

As others have noted they've been found colluding previously, more than once. Even admitted guilt and some people went to jail over it.

And the reason the market is so small is because of players who did as you propose, the DRAM market crashed on them (not due to them but due to broader market dynamics), and they went out of business.

So there is, quite perversely, strong incentive to collude if possible and if not to abide by an unstated agreement to be extremely slow to expand DRAM capacity.

That said...

SK Hynix has two fabs under construction, one set to open later this year. One for 80k wpm, another 360k wpm. A third was recently approved to start construction.

Micro has an overhaul of Fab 6 in Virgina near done and slated to produce first wafers this year. It also has a large expansion in Boise nearly completion for 300k wpm. They're also building a "megafab" site in New York (my guess is >1m wpm) and expanding at their Japanese plants (40k wpm).

Samsung has two under construction, P4 for 100k wpm.

Nanya - historically a secondary player focused only on producing non-leading-edge DRAM is making a play for the leading-edge market with a new fab dedicated to DDR5 and HBM.

It appears they've all decided that the demand is sufficiently sustained that building new fabs is worth doing... but it takes time. If my guess is right Micron is making the largest play with their New York site at 1-1.5m wafers per month. That would give them a large chunk of the market.

(wpm = wafers per month)

I don't think your argument is a genuine one in this case.

The government is and always has been the primary driver of science research funding. There simply isn't enough of a known payoff for private investors to bother and private philanthropists have never come close to making up the shortfall. Or perhaps more accurately the government stepped in because private individuals would not and the risk was falling dramatically behind the world in research.

Now new administrations have always been able to influence grant awards. But rarely have they been so ... bad at their jobs? Who really thinks a keyword list is the right way to filter grant applications?

People have always been driven by ideology but the current US admin is just so stupid. Shockingly incompetently stupid. Even if you agree with their ideology do you really want this clown show being the ones implementing it in such a ham-fisted idiotic way?

I agree and I'm glad we have the alerts. I'm just saying we may not get any warning and that's nobody's fault.

I'm also thankful that we have at least some seismic standards and have continued to increase them.

Look at how a country like Chile fares where they don't just have standards they enforce them. You can't bribe your way out of proper seismic engineering for the most part. Then look at Columbia which technically has NSR-10 but enforcement is... not exactly stringent.

The results are obvious. Where standards are lax or just as importantly enforcement is lax structures collapse and people die. Where the standards are good and taken seriously damage is much less severe.

Even here in San Francisco things are not all perfect. Sometimes things slip through the cracks like the Millennium Tower. Others are known: the city has a special seismic hardened firefighting pipe network with big metal rods that clamp the pipes together to prevent joints from coming apart. With special seawater pumps and even ship-based seawater hookups so if all else fails they can pump vast volumes of seawater into that network... yet that network has absolutely massive gaps that have been known about for decades and still exist but there are no plans to expand it whatsoever. You'd think the city would continuously invest in this system to slowly expand it over time even if just a bit. But nope.

A difficult lesson to learn but one that led to a 10x increase in income for me.

You can fight the system all you like. You might make a difference but you won't be rewarded for it.

Or you can accept the game and play it. If you don't like the current game switch companies and play a different one.

Success is almost always a combination of both. No one is self-made but that tends to irk people who believe they are because they've spent years doing a lot of hard work. Without that hard work they'd have failed! Yet hard work alone is not sufficient.

Where you grow up, what schools you attend, who your childhood friend group is, luckily not getting hit by a bus and put in the hospital, etc. A whole host of factors conspire to grant or deny you opportunities.

However once the opportunity arises you have to seize it. Many people have lots of opportunity yet accomplish nothing because they lack the skills, will, or both to do so. Conversely many people with the drive and skills never hit the right opportunity and end up nowhere.

For whatever reason the broader culture seems to bounce back and forth between both extremes. The whole "pull yourself up by your bootstraps" and "get a job you lazy bum" crowd acting as if you can get rich purely with hard work. Then on the other side there seems to be this well-intentioned idea on the left that if we simply give every under-privileged kid opportunity they'll be successful.

