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

Spotify hi-fi was announced a few months ago and is launching soon.


Rough for Spotify - since Apple is including theirs at no extra cost, it becomes tough for Spotify to justify charging extra for it. So it goes from a feature that makes them money, to a feature that instead costs them money in engineering resources and bandwidth, before they even release it.


That depends if you believe Spotify users will switch to Apple just for that feature... if they won't, Spotify can likely still charge extra for it. Folks are pretty tied in to ecosystems at this point: if you've bought audio hardware that supports Spotify but not Apple, you're not going to switch.


It's not a good look for them. It looks like they are afraid that Spotify will solidify its position.


The EU (and Japan and Australia) has a range of rules now for increasing pedestrian safety, including softer materials on the front bumper, hoods that give so that they cushion the impact if a pedestrian is hit, and limiteing sharp protrusions / edges on the front of the vehicle. The new G-wagon, for example, has turn signals that flex down into the vehicle on an impact. https://www.autonews.com/article/20120423/OEM03/304239967/eu...


" It is a very feminine point of view to believe that one has intrinsic, objective value."

I think you need to take a while and consider why you believe that statement is true, and how it might be impacting the rest of your view on this topic.

Have you contemplated that you might be in the wrong here, and that your approach on this topic is very, very heavily based on your first person view?

It's extremely hard to control for the success of corporations vs. any single factor, much less "gender balance". For example, more male-founded startups get funded - but it's been shown that's in large part because they're male (not to mention their investors usually are), not because of any inherent merit of their business. Similarly, consumer startups often have an easier time getting funded because they're easier for partners to explain the rest of their firm, but it doesn't actually mean they're better investments than an niche enterprise play that's harder to explain to a layman.

More starts = more exits, more role models => more male founders and more all-male startups. Nothing in that cycle actually proves that that men are better at founding or running tech startups, and say "show me the data" is a poor response, given that we don't have an alternate universe where there's no gender bias feeding into those patterns.


I’ll assume that you agree that objective value is not possible to define and that therefore no one has objective value. Are you taking issue with my view that women are conditioned to believe that they have inherent, objective value? Why do you think Instagram is so popular with them, where it is sufficient just to post a photo of oneself to be valued?

Also what do you mean by inherent merit? It doesn’t exist, just like objective value. It doesn’t matter that a study has shown that men are hired or receive funding because they’re male over inherent merit, when the study has already necessarily failed at defining inherent merit.

You’re either being intellectually dishonest, or you need to read what I wrote again.


Yes, I completly disagree with your sweeping characterizations of entire genders, on both sides of that discussion.

A key part of the problem is that society often propogates such sweeping generalizations to children, reinforcing existing biases based on stereotypes rather than individual merit. This is part of what diversity policies are intended to address.


Language and the market are generalisations. These are the only ways in which humans can interact. The market is neither right nor wrong. It is the status quo. Women put more selfies on Instagram than men. It's a generalisation and also a fact.

You're still propagating the idea that individual merit exists and can be defined. It can't. If you think that hiring based on your subjective, generalised idea of merit would lead to grater gains, it's up to you to show it with real results. You clearly stand to make a lot of money. Until then, the status quo remains.


Medical bills are (according to some studies [0]) the leading cause of bankruptcy in the US. So not only do you have to recover from a serious accident or illness, and maybe the loss of work time associated with it, you also are left with either a crippling level of debt or the total loss of your assets, plus a lien on future income.

Additionally, the less money you make, the worse coverage you probably can afford, leaving you even more exposed.

You're absolutely right that it's insane that Americans talk about this as if it's just the only way things could work.

[0] https://www.cnbc.com/id/100840148


It's an interesting question, but rationally it seems like BTC should decline during a market crash / recession. It's arguably a way to store value as other assets decline, but given that it has no yield, the rational thing to do after that decline is over would be to cash in your BTC and use it to purchase those other assets that have declined, such as real estate, bonds or stocks, and now have either attractive yield or a strong potential for future asset growth.

(And you see this kind of rebalancing effect in general when a major asset class declines - eventually it pulls down other, unrelated asset types because as it goes down in price, it becomes a relatively more attractive investment.)


I wrote a bit about my opinion in a reddit thread a while back. It is here if you are interested, but a fair warning it will probably be quite an unpopular opinion with the hacker news crowd:

https://www.reddit.com/r/BitcoinMarkets/comments/7xq7mp/comm...


The problem isn't that it's unpopular, it doesn't seem (from the post) like you have some kind of underlying economic or financial principle supporting your argument.

* why would it have an inverse relationship with other assets? It's not a short. It's still valued based on purchase price vs sale price, adjusted for risk. If risk has risen and nothing has changed about purchase or sale price, why would it rise?

* if Bitcoin was at a market-clearing price before a downturn, and other assets are now much cheaper, why would BTC then be comparatively more attractive in terms of expected investment returns?


Well can we at least agree that no one knows with absolute certainty what will happen to Bitcoin and other crypto currencies during the next recession?

The reason I believe it will be inversely related is because it is still a relatively new market whose market cap is minuscule compared to other established markets.

During a recession people will naturally look for other places to put their money and gold will be the obvious choice. Bitcoin has all the same properties as gold except it is cheaper to acquire (fees), cheaper to store, has a fixed mining rate (with gold more supply is created when the price increases which drives the price back down), and has only 1% of gold’s market cap. The way I see it, it would be silly not to at least hedge a small amount of money into BTC/crypto.

The other economics reason is pure supply and demand. The supply of Bitcoin is shrinking every day as more people decide to hold it long term. Many more get lost or stolen. Every couple years the block rewards get cut in half as well.

All of that said, the rise and fall in the last year has definitely shaken the public’s confidence so it may take a long time before new money starts flowing into it. All of my timelines are years down the road. (10-20 years)


Obviously, nobody can predict the future, but the whole point of talking about what's likely to happen is to make reasoned guesses based on the best available data and economic models.

The nature of a hedge, in particular, is that you do it before a market downturn, not after. Once the market has dropped, you want to unwind that hedge, which in the case of gold or Bitcoin means selling. If you look at gold's performance during recessions since we unlinked it from the dollar, it's as likely to decline as rise. That makes it a poor hedge against recession.

In terms of supply and demand, it's true that supply is constrained, but that's only half the equation. Demand is extremely volatile, as there's no inherent "consumption" of crypto currencies (aside from lost wallets and other breakage), not is there any production occuring that requires btc to continue. The demand is entirely composed of people buying it with the expectation of a future sale at a higher price... that's speculative demand, and it's not a stable or dependable type of demand.


Is the supply really "limited"? Sure, there's finite number of "coins" that can be generated. But unlike physical objects, there's not really any special property of that unit. Also, you can just generate a new currency of more units and similar utility (as has happened many times recently.)

Objects in the real world have utility that's directly linked to their unit value. That's not necessarily true for Bitcoin... if it goes up 10X, I can just use 1/10 as much and get the same exact transaction result. So why is it "limited" from a supply/demand perspective?


> Objects in the real world have utility that's directly linked to their unit value.

But that's not true of money...whether it's official fiat money, many local currencies, or even gold, which has some utility but not near enough to support its total value.


I'd say that an analogy with diamonds works: There is an infinite supply of lab-made-diamonds, however they are not fungible with real diamonds. Them being substitute goods, lab-made-diamonds may have negatively impacted the price, but real diamonds are still considered scarce.


> There is an infinite supply of lab-made-diamonds, however they are not fungible with real diamonds.

Ada Diamonds sells lab-grown diamonds and jewelers can't tell https://news.ycombinator.com/item?id=17228369

De Beers admits defeat over man-made diamonds https://news.ycombinator.com/item?id=17183603


There is nearly an infinite supply [for all practical purposes] of mined diamonds, for that matter. And lab-made diamonds are real, and aside from the early versions being a little too perfect they are indistinguishable from mined diamonds.


There's an infinite supply, but there's non-zero cost associated with accessing incremental units in that supply. (And some would say diamonds are an artificially constrained market anyway.) With crypto, there's arguably an unlimited supply at zero cost.


> With crypto, there's arguably an unlimited supply at zero cost.

I like that one. Makes it seem like its a perpetual motion machine and we all know how that ends.


The mining cost is actually quite high for bitcoin, there's a friction involved that's being overcome.


They pretty much are fungible with natural diamonds. High quality synthetics can't be told by eye from equivalent quality naturals.

If they're < 1ct., it doesn't pay to have them analyzed.

"Real" diamonds have always been considered scarce, but only because of marketing. De Beers, AlRosa, etc. have very large reserves of natural diamonds. Over time, they will become genuinely scarce, as the industry agrees that economically recoverable deposits of natural gems will decline after about 2020.


This analogy isnt complete though because it doesnt take into account network health. Its easy to make a cryptocurrency, but hard to build a network large enough to defend against various attacks. There are many, man coins out there suseptible to double spending - bitcoin is the most secure, and this security adds some value in itself.


