The ark of the covenant was itself a cultural fusion of Egyptian New Kingdom solar barques, used to carry idols, with the practices of the aniconic Midianite Yahwists of the Sinai region. The proto-Levites brought this fusion with them into Canaan, where it melded further with the endogenous Canaanite worship of El.
It's a cool bit of history, but connecting it to diffusion models feels like a bit of a stretch!
I read it more like a complex joke, or rather, a stab at the problem of why generative material of that caliber feels so fundamentally wrong to some of us.
Weird. Costco is never the cheapest price for Coke in my area. Or the previous state I lived in. At least if you bought during sales at competing Supermarkets.
Where they win my Diet Coke business is that they are consistently a "good deal" - never the cheapest, but never the most expensive. And I don't need to wait for a sale to stock up.
I still buy soda from the local grocery stores when I happen by a deal when shopping for other things. But I don't go out of my way. The specials are usually much better than Costco though. By 25% or more per can.
Honestly this goes for most things I buy at Costco. Even in bulk, their per-unit costs beat the standard pricing at competitors - but never the sale prices. It just removes mental overhead. And the bulk prices are always better at a restaurant supply store. Where Costco wins in this area is curation so I can reduce mental load, but then again Costco does the bullshit where they stock something for 6mo out of the year and you spend an extra 20 minutes searching the store for it the next time you need it only realizing they stopped selling it. Then it appears randomly a month later.
They are a decent middle ground for me. I can buy 10lbs of chicken breast of decent quality for $2.99/lb I can toss directly into the freezer pre-packaged into meal-sized portions. Or I can buy 40lb boxes of bulk chicken breast for $1.39/lb from the supply store and spend a couple hours vacuum packaging it myself.
I definitely have a love/hate relationship with the place! I see Costco as the upper middle class consumerism version of nerd sniping.
I shop Costco because I don’t want to have to chase sales all over town. I did that in my 20s when money was tight but now I value my time more than saving fifty cents on chicken.
The soda thing is pretty crazy. I’ve mentally tracked it for about 20 years and Costco never beat supermarket sales for. Are brand soda. It was almost absurdly overpriced for something at Costco. Now their every day price is cheaper than the cheapest weekly deals.
20 years ago you could routinely find 12-packs for $2-$2.50 when bought in bulk. Today $5.50 is the best I’ve seen (48¢ a can). Costco is $15 for 35 (40¢ a can).
In my area, Costco is $15/35, BJ's is $19.49/35. Target is $8.99 normally, but the go on sale for 3/$20. (That's like one every 4-6 weeks, where it used to alternate between Coke and Pepsi bottler products weekly)
Sometimes other markets do the similar, but the prices are keeping up. I suspect that Costco has longer-term contracts, and the other retailers are stuck eating the increased costs for aluminum cans.
I don't know if this is how Costco itself thinks of it, but you can get to a pretty decent understanding of their financials by assuming they sell everything at cost and pocket the membership fees as profit. Their actual profits tend to be about 50% higher than that implies, but given the size of the rest of the business that means the membership fees are only about 2% of the revenue overall it's not a bad approximation.
They can twiddle with dropping a bit of price here and raising it somewhere else and still have the same ratio of cost.
Of course you can still think of things in terms of the cost per item versus the profit per item, but the stability of that ratio over many years makes me wonder if maybe Costco thinks of the membership fees that way.
It would be illogical to make gasoline a "loss leader" because, surely, there are people who visit the gas station and then jet out of there without going anywhere else. A loss leader item is something that gets people in the door, puts their butt in a seat, and gets them to make a transaction that will include piling on other stuff. So if your Jumbo Jack is a loss leader, then you can charge $2.00 for sodas and $3.00 for French fries, and the same customer will probably also grab a $6 Oreo shake.
Someone who's filling up for gas may or may not go inside the Costco store proper. But the gasoline isn't going to influence their buying habits. If they "just need gas" that's all they'll get.
Conversely, I cannot see how a $1.50 hot dog is doing anything but losing money. It's a classic candidate for loss leader.
The average Costco ticket is $150 out the door. The retail margin is 11%, or $16.50.
Say you fill the car with 15 gallons, and they lose $0.25/gallon. That’s $3.75. If 1/3 of the gas station customers go in the store, they’re making $5.50 from retail, less the gas losses they net $1.75.
