Hacker News new | past | comments | ask | show | jobs | submit | more gibybo's comments login

>We can think about "what" the initial state is ? Is that initial state self contained ? or was it under effect of something else ? Will the computer "generate" the initial state, or the programmer (aka god) has to explicitly hard code it carefully to create an apparent stable universe ?

They have infinite computing power, just simulate all possible initial states?


>25/5Mbps satellite circuit

>data cap of 80 / 150 MB

So if you had the full 25 mbps available, you would hit your peak cap in 26 seconds? Do you only use it for email/text based applications?


Yikes, and thanks for picking that up.

I meant 80 / 150 GB -- I'll correct my earlier post.


>Pharmacy is about 10% of total healthcare costs

That's good to know and something I've been curious about. Do you have a source that breaks down the other 90%?


Ok, so I work in healthcare in the US and I read a lot of papers and think "well yeah, but that's not very actionable information".

The best analysis I've found so far is from McKinsey.[1] They took a very clever approach to benchmark US healthcare spending to other OECD countries. They adjusted for US GDP (richer countries spend more on healthcare), then broke down costs by category and figured out how much more/less the US spent compared to OECD countries. Very eye opening if you check out the graph at the bottom of page 4.

Findings: 1) out-patient care is driving ~70% of the extra cost, 2) in-patient costs are ~10% higher than OECD, 3) the US spend 50% more on drugs, but it's a small part of overall spending, 4) administration is 200% higher, but again, it doesn't add up to that much, 5) the US spends less on long-term and home care and 6) the US invests 50% more in healthcare, but it's not a large absolute amount.

[1]https://www.mckinsey.com/~/media/mckinsey/dotcom/client_serv...


>Ask your representatives to make stronger labor laws.

Lawmakers don't do things when one person asks, they need significant public pressure. Publicly showing displeasure with how a company operates is usually the first step for that.


I thought electric charge in conductors already moved very close to C? https://en.wikipedia.org/wiki/Speed_of_electricity


Exactly. If we're pedantic about it, a charge signal moves at the speed of light. But the signals are what computers use.


Only if you consider 70% or so to be close. There's some room for improvement over copper wires. Now, if there are any physicists here who want to jump in, I have a question about that. I heard waveguides are dispersive, would sending pulses of light through tiny channels slow it down as well?


Some people working in optics say it works "at the speed of light." That's true of course ... but the speed is no different from using copper.

Unfortunately something like 0.7c is about the fastest speed of EM wave propagation in an optical waveguide or along a copper waveguide. Another comment here gives a slightly faster example with n=1.3, which is maybe achievable in some kind of polymer. Or in highly purified water, for what it's worth.

You can get a mild speedup, 40% or something, by moving to free space. But that is an unbelievable can of worms, taking all the signals out of the waveguides and somehow still getting 1B signals going to the right place. The 40% speedup doesn't remotely pay for giving up solid state waveguides.

"Dispersive" fortunately doesn't mean a meaningful slowdown. It just means that a transmitted bit will travel at a range of slightly different speeds. If it goes very far, the shape of the pulse will get messed up. But that's a problem people are already pretty good at solving.


But light in a dielectric medium also travels slower.

If n=1.3 -> v = c/1.3 = 77% of c


Not a physicist, but you may want to look for 'hollow core fiber / photonic crystal fiber'.

Where some are said to reach up to 99.x % the speed of light in a vacuum.


Laser physicist here, did some work with PCF: Expensive as fuck, highly polarization dependent, almost impossible to splice.


Nonetheless it feels like at least every few months or even weeks a new announcement appears in pop-science sites like eurekalert and phys.org. Feels a little bit like the always around the corner next big battery tech.

Most fascinating thing i've read years ago they'd be the prime candidate for manufacturing in space, because real vacuum.


Out of curiosity, how would you prove you didn't give consent? Can't the cop just say you did?


They can certainly say you consented. But most likely it's going to get recorded. If it's not recorded and the LEO had a body cam the DA, judge, and potential jury are going to start questioning the LEO why. That takes the heat off the defendant and puts the entire case into question.


If you are quarantined, your food selection may be severely restricted. I'm sure you'll be able to get food, but it may not be the food you normally like to eat.

Additionally, if the virus is rampant in your area, you may not wish to go to grocery stores and risk exposing yourself.


Not a single wafer, but the sensor built for the Large Synoptic Survey Telescope is a ~64x64cm 3.2-gigapixel CCD, made of 189 4x4cm chips on the same sensor bed[0].

I suspect the cost was in the millions of dollars for the one-of-a-kind sensor, but I bet you could get a great deal on a second one since they've already done the R&D!

[0]https://en.wikipedia.org/wiki/Vera_C._Rubin_Observatory#Came...


Absolutely, in fact it's pretty easy and is routinely done by amateur and professional astronomers alike. We've had satellites, airplanes, and meteors obstructing astronomical images ever since we began taking them.

A very simple and common way is to stack all of the static images aligned on top of each other such that you have a set of values for each pixel corresponding to the same region of sky. Then sort the values of each pixel by brightness and keep the median value for that pixel. Or throw out that top and bottom 10% and average the rest, or throw out the top and bottom N, etc.

This is a standard feature of essentially all astronomical image processing software and has been for a long time.


Sure. But increasing the number of frames with tracks increased the number of datapoints you need to throw away, reducing the effective time on target and therefore sensitivity of your observations. Astronomers are not idiots. They know how to deal with obstructions. But that is not free of cost. Neither in observational limits nor computer time nor required brain power.


Sure it's not without cost, but in most cases the cost is not very high.

Remember you don't need to throw away the whole frame, only the pixels that were obstructed. One track across a frame will obstruct less than 1% of the data in that frame.


Why did the factory owners pay for (or at least allow) those sorts of books to be read? Wouldn't they view it as against their interests?


I think most lectors were paid by workers pooling some of their money, not by a factory owner.


The capitalist didn't always know what was going on in their factories. Besides that, workers used to pay somebody to read for them, or they would select someone from their own group.


Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: