MystikIncarnate

Fair enough. I haven't used spectrum, so I have no opinion. I'm not in the right country to subscribe to their service, so there's that.

Have a good day.

It's that recent. Jeez. Feels like it's been a thing for months.

Wait, what day is it? WHAT YEAR IS IT?

OH GOD

IMO, the post is centered around proton VPN, and since that's a public VPN service, it's the focus of the discussion.

Private VPNs are a very different story.

Yes and no.

Modern HTTPS connections send the URL you are connecting to in the initial hello, so the remote webserver knows what security certificate to use when you connect. A lot of web servers host multiple sites, especially for smaller webpages, and so it doesn't assume that since you connected to that specific webserver, that you're connecting to the site that the webserver is hosting, even if it's only hosting a single site.

This can leak the data to anyone sniffing the traffic.

You can also determine some traffic by IP address, this is for larger web services like Facebook, youtube and other sites of similar size. They load balance groups of IPs for their traffic, all are serving the same data. So if you connect to an IP that's owned by Facebook, for example, then your actions can be easily derived.

Since the connection is still secured by TLS, the content can't be deciphered, but the location you are going to absolutely can.

It really depends on a lot of factors.

All ISPs are legally obligated to forward that shit to you. The alerts are not from spectrum, they're just relaying the information.

Right now, copyright owners do not have legal permission to find out who you are directly without a court order. They would only seek that information if they were planning to file a lawsuit.

Media companies know, from the Napster incident, that such actions can backfire stupendously. It's rare that they even bother anymore. I can go into detail on why, but I'll leave it out for brevity.

So they send the notice to your ISP, who is legally obligated to match the information on the notice to the subscriber and forward the notice to you.

For many, this goes to an ISP provided mailbox, which most people ignore the existence of it. Clearly spectrum operates differently.

The notices are from copyright holders who have no idea who you are, and can't determine that information unless they intend to sue you. So those can be, for the most part, ignored.

It's not your ISPs fault that you got those. They couldn't give a shit less about what you do on their service, or what you download. They just want you to pay your bill every month and keep the gravy train rolling.

When speaking to the overall system, there are always inefficiencies with all forms due to the conservation of energy laws.

Similar arguments can be made regarding batteries, as resistance in the wires that connect the cells in a pack together waste power as heat. While overall this may be minimal, the physics provide hard limits here. Unless a superconducting material is made commercially viable without needing to be super cooled, these limits will always be nontrivial.

My entire point is, battery tech has reached a high level of development and there is significantly more we're trying to achieve with the technology (whether solid state or otherwise), meanwhile, I would argue that hydrogen hasn't even reached the same level of development as battery technology, yet everyone seems to think it's a dead end.

It's hard to argue with the energy density per kg of hydrogen as a material. It's possibly one of the highest specific potentials of existing technology. What we should be doing is trying to create power from that with as few losses as possible. Fuel cell technology was, in my mind, the first real push in that direction, when it didn't immediately pay off, we gave up. Meanwhile, alkaline and cadmium based batteries were much worse, but we used them, and continued using them for decades before lithium based batteries became more commercially viable.

I see battery research as looking for the last, most efficient type of battery, while hydrogen isn't even half way through the possible research we could do on it. Forgetting hydrogen, while it's in the infancy of the research, for batteries that are very nearly as efficient as physics allows for, to me, is doing ourselves a disservice as a society.

I have no idea what further research into hydrogen will yield. Maybe you're right and it's going to go nowhere, maybe not. We don't know unless we keep trying, same with batteries, same with kinetic storage (flywheel/gravity systems), same with thermal storage..... There's just a lot of material science we can experiment with that wasn't really something that was possible before now.

I still think it's worthwhile, clearly you disagree. I appreciate the discussion either way.

Have a good day.

H2 from natural gas is more efficient, but obviously creates pollution. Because of the relative efficiency and the prevalence of natural gas in society, most companies have gone to natural gas conversion to hydrogen, as it's easier to implement, not because it's greener.

To touch on it, when I'm discussing economics, I'm talking about the discipline of economics, not specifically the economy. The money economy is only concerned with the dollars and cents of everything, economics as a discipline, considers all factors, both in and out, and the adverse effects of everything, both financial and sometimes not financial (since nonfinancial effects can affect the future financial viability of a system).

I'll be clear, storage isn't the debate on hydrogen being inefficient. Hydrogen storage is more efficient than most other storage systems. The materials are minimal, a pressure tank with the appropriate seals and safeguards, and the tank can output 100% of the hydrogen that goes into it. There's no concern with cycle life, as the system can cycle infinitely as long as the structure of the container isn't compromised. The waste produced when a storage vessel is no longer suitable, is essentially metals that can be fully recycled or otherwise reconstituted into other items without any degradation in the quality of those items, with few exceptions.