Real humans and real lives are far more complex than such simple notions.

I think that overstates the case for the parametron though.

Even compared to contemporary vacuum tubes they were slower. Contemporary bleeding edge vacuum tube computers could reach Mhz frequencies. Parametrons were 10-15kHz.

The supposed power benefits depended on state. At idle they were significantly more efficient but the more work they did and the faster you tried to run them the exponentially more electricity they required (and heat they generated).

Lastly because the early ones were lower frequency they required physically larger inductors. Contemporary vacuum tube computers took up similar amounts of space due to cooling constraints but parametrons were bulky in ways you couldn't fix.

Parametrons were a dead-end technology almost from the get-go due to physical scaling limits. Even the earliest use of discrete transistors blew them out of the water beyond any hope that some magical physical scaling would rescue them.

Abandoned technologies are usually abandoned for very good reasons not just because they happened to lose the race or get backed by the wrong companies. No amount of investment was ever going to help parametrons catch up with transistors - just as no amount of investment was going to help vacuum tubes catch up either.

That's unavoidable. We now have shake alerts in CA but if the quake happens on the Hayward fault across the bay San Francisco isn't going to get much warning no matter what anyone does. By best estimates we may get 15s of warning.

The traditional model of compilation is the biggest issue holding back fast incremental compilation IMHO. Swift suffers from this as well.

Even a simple change to one file results in re-parsing the whole library because definitions come from anywhere and we have to obey the 1970s single file compilation model. The result is a driver spawns 8 threads and each one wastes time re-parsing every file in the library looking for definitions. AFAIK Rust doesn't really track dependencies at the file or function level either so it doesn't really know what changed.

To me compilers should be content-addressed databases. Each declaration and its associated content generate hashes that roll up to its containing type or namespace, then to the file, then to the library as a whole, along with hashes of the dependencies. Changing the type signature of a single function should result in the compiler being able to cheaply determine whether that has any visibility and if so to what other files in the same library or if it affects the public interface.

A file that hasn't changed and whos inputs hasn't changed should re-use the IR from the prior compilation. Even for an individual type that should be the case so changing the internals of a function in a struct only regnerates that one function and nothing else. The compiler knows deterministically that change can't have affected anything else.

That has major benefits for code completion and editing as prior compilations can feed into generating errors or suggested corrections.

Then you can take things a step further and JIT a changed function, injecting the new machine code on the fly so long as the shapes of the types don't change. Very useful for debugging.

Compilers are mostly held back because the people who write compilers are stuck on certain ideas about how compilers should be written.

I didn't leave the Republican party, the party left me. Then again I'm some kind of crazy nut for believing that principles matter even if you lose the election and that cruelty isn't something to be praised so what do I know.

Also note this is a side-effect of the fact that mammals lost true 3-color vision then primates re-evolved it by mutating the yellow cone into green/red cones. If you look at the sensitivity of red/green cones they almost overlap, as opposed to blue cones which have a more reasonable placement on the spectrum line. If you look at say birds with 3-color vision their distribution of cones is more like our blue/green spacing in the spectrum.

I believe the leading theory is that red light is given off by a variety of fruits when ripe so arboreal ancestors with that mutation could much more easily locate food. To most mammals red ripe apples and green unripe ones are all just shades of yellow so the mutation would have been like a superpower: the equivalent of eagle-eyed vision.

One thing you learn if you study the Amish or interact with them personally: there is no "Amish generally" or "the Amish usually".

Every Amish congregation makes their own rules. The local deacons vote on changes. Each "Fellowship" of congregations has tenets and rules they require to be "in fellowship" but otherwise don't have control over the individual groups. Splits are often driven by disagreement over the fellowship rules.

There are Amish that own cars but the group might specify that each family may only own one car and it must be model X and of color Y with a specific set of features. Most people are aware of the groups that prohibit cars - requiring use of horse and buggy.

Some Amish allow electricity, but only connected to the barn or only when provided by solar power. Or they might only allow it to the extent required to comply with regulation - for example refrigeration for milk on a dairy farm. Some allow cell phones, but not in the home (you leave it in the barn or a lockbox). Some prohibit them entirely. Some Amish groups allow machinery but apply restrictions that don't allow use of the machinery to end up with one farmer greatly out-producing others. So when a farmer gets old they may allow them to buy a tractor for fieldwork but require it be pulled by horses.

You might also be surprised to learn many Amish experiment with new technology. Often the deacons will appoint a "nerdy" (for lack of a better term) person in the community to use some new thing. They'll evaluate its effects and costs. Then hold a vote on whether to allow it and what the rules will be. Sometime their answer to changes is not "no" but "not yet".

The decisions are nominally driven by a desire to be independent and to avoid negatively affecting family life rather than rejecting technology directly. Most don't allow electricity, cell phones, or internet inside the home because it would lead to people not interacting or connecting with each other. When power tools are allowed they're often pneumatic so they don't need to be tied to the grid and thus dependent upon "the English" - though with Solar and batteries that has rapidly changed.

Most don't allow extensive use of farming machinery because it could lead to your neighbor out-producing you, leading to them getting much richer and having more than you leading to discord in the community. If cars are allowed they may require everyone to buy the same car so there is no jealousy or "keeping up with the Joneses".

Amish are just people so they certainly have assholes and jerks but many are friendly people and if you treat them genuinely they'll repay the favor. They will not be offended that you use a cell phone or drive a car.

Surprising how often things boil down to "they're just people and most people, if treated with respect, will repay in kind".

Well studies show that early school start times are bad for kids anyway so schools should fix their schedules. Kids also naturally like to stay up later these days so more daylight would better match that rather than keeping awake with artificial light.

I also dispute that site's claim that everyone starts before 8am. Here on the west coast I'd say most jobs start at 9am unless they are retail. A lot of schools have also switched to later start times.

Not to mention: DST is already in effect for 65% of the year.

Permanent DST has been introduced in congress every year since 2018. A few years ago it passed the Senate but the house didn't take it up. This time the house has passed it but who knows if the Senate will take it up. I'd prefer to be on permanent DST but I'm not going to hold my breath waiting for it to happen.

I liken it to the story of Noah. Whether that was the mediterranean re-joining the Atlantic and thus oral re-tellings from a much much earlier event or merely a localized flood you can certainly imagine someone preparing for a flood and surviving localized or wide-spread destruction. But two of every animal? That's not a stable genetic population. Hell there are 40,000 or more species of spiders! There is simply no possibility that you could even fit enough animals on a boat of any kind to make that story work. If it did happen the immediate result would be complete genetic collapse and extinction. The idea is abject nonsense but the core story probably happened.

It is easy to imagine a large group of slaves escaping or being freed from Egypt. Maybe they or their ancestors were war captives. But wandering the desert for 40 years? Yeah right. Even if you want to grant miracles the idea that all of Egypt would even know about such events at that time is bananas. Information didn't travel that fast. Probably one group of people in one city. And the antagonist could easily have been a local lord. Over time it became the Pharaoh and the 18 months of wandering turned into 40 years. Only then it was written down.

You don't have to do the exercises or follow along with the book. No matter if you are new to the field or have decades of experience: almost everyone would benefit from watching these videos if they haven't already (or done an SICP course).

Most piston aircraft engines are still air-cooled which really means air and oil cooled. The oil is a big part of getting heat out of various parts of the engine.

That also makes them harder on oil as the piston/rings have larger tolerances so they don't expand and bind up during operation. That means greater blow-by at startup and when operating at lower temps which puts a lot more combustion byproducts into the oil. Ultimately you want to run an aircraft engine in the upper part of its range (65% power) continuously and don't let it get too cold.

This is also true because 100LL still contains lead and at lower temps the lead combustion byproducts precipitate out of solution, coating everything in metallic lead, lead oxides, and various other lead compounds all of which are really bad for engines. Converting to unleaded nearly doubled engine life in autos.

Many modern engines have valve rotators and hydraulic lifters. Oil pressure is fed to a lifter that sits between the valves and the cam and automatically takes up for any variation in the system, ensuring valves operate correctly. If you ever wondered why car engines don't need to have their valves adjusted every 20k miles anymore - that's why. In some engines if these leak down after shutdown it can cause trouble starting because the valve timing will be off until oil pressure re-fills the lifter.

