LetMeEatCake

Streaming services have an enormous amount of fixed costs. It might cost them several billion dollars/year to operate the necessary infrastructure even with zero customers, but the marginal cost to serve a customer might be on the order of $2/month on that $10/month subscription.

It's why streaming and digital storefronts are such a sink/swim industry. Either a company gets over user number+sales threshold to override their fixed costs, upon which they become profitable and all further growth makes them exceedingly profitable. Or the company fails to do so or barely does so, and makes somewhere between giant losses to minimal profits.

From a quick search, Spotify's user count should have grown somewhere in the neighborhood of ten times over since 2015.

This is not a cost increase that is mandated or justified by inflation. It never is. It's a cost increase from a very, very, very simple fact: companies want profit, and Spotify's leadership has concluded that they will gain more profit by increasing prices than they will by not doing so.

The primary will almost certainly be de facto decided by the end of March. There's a good chance it will be de facto over by mid February, if any candidate dominates the early states.

Of the 2467 delegates up, 1250 of them will have been distributed by March 12. Republicans rules on delegate distribution heavily favor the candidate in the lead as well. No primaries/caucuses have been scheduled yet after March 12, but expect the bulk of the remainder to be in the other half of March and all of April.

The primary is all but certain to be over by May 2024.

But why male models?

You'd be surprised at how many fabs there are in the US.

• TI has something like a half dozen to a dozen, predominantly in Texas
• Intel has more fabs than you can shake a stick at, mostly in Oregon but also Arizona
• Samsung has a fab in Texas
• GlobalFoundries exists in New York and Vermont
• Micron is in Idaho
• Wolfspeed has power electronics fabs in North Carolina and New York

And so on. The US has a lot of fabs. For best countries in the world to build a new fab, the US would rank somewhere between first and third place — and I think there's a strong argument for the answer being "first place." Unlike Taiwan and South Korea, US fab jobs and experience are not almost entirely dominated by one or two companies. The US isn't located in one of the most geopolitically risky parts of the developed world. The US has a huge population and plenty of money to put into fab expansion.

The only issues here are (a) the US has gotten worse and worse at large scale construction projects, and (b) TSMC wants to pay workers like shit and treat them even worse, which doesn't fly for technically skilled US workers. You can treat US technical workers workers poorly, but not as poorly as in much of Asia, and you definitely cannot do it without paying them very well.

As a vegetarian myself, I've thought about this a little bit.

I think it ultimately boils down to the fact that going vegan requires a lot more work from an individual. Avoiding meat might be a pain in the ass to implement at times, but the actual intellectual process is straightforward. You need to watch out for soup stocks, cheeses with rennet, and meat sauces basically. Everything else, at least in my experience, is obvious. Converting a recipe to vegetarian doesn't require too much thinking. A lot of foods are just innately vegetarian and won't be labelled as such: there aren't "vegetarian pancakes" or "vegetarian pies" out there — they're just expected to be vegetarian unless someone made a meat version. Only a small handful of pizzas will be labelled vegetarian even though most are or trivially can be made such. It's easier to find/adapt recipes that are vegetarian compatible.

Going vegan is just a full extra process. Eggs, milk, butter aren't visually obvious. Even bread isn't certain to be vegan-friendly. The ingredients being removed from a recipe cannot be simply removed, especially with baked goods, without risking the entire recipe becoming a disaster. If you take a cookie recipe and remove the eggs and butter, you're going to be disappointed; you need to find a recipe designed from the ground up to not use eggs or butter.

The extra restrictions on vegans mean they need to be much more specific about their foods than vegetarians.

I find it infinitely more usable than all of the other storefronts I've used or seen. Interacting with my library is easy and straight forward. Buying games is easy and straight forward. When it opens I'm not inundated with ads for games I don't care about, or ads at all.

Err... the first sentence that you quoted directly refutes what you said.

If there hasn't been an election for 60+ years in the town, that means everyone has "assumed office through paperwork." He's the only one that's faced resistance. This is entirely a result of racism. Don't pretend otherwise.

On the timescale of 27 years, grid-scale storage is going to be a complete non-issue. There's already a decent amount of work being done at that level right now and battery tech has been improving at a consistent pace. Renewables can work quite well as-is with a good mix of location and source. Offshore wind is more consistent wind speeds, solar locations can mitigate light cloud coverage, solar output peaks during the times of greatest human use, and land based wind is typically dispersed over large areas.

I'm a huge proponent of nuclear power, but as things stand it isn't going to be necessary on these time tables. The value in nuclear is that it's another thing we can build now without needing to wait ten years for battery prices to continue to decline or for manufacturing capabilities to ramp up. Building 10 GW of nameplate capacity wind+solar is great. Building 10 GW of nameplate wind+solar and 5 GW of nameplate nuclear is better! That's the advantage of nuclear today, and we should fucking make use of it. That doesn't make it mandatory in the long-term.

I'd expect Zen 5 in 2024. AMD has been on a decently consistent release schedule for Zen CPUs.

Zen 1: Mar 2017
Zen+: April 2018 (+13 months)
Zen 2: July 2019 (+15 months)
Zen 3: November 2020 (+16 months)
Zen 4: September 2022 (+22 months)

Lots of clustering around ~15 months, with Zen 4 as the major exception. Zen 4 had to run through the whole pandemic supply chain gauntlet to get released, which explains most of that delay to me. Especially since the supply chain issues hit semiconductors hardest. In theory that'd put Zen 5 in early 2024, but I'd guess somewhere from late spring through early fall.

I want to know when the 3D cache variants of Zen 5 are coming out. I built a good Zen 3 system about a year before those were available. Don't feel like I can justify the expense of going to a 5800X3D when I already have a 5800x, but I'd love to have the 3D cache. Once the Zen 5 versions are out I hope to make a new system around that.

I doubt anyone will complain if Blizzard's games are brought to other storefronts too.

I like Steam. Steam has the best features, best UI, good sales, and while they are not without faults (systems can stay unchanged for a long time!), they are run by a company that by and large respects its userbase.

I don't mind if games are brought to Steam and any or all other storefronts. Put it on GOG, Windows Store, EGS, Itch.io, battlenet, Origin, Uplay... You name it, I approve of it going there also. If those other storefronts want me to use them, they need to provide a comparable or superior experience. GOG comes the closest, but its inability to get games in a timely or predictable manner, if at all, is too much of an obstacle for me.

That's not what they said.

What people are calling "AI" today is not AI in the sense of how laypeople understand it. Personally I hate the use of the term in this context and think it would have been much better to stick with Machine Learning (often just ML). Regardless, the point is that you cannot get from these system to what you think of as AI. To get there it would require new, different systems. Or changing these systems so thoroughly as to make them unrecognizable from their origins.

If you put e.g. ChatGPT into a robotic body with sensors... you'd get nothing. It has no concept of a body. No concept of controlling the body. No concept of operating outside of the constraints within which it already operates. You could debate if it has some inhuman concept of language, but that debate is about as far as you can go.

Actual AI in the sense of how we conceive of it at a societal level is something else. It very well may be that many years down the line that historians will look back at the ML advancements happening today as a major building block for the creation of that "true" AI of the future, but as-is they are not the same thing.

To put it another way: what happens if you connect the algorithms controlling a video game NPC to a robotic body? Absolutely nothing. Same deal here.

I always hated that argument from people.

Even if they're right — which we all know they are not — it wouldn't matter. Climate change is going to devastate human life if we do nothing. If, somehow, the source of the warming wasn't human-caused, we'd still need to find a way to counteract it. It's not our fault doesn't prevent it from being our problem.