The discussion is entirely around how hydrogen is created, and how it is converted back to whatever energy format that is desirable, such as electricity. Coming from electricity, electrolysis is about 70-82% efficient, with 1kg of hydrogen, which has a specific energy density of 143 MJ/kg needing about 50-55 kWh of electricity to create. The most inefficient part of the system is conversation back from hydrogen to electricity, where internal combustion style generators are common (basically a slightly modified natural gas generator), but less efficient than fuel cells. Fuel cells generally have 40-60% efficiency.

Batteries on the other hand have much higher efficiency, but never 100%. Since they're generally not self regulating, systems for battery management are required. Charge controllers and voltage conversion (or inverters) reduce efficiency further, but generally battery systems are considered to be better than 90% efficient. The downside with battery systems is the relatively short life of the battery and the large amount of waste produced, in comparison with something like hydrogen.

Hydrogen can achieve much higher energy density and the container weighs next to nothing when empty, while batteries weigh approximately the same whether charged or not.

My main argument for hydrogen surrounds the fact that we're pretty close. 80% efficiency in hydrolysis and 60% on fuel cells, with storage being significantly cheaper on materials and significantly better with cycles, with much less to recycle when the system is replaced or otherwise decommissioned. You can pack a lot more energy in the same volume of space using hydrogen compared to batteries because it can be significantly pressurized to several atmospheres.

There are benefits here that batteries simply cannot match. If we can get the fuel cells and electrolysis to a level that's comparable to batteries with efficiency, then hydrogen would really become the better option.

With over 8.2 billion people on the planet, we certainly can research all of these options at the same time. Only a very small fraction is even doing the work right now. That number can increase a lot, but we choose to pursue what is financially profitable rather than purely looking towards scientific discovery. Capitalism at work.

If companies can't sell it, they don't care. So it doesn't get done. We should do it anyways because there's potential here.

That's fine, Qualcomm has followed suit, and Samsung is doing the same.

I'm sure Intel and AMD are not far behind. They may already be doing this, I just haven't kept up on the latest information from them.

Eventually all processors will have it, whether you want it or not.

I'm not saying this is a good thing, I'm saying this as a matter of fact.

IMO, there's two main factors at play. First, the speakers in most stores suck. They have to buy them at volume (quantity, not loudness), and install them everywhere. The primary reason they have them is for paging, so they can make announcements and request that people go places. Music just gives the speakers something to do while not doing announcements.

Due to the amount of speakers they buy, and their primary purpose being for announcements, they don't exactly buy high quality speakers. If the store has existed for a long time (maybe 10+ years), then it's likely they're analog, so the quality is also affected by the amps they're using, and the cables, etc.

As long as the system can still do paging/announcements without issue, the business really doesn't have any reason to spend money on upgrading it.

For the most part, most companies have connected these to some kind of satellite radio or music streaming system (like Spotify, but more business centric). It's just plugged into the ancient sound amps for the analog system, often by someone who isn't an audio expert, so levels are often all over the place, sometimes to loud and blown out, other times too quiet and details in the music are too quiet to be heard.

As long as the speakers still perform the announcements/paging that the company requires, they don't care if the music sounds bad.

There's a lot more to say on it for contributing factors, but the main drivers for it are not to play music. With the shift to digital and everything needing to update their music providing device, coupled with untrained people doing the connections for the new music solution to an ancient speaker system, it's unsurprising that it sounds like garbage.

This is the way.

Both apple and Google have integrated machine learning optimisations, specifically for running ML algorithms, into their processors.

As long as you have something optimized to run the model, it will work fairly well.

They don't want to have independent ML chips, they want it baked into every processor.

There's a lot I can say here, but to be terse, the money paid into (un) employment insurance is more than what is paid out normally, since some people will pay for it all their life without ever collecting, that money isn't just stored indefinitely, it's used for other things.

As a result, if a large portion of the population suddenly find themselves without work, the system will be unable to sustain itself, whether "short term" or not. All systems that rely on EI overflow funds would suddenly have a deficiency in their money flow, and considering they the people pay most of the taxes while billionaires and corporations get tax breaks so that they pay nothing, the entire social support systems would collapse quickly, as the country plunges further into debt, devaluing the countries currency.

The entire economic model is built upon things maintaining and continuing mostly as they are, pull any thread too far and the whole thing unravels.