I'm employing a friend to work on a project of mine. Chances are very good that this project is eventually going to be scrapped, at which point he will have demonstrably generated $0 of value.

You know.... the labor that we rely on Immigrant Labor for, that we both underpay and undervalue. That if you ask 1,000 Americans to do, you won't find 10 that will be there in a week.

This is kind of a self-fulfilling prophecy. We underpay this because we're bringing thousands of immigrants in to do it for us. If we stopped doing that, people would pay more for it, because they would actually have to compete. If we keep doing that, it will keep earning virtually nothing, because it has a large captive employee base.

By "election", I mean "actually getting elected and what you do after that". Obviously you want a full plan for everyone you're putting into position, yes? You don't want to step into the presidency and say "well, shucks, didn't really plan for this".

If the goal is to take a bath, use the left image. If the goal is to build a bathroom, use the right image.

Seriously. The right says this about Biden, the left said this about Trump, the right said this about Obama, the left said this about GWB.

I wasn't politically aware enough to pay attention to Bill Clinton . . .

. . . but I've got a friend who was, and she says that, yeah, the right said this about Bill Clinton.

And she wasn't aware enough to pay attention to Bush I . . .

. . . but I've got another friend who was, and he says the left said this about Bush I, and also about Ronald Reagan.

We've gone through forty straight years of people panicking over how the President is going to revoke the rule of law and declare themselves Dictator.

There's no "Project 2025" for the democratic side

There certainly is; there's no way people aren't planning for a prospective 2024 presidential election.

Pretty accurate, from what I understand.

It's also worth noting that you need to get all the other forms up to a tier in order to get Dragon to that tier, but nothing stops you from doing uneven form tiers beyond that. Dragon is good . . . but it's always going to lag behind your other forms because of that.

My headcanon is that Belle was disappointed when the Prince turned back into a human.

ZorbaTHut
3
89TB usable

That chart seems a bit off on its historical data since it lists HDD prices at $13 TB in 2023 which they are not that even now.

There's a number of drives in the general vicinity of those prices, though I'm not completely convinced those are actually new. Shucked drives occasionally beat those prices as well.

It's not entirely clear what price we should be using. "Most common price" is probably wrong - shouldn't we care about the cheapest? At the same time, "the absolute best confluence of sales" is maybe not really useful either.

All that said, $13/tb does pass my sniff test, fwiw.

(And if you're willing to buy used drives, well, I buy most of my drives at around $8/tb.)

Also as the people in that thread rightfully point out, it's going to plateau at some point... or do you think companies are going to be giving us free drives in ~6 years?

Anyone saying this is having trouble with the concept of a logarithmic chart. No, hitting the bottom of the chart doesn't mean it hits zero, that's not how logarithmic charts work.

On this specific chart, the bottom of the chart is $5/tb, which you can clearly see on the appropriate axis.

I've got a 6-year-old who can, technically, get her point across. I don't think "they can eventually be understood after significant work on all sides" is the bar I want to hold the President to.

Coherency isn't a binary.

ZorbaTHut
2
89TB usable

Here's a chart of SSD prices over time . If I'm eyeballing the curve properly , this suggests that expected price in 2027, following historical trends, is a bit less than $13/tb, which itself is less than "20tb for under $300".

The chart is SSDs, not NVMe, but they're not that far off from each other. This actually seems pretty believable.

So, no, prices actually tend to drop faster than this article suggests.

(Of course, SSD prices actually went back up this year, which puts a kink in the chart, but it still ends up in roughly the right ballpark; I'm not going to be too upset if it turns out they're 10% off.)

ZorbaTHut
2
89TB usable

But . . . storage costs keep dropping? And have been doing so for decades?

I remember when I had a 40MB hard drive and it was both expensive and seemed incredibly vast. Now I've got over a hundred terabytes of raw storage and I'm buying more at under ten bucks per terabyte.

Yes, storage is probably going to be cheaper in 2027 than it is today. What would make you think otherwise?

My headcanon is that Claptrap and Bowser are literally the same person and you cannot stop me.

Sure, but this still doesn't solve the fundamental physical issues. You can't get acceleration out of nothing, and you can't "store" acceleration; if you have a ship meant to rendezvous with another ship, they must be traveling at the same speed at the rendezvous. And foreseeable ion engines are not suitable for boosting humans up to interplanetary speeds, they just don't have enough thrust.

It applies only to actions that are within the President's duties, which is pretty restricted. It's not like he can just go out and shoot whoever he feels like.

And, again, we had this before. This was always how the position worked; if the President is working within their job description, they have virtually unlimited power. But "working within their job description" is doing a ton of work here.

You can't "accelerate in orbit". Accelerating pushes you out of orbit; you'll have left orbit long before you reach interplanetary velocity.

Also, what's the point of the ion-engine craft in this example? If you have to accelerate the shuttle to interplanetary velocity anyway then you might as well just use the shuttle.

(With the exception of an Aldrin Cycler, but the point of an Aldrin Cycler is that you accelerate it only once and then just let it keep cycling; you gain huge amounts of living room for the cost of huge amounts of resources up-front and smaller amounts of extra energy on each transit.)

Example: Steamworld Heist, which is a turn-based strategy platformer.

The big problem with ion engines is that "gain speed more slowly" is an understatement. The acceleration difference is so massive that even for interplanetary trips, they end up far slower than conventional rockets. Fine for cargo, less fine for humans.

This is like saying "you can get terrible bikes at walmart, therefore all bikes are useless".

ZorbaTHut
1Edited

There's been at least one test of something vaguely similar. Take Stockfish, the world's best chess system, and run it on simulated hardware that's much, much older. What do you find out?

While it took the supercomputer "Deep Blue" to win over world champion Gary Kasparov in 1997, today's Stockfish program achieves the same ELO level on a 486-DX4-100 MHz from 1994.

The number they eventually come up with is that modern Stockfish is about 100-1000 times more CPU-efficient than Deep Blue was.

This isn't actually an abnormal result, from what I understand. A huge amount of our computation gains aren't actually from better silicon, they're from better algorithms; I've seen similar results in things like sorting and analysis, although it was years ago and I have no idea how I'd fine it.

Certainly GPT-7 or its equivalent will have a huge amount more processor power aimed at it than GPT-4 did, but it'll gain far more from progress in the field of AI than it will from simply getting more raw computation.

I think another generally-unremarked aspect is that it was really long. I've always wanted to read giant meaty books and they essentially didn't exist, then Worm showed up and it's like, holy shit, I've wanted to read this book for decades and I finally can.

Today it's just not as noteworthy that it's really long. By modern webfic standards, it's not even particularly long! But back then it really was a one-of-a-kind gem, providing a depth of story and character development that virtually nothing else could match.

(also, man, long books are the best, I am very happy with the world right about now)

I cannot fathom any reason I'd have posted there.

Because you happen to have clicked on it while looking at the Reddit front-page, or clicking Other Discussions, or someone linked to a thread there and you clicked it without even realizing you were on another subreddit. bestof, for example, links everywhere, and all it takes is one response and you're on the get-a-ban-message ticker.

ZorbaTHut
-1
Social Democrat

Do you disagree that happened?

Because "Obama argues that the US government can kill any US citizen" was a short burst of news back then, and then everyone forgot about it, and I don't get why people are freaking out about it again now when - I will reiterate - nothing has changed.