Rotators are little spring mechanisms that compress and when uncompressing try to rotate the valve in one direction. This causes the valves to rotate a tiny bit with each cycle. Often there are hot spots and exhaust valves especially often have no good way to shed heat yet are exposed to extremely high temps - so they shed heat when they close and are in contact with the head. If they don't rotate the slightly hotter spots will continuously build up heat eventually destroying the valve. The rotator keeps that from happening. (Some engines use sodium filled valves to help transport heat away from the valve face).

I always found it surprising how tiny variations in wear or even a few degrees of excess heat can end up destroying an engine.

That's a dangerous game to play.

It requires them to bet that:

1. This AI-driven demand cycle is a temporary blip

2. None of their competitors will make massive investments in capacity.

3. No state actors will do so despite evidence to the contrary.

Many in silicon have gone out of business due to the boom-bust cycles so being a bit conservative is appropriate but at some point not expanding production becomes an existential crisis.

Or ya know enough governments with the power get really angry and start forcing price controls, nationalizing, using their state spy agencies to steal your secrets, or use unofficial back channels to make your life extremely difficult.

Wasn't this also one of the last chips laid out by hand (literally the masks were cut and laid out physically)? Or am I thinking of something else?

I sometimes wonder if some design decisions were made on that basis.

> That said, (before things like SPF) it was easy enough to deliver email to anyone you wanted even if you didn't have your own real email account and SMTP server

Yup, this was also a fun exercise. Use nslookup, find the MX records, then telnet to them and deliver some mail destined for domains that email server handled. At that point you're just a slower version of a mail daemon.

In 2006 when FIOS first rolled out they assigned ARIN IP blocks to anyone who requested a static IP and I hosted my own webserver and email server on the web. So for a long time I had my own netblock and my email server had been around for so very long it was in everyone's legacy trusted list. Even gmail never bounced emails from me, despite not having SPF or DKIM and whatever else. Though I did eventually set those up. Only authed users could send email and that was a limited list so my netblock never actually delivered a spam message.

Funny enough I also setup default routing so all non-registered addresses went into a separate mailbox for me and I used the "companyname@mydomain.net" for everything. For many years I knew about every single data breech before anyone else. Often months or years ahead as spam would arrive out of nowhere to that address.

When I finally moved to the bay area to work in tech for real the entire world had changed. You couldn't just get a netblock assignment anymore. And lord knows even if your ISP wasn't deliberately sabotaging your ability to run a server every other system on the internet would assume you're a scamming spammer. I had to give up self-hosting on-prem.

Ah this one was so long ago I forgot about it.

The realization that (on DOS) "copy con file.txt" as the world's worst text editor or "copy file lpt1:" was treating physical devices as files. Everything on the computer was arbitrary so you could make anything behave like anything else (for some definition of behave like)!

Not a great insight I'll grant but a key one that everyone has to go through to be any good.

> This is false. They slow down disease progression by about 30%, as measured by cognitive outcomes. This is discussed in the article.

Perhaps I am just not well-informed but 30% slowdown in progression translates to sufferers have some mild improvement in cognitive tests and live a few months longer.

Maybe it is simply too early to tell but I would naively expect something much more significant. Perhaps this is the sort of thing that requires much earlier treatment to demonstrate better results.

I'm not saying amyloid beta research should be terminated. Merely that everyone in the field should be willing to entertain other ideas.

I don't think that's quite true. Unifi 10GbE switches are cheap enough I have a 24 port PoE+++ one in my house and my 3 main WiFi APs are 10GbE connected. My MBP uses a 5Gbps Thunderbolt adapter that runs relatively cool as well. All of this is over the existing Cat5e wiring.

I'd say 10GbE has arrived. It is relatively cheap, most of the time works over existing 1GbE cabling, and gracefully degrades to 5/2.5/1Gb based on conditions when it can't reach 10Gb.

