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

It’s not that simple: YouTube has a huge natural lock-in because videos use lots of bandwidth and Google’s advertisers are willing to pay for that.

Peer to peer systems struggle to offer a competitive a competitive experience because few people are able/willing to volunteer significant bandwidth or take on the personal liability for serving content which is copyrighted or otherwise illegal. They’ll work okay for a hugely popular video but fall off the cliff for the long-tail of niche interests and there aren’t enough people who care about it enough to make “YouTube but slower and less reliable” work in general.

Yeah. Essentially, YouTube is all but a natural monopoly given today's legal and technical environment. Unfortunately, it doesn't get the oversight that natural monopolies are supposed to have to prevent abuses like these.

That’s my take as well: regulation and liability for errors are the most realistic way to fix this. Even if ISPs were required to offer symmetric bandwidth I don’t see anywhere near enough people even being interested in serving other people’s content to make P2P work, which makes me feel old for remembering how cool the BitTorrent launch seemed before I understood the social aspects of the problem.

The reason you can't fork it is because the copyright owner wants to maintain control. That means that the content will never be on a blockchain even if we see the 8+ orders of magnitude efficiency improvements which would be needed to do so and when you've built the systems which would be necessary to hold and authorize access you don't get anything except expense and support burden from a blockchain.

Oh, my point was that we can and already have (ok, a DHT is not a blockchain, but it's the same sort of deal). I pay for Prime for the shipping, so I could stream it, but I still torrent the content because it's just more convenient to have the file.

The only thing that DeFi changes about this scenario is that it's going to get easier for me to pay the people who make the content available in the format that I prefer. When it's sufficiently mature, that'll mean paying the artists as well. It's just a shame that we're not there yet.

How many people even say they want micropayments? It seems like a relatively niche position even in my circles — most people seem relatively happy with the idea that their monthly costs are fixed and they don't need to think about how much it'll cost to open a link.

The proposed web3 model seems far more likely to be like those abusive micro-payment games with constant prompts to pay for something, with a page's worth of content moved behind successive paywalls since publishers aren't going to switch to a system which pays even less and you'd need to ask permission before charging someone.

This also adds problems with people who don't have money or aren't allowed to spend it (children, elderly, etc.) or the same scandals when those people are taken advantage of, not to mention the privacy impacts of effectively creating a super-cookie which can be used to track people all over the web.

You'd save an order of magnitude more water by cutting back agricultural and landscaping usage. If those apartment buildings are professionally designed, they're likely to be more efficient than random houses with ±30 years age on fixtures.

What seems to be working well in LA is building up buses. When I was out last June I noticed buses with dedicated right-of-way (physically separated) and that seemed to be running well during peak times, which is quite eye-opening when you count and realize that the bus which just went by was carrying more people than all of the vehicles in the 4 lanes of stopped traffic.

there’s only one truly dedicated busway here in LA (orange line in the valley). some other buses (like silver line) have sections of busway, but are mostly traveling on shared streets. there’s been a back-and-forth on dedicated bus lanes and some plans for more busways, but there is a lot of misplaced pushback from drivers on this (that it would increase traffic).

i do think that buses are a good lower-cost mass transit option, but without the reliability/security that comes with dedicated infrastructure, it will be hard to get more angelenos to rely on the bus for daily use. currently metro is focused on removing fares, but all that will do is lower their budget (modestly) and worsen service, which will be counterproductive to the goal of increasing ridership.

I was out around the G Line, and not for a long trip but it definitely seemed to be working there. The traffic complaints are always amusing because it really shows how people don't want to accept that the current system just doesn't work. I grew up in Ventura and have family around LA, and there have been so many road projects which made traffic better for maybe as much as 6 months until usage increased to match.

There's certainly a lot of iconography around that but I think that statement is dramatically overselling the strength and universality of that image. Millions of people have chosen suburbs since that was so heavily subsidized in the post-WWII era and limited supply meant there weren't good alternatives in many cases, but I would be careful assuming that this a deep preference rather than picking the default option.

In particular, I think you'd see very different responses when people were asked to choose realistic options: it's easy to say you like the idyllic image realtors promote with luxurious houses, perfect landscaping, happy kids at great schools, etc. but quite another to say you want to spend a couple hours a day stuck in an expensive car because there's almost nothing to do within walking distance, take on additional expenses (landscaping, lower density requires higher taxes for infrastructure, etc.), and otherwise live the actual suburban lifestyle for most people rather than the postcard version.

Proving the correctness of a program is a famously hard CS problem, not “bare minimum literacy”.

> As to how they would know if it's the real smart contract: they would see what it was via their wallet after interacting with it the first time.