They make money, deny their competition business, and lock you mentally into saving money on gas and renewing that membership. That membership is half their operating income.
I’m probably being pessimistic. One thing with Costco is they are incredibly efficient, so I’m sure they’ve baked out cost and overhead. You can see stuff like how they provide the minimum number of garbage cans.
One good example of DoD which isn't CPU-cache-related is relational databases. When designing a CRUD application, the Data-Oriented way of doing it is to figure out what data you will be storing and how to organize that data to minimize access times, which is what you're doing when you design the DB schema.
An example of failing to follow DoD is the N+1 query problem: a programmer builds an abstraction that operates on individual DB rows, but "where there's one, there's more than one": you will inevitably be running that code in a loop so that you can process multiple rows. If instead the programmer had abstracted over groups of rows, then per-item query overheads suddenly become per-batch overheads.
I consider it generally the ideology of anti-OOP. While OOP ideology teaches you to structure the program after the problem it solves, DOD ideology explicitly teaches you to throw all that away and think about what runs fastest on the computer. Maybe it should be called Computer-Oriented Programming or Hardware-Oriented Programming. The specifics very a lot but the top-level ideology of "fuck OOP" is consistent.
> While OOP ideology teaches you to structure the program after the problem it solves
I completely disagree with this characterization. OOP teaches you a synthetic set of concepts (go4) and then asks you to solve problems in terms of that.
And the reason why the canonical bird as a subclass of animal doesn’t work, is it’s extremely difficult to divide the world into strict categories (are you Aristotle). So the solution is to organize virtually rather than around natural traits.
The most natural way to solve a programming problem is a big list of instructions with if/else and goto. It’s very learnable, even for young children.
I responded to that. It’s nonsense which doesn’t work:
> And the reason why the canonical bird as a subclass of animal doesn’t work, is it’s extremely difficult to divide the world into strict categories (are you Aristotle)
What behavior are you going to out in vehicle? Here are a few potential subclasses you’ll need to consider:
I actually see ECS as a subset of the relational data model. It's effectively binary relations, in database / E.F. Codd terms.
When you look at it from that angle, rather than as an optimization technique, it makes it clear there is actually an elegant programming model here.
Unfortunately game engine programmers tend to think databases are super uncool and not relevant. They could actually learn a lot.
"flecs" pulls in some concepts from the relational algebraic world in that it has some sense of joins, etc. but it's a bit ad hoc.
The ultimate "data oriented design" game engine could be a high speed, GPU/SIMD accelerated, in-memory Datalog engine. And then the game world expressed in Horn clauses and logic.
Lisp is the perfect family of languages for a config. Homoiconic for seamless config generation. S-expressions for shared human and machine readability.
I actually enjoyed the first hand experience of an early-ish Google described at the beginning. For those with enough tenure in tech, was this ethos unique (at the time)? Or was it typical of successful, tech-heavy startups? I could believe an early Sun might've felt similar but I don't know.
Early Microsoft (well, and current micorsoft) did not/does not shower their employees with massive stock grants.
Google was one of the first tech companies to pay so well that people switched industries to get into tech both to work alongside brilliant people from various backgrounds and to make a lot of money.
Before Google, it was very rare for an independent contributor/software engineer in tech to retire early if you didn't hit the startup lottery. Sure, people who joined Microsoft pre-IPO did well, but it's not the same as now where a good 10-15 year run at a FAANG allows many to retire early (if they choose).
It did not shower people with grants but I met plenty of people who made a killing with post IPO stocks and pay at Microsoft. It also had a great culture with vibrant campus life, lectures, free soda and other things that were uncommon at the time.
Yes, this is something almost all other anti-bot/fraud prevention solutions already does, and there are already bypasses developed simulating human mouse movement.
Not sure, but I struggle with skepticism for anyone who blocks archive.today, which cloudflare does, along with nextdns and others. Being blocked by such a large... apologies in advance for 'lack of better word' vernacular, cartel, is a near death sentence.
Although the blocking of archive.today goes back years, as can be verified through forum searches and archives with nextdns and others, I was not aware of this and have no excuse to dispute it. But for the record, the blocking predates 2026 by many years -- and my own records also verify this. That said, I think I need to learn more.
It's a cool bit of history, but connecting it to diffusion models feels like a bit of a stretch!
reply