Yes to be 100% guaranteed of getting 10Gb even in bundles of 100 cables running over noisy fluorescent ballasts to a full 100m you need Cat6A but in many environments Cat5e or Cat6 is more than sufficient. It works so well if you fail to get the full 10Gb I humbly suggest you re-do the terminations on both ends before considering replacing the cable.

As a kid in the late 90s my mind was blown when I realized I could telnet to port 80, 25, or 110 and interact with the servers manually.

Simple get: GET / HTTP/1.1 Content-Type: text/html User-Agent: l33t hax0rs lol X-Funny-Monkey: farts

For sending a mail message on port 25: HELO mail-from: whoever@whatever.com mail-to: sysadmin@yaya.com <other headers> <blank line> Body of the message yay. <two blank lines to end>

POP3 was so long ago I forgot but you could list the mailboxes then get individual messages and so on.

This revelation was the beginning of "there is no magic" for me. The realization that every part of the computer was built by human beings and was at some level understandable if one undertook the effort.

Perhaps most people in the future won't bother. They'll just let agents do it all. I'm sure that will leave some interesting holes in various systems for people willing to actually learn how they work without the filter of a model (or its safety rails).

Let's put all of that aside for a moment.

When the first drugs targeting HIV arrived the results were undeniable. Yes the drugs sucked for various reasons and yes HIV would evolve resistance. But the data demonstrated a very clear link that these drugs suppressed HIV and suppressing HIV made people live longer. Or consider mRNA and COVID, a great success story where the technology was put to good use and the results are obvious.

On the flip side we have certain cancers like certain breast cancers, melanoma, etc that never had a "wow" moment where some miracle turned them from highly fatal into treatable but we have seen decade after decade treatments improve and survival rates march ever upward such that what were once almost guaranteed death sentences are now often very treatable.

These are two disease treatment models worth keeping in mind. Sometimes major leaps are made. Sometimes progress is slow.

Now if we consider amyloid beta therapies: we have treatments that target amyloid beta with varying degrees of success but at least some show definite reductions in amyloid beta plaques. To the best of my knowledge that has not shown to improve outcomes in Alzheimer's patients to any meaningful degree.

That concerns me and I think justifies some skepticism of the amyloid hypothesis. The data is messy but if amyloid beta were a symptom not a cause that could certainly fit the results we are seeing. That doesn't mean the amyloid beta hypothesis is wrong but I think skepticism of the "state of the art" in the field is warranted given the pathetically ineffective progress made to date.

By definition a free market in healthcare requires forcing some people to die from a treatable injury or condition to discover the optimal price. That's part of the definition.

That's by design.

If currency doesn't devalue then stuffing it under a mattress looks like a reasonable alternative to investing. If we hit deflation you can receive gains for "free" and borrowed money becomes more expensive over time. Neither of which our economic system is setup to handle.

We punish people who hoard cash by devaluing it thus encouraging them to put the money to work.

> with the US actively sabotaging their access to information and hardware

More like they didn't see the need and didn't invest in early computing and so lagged behind. Something the US very nearly almost did as post-war a lot of people didn't see the point. Software engineers (though they weren't called that) were seen as unimportant secretaries who just typed letters into the computer. Grace Hopper's Navy computing unit famously had to raid other offices at night for resources.

> I suspect this was partly to facilitate cloned hardware, but i do also suspect they wanted their systems to be approachable by engineers from around the world, too, so diverging too much would have been detrimental.

Computing advanced so quickly it showed the Soviet-style communist system for the lumbering boondoggle it was. By the time the central committee deigned to allocate resources for computing they were a generation behind and that only got worse. They stole western design and software because they didn't have the economic leeway to do it themselves.

You need to remember: by the 60s/70s the western economies were taking off in says Warsaw countries were never able to match. They simply did less and did it less efficiently across almost all sectors of the economy. No one had to cheat them or restrict them. They shot themselves in the foot repeatedly.

There were no startups. You had to petition the central government for permission to build more computers. They would assign you a quota. You built that many computers. In the mean time 12 new startups were founded in the US and two of them came up with new chip designs and shipped them to customers. Doesn't take a genius to understand why the soviet sphere was a compete non-entity in computing.