In other words, the system is not safe to use. People will reliably be fooled into thinking that they're interacting with someone else — the difference is that if you go to amaz0n.com and enter your credit card info, your liability is capped at a low amount and will likely be zero because the regulated financial industry has a fraud handling mechanism better than “the people who profited from you buying their tokens will mock you for being phished”.

You're still required to keep records for things like tax purposes. More importantly, however, this is a service specifically designed to launder money which is going to look like a public declaration of intent to law enforcement types.

Since this costs money to use, most people are not going to use it unless they're trying to hide something so the big risk I'd worry about is similar to the risks of running a Tor exit node in your house. What happens when someone else using that service is investigated for some serious crime? Anyone who has transactions going to or from that pool is going to be under suspicion and it's really hard to _prove_ that you weren't knowingly helping them launder money when you have a public log of transactions involving the target.

This is like saying that Tor is specifically designed to buy drugs. It is designed for privacy, just like Tornado.cash is. Privacy can be used for many things.

People who really care about privacy don’t use public blockchains. Leaving a permanent public log of your transactions for analysts is reckless in general.

In this specific case, you’re talking about a service people have to pay to use. That lowers the pool of people using it considerably which makes techniques like timing analysis easier and increases the odds that your transactions will be mixed in with someone else’s criminal activity.

Plenty of people concerned with privacy use public blockchains - you don't need to dox yourself to use them, just need a private key. Unlike traditional finance, where you just have to hope that your PII data won't get leaked one day with all your transaction history.

Timing analysis is a real concern, that's why they warn you about it on the front page and ask to wait before you withdraw.

It's not just timing analysis — look at their long list of difficult measures needed to make these transactions private and ask whether that's remotely plausible for widespread use:


Very few people are so ideologically committed that they're going to pay extra and live with those constraints, which is a major problem for a protocol which is critically dependent on volume to deliver privacy and repudiation.

> Plenty of people concerned with privacy use public blockchains - you don't need to dox yourself to use them, just need a private key. Unlike traditional finance, where you just have to hope that your PII data won't get leaked one day with all your transaction history.

This is confusing a number of things. Most PII breaches are not the banks but the merchants who collect things like addresses because they need them for shipping or to satisfy legal requirements, and paying with a blockchain won't change any of those needs.

Similarly, very few people have a way to generate and spend a significant amount of cryptocurrency entirely for anonymous online services and will thus need to identify themselves for most transactions — a cryptocurrency exchange isn't exempted from Know Your Customer, companies which have to deal with abuse are going to want to prevent sock puppets or shell accounts, airlines aren't going to lose interest in checking your identity, buying a house without showing where you got the funds is going to attract a lot of attention, etc.

That link to the real world is the common reason why these promises don't pan out. In general, ask yourself how a particular activity would go if you showed up with a suitcase full of cash and refused to say where it came from. Cryptocurrency will be exactly the same in all but a very few cases.

Timing analysis is the main thing to worry about, if you're just looking to get some anonymous ETH such that people looking at the blockchain can't track it easily. Those difficult measures are not a requirement and can be ignored depending on your threat level.

And I don't know why are you talking to a strawman about a suitcase full of cash, etc. I just said that Tornado.cash offers privacy and that privacy is not always used for evil things.

I'm skeptical that it's all that much better for any group of user. I search for normal things all of the time and it doesn't do a good job at returning relevant results for things like businesses, recent movies / TV shows, etc. and natural language syntax doesn't make the results less bad.

What it looks suspiciously like to me is a lack of an effective feedback loop for user frustration — if it takes a number of queries to get correct results or someone doesn't stop using Google search entirely, this would be easy to confuse with improved engagement and I'd especially believe that managers whose job it is to get a number to go up are not in a hurry to question whether that growth is meaningful.

It looks like a turn of the century art project or album launch site at first: there's nothing on the page mentioning Square until you click on the arrow at the bottom which does some animation before showing the copyright notice and the pseudo-Python on their home page doesn't look like a $100 billion dollar company or a tech company (it'd be valid syntax).

There are effectively three tiers: managed storage services like S3 (object storage), EFS (NFS NAS), or FSX (clustered filesystem) where most of the decisions are made for you; the mid-level EBS (SAN) service; and storage-optimized instance types with local disks which you manage.

This custom SSD hardware family is what powers the EBS (cloud SAN) service, which allows you to pay for the performance level you need, where they give it both higher absolute performance and [now] better worst-case latency.

This announcement is saying that you can now get your own instances with the same performance characteristics for situations where you need better performance than a SAN can deliver and/or the robustness benefits of using per-node storage rather than a separate networked service.

The other part of this announcement is the implicit message it sends about the competition: they're telling everyone that their storage performance is more consistent than their competitors and increasing the number of areas where they can say they have an option which a competitor does not. Noting that this was driven by the EBS storage team is also a reminder that they have more people working on lower-level infrastructure problems than you likely do.

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