Maybe you're jumping to the conclusion too fast. Most coins don't have publicly declared vulnerabilities (ie there might be many 0days). But it might be untrue that bitcoin price is not being manipulated (an attack itself on the network) based on multiple articles on HN alone.

So bitcoin might not be too unique or valuable.


Diamonds aren’t scarce. They are just unevenly distributed, and a few cartels own the means of production. The resale market for all but the most valuable diamonds is complete crap, because deBeers keeps bringing enough new diamonds to market every year to meet demand.


while the supply of digital coins is infinite (you could fork btc, or any other chain, or create different coins that could then be forked) the supply of Bitcoins is finite & knowable. this is a component of the argument by "maximalists" who suggest that alternate coins/tokens are bullshit by design, and anybody who thinks they aren't is necessarily in favor of inflationary money.

> Objects in the real world have utility that's directly linked to their unit value.

Could you explain what this means to you? I don't understand what you're saying


Bigger diamonds are more desirable than larger diamonds, not because of the price, but because they look better and are sparklier and harder to lose and all sorts of other real-world metrics.

If I want $1000 of bitcoin, I don't care whether that value is in 1 BTC, 100BTC, or 0.0001BTC. They are otherwise identical to me.

Therefore, the fact that the total of all bitcoins in circulation can't exceed 23 million is irrelevant unless you care about the value of 1 BTC, which is only the case if you are trying to sell them for more than you paid.


What he said above: unless you have a long or short position, you don't care whether you get a full unit or some fraction thereof.

There are some elements of behavior economics that counter this - people like to buy lower-priced coins and stocks because they get "more" of them, from a unit perspective. But rationally, there's no reason to consider 1 BTC @ $10 any different from 0.1 BTC @ $100, hence the number of "units" seems potentially irrelevant.


but is the same not true of trading USD for EUR? If I want $1000 of EUR, I don't care whether that is 1 EUR, 100, or .01. The value is in the ecosystem.

Bitcoin is still at the beginning of the beginning of any sort of ecosystem. Everyone expects way more from it than is possible. It is/will deliver on the things it is designed to.

You're correct that I don't care about the value of 1 BTC (or 1 EUR) unless I can sell it, but you imply only selling it for USD again. When it is widespread enough to be a currency, you can 'sell' it for goods and services.


Yes, but people argue that the finite supply of Bitcoin is a fundamental advantage vs. currencies. The supply of a currency isn't really infinite, but it's quite flexible when you consider that giving credit effectively "creates" more currency and expands the monetary supply. (I'm not clear why BTC won't eventually have this issue as well if folks start lending / borrowing it.) All the BTC really avoid is increases in the monetary supply via the printing of additional currency, which is arguably a feature of traditional money, in that the central government can intentionally tighten or ease the monetary supply to help manage economic volatility.


while the supply of digital coins is infinite (you could fork btc, or any other chain, or create different coins that could then be forked)

This is bull. It's not zero cost to fork a cryptocurrency. It's not zero cost to mine. The supply of a digital good might well be very large. However, it will most certainly be finite.


You could fork the cryptocurrency and let miners decide their reward.

Then you're limited to "What's the biggest number that a miner can name in the finite amount of memory available?", which has the ultimate limit of a Busy Beaver function.

BB(n) for n > 1500 behaves like an infinity in many important ways.


Then you're limited to "What's the biggest number that a miner can name in the finite amount of memory available?"

Then just apply my original argument. It's not zero cost to mine. It's not zero cost to fork. The supply of cryptocurrency will be very large, but it will be so many x orders of magnitude smaller than BB(kerjillion), that x itself might be larger than the number of every cryptocoin ever made and every cryptocoin that ever will be made.

The useful thinking is about the computational limits of our current civilization, not infinities or redonkulous numbers.


You don't need to fork it, or mine it. You just use a smaller amount.

My point was that other traditional "store of values" have inherent value based on their utility per amount, which means that the finite supply has implications for people who want to use it. Bitcoin doesn't have that - it's purely useful in terms of what you can trade it for (much like a dollar.) Hence the fact that there's only so many "units" is kind of uninteresting, because a fractional unit is no less useful.


Interchange rates on Visa are >2%, and there's definitely cards that offer 2% cash back on everything, with the exception they generally bar "cash-like" items like cash withdrawals and t-bill purchases, to avoid holders just churning purchases for the rewards. (There used to be a way to make tons of points for buying like $10K of t-bills on your Visa, then immediately reselling them for $10K, because the Treasury wasn't charging the interchange fee.)


I suspect this is less about the credit risk of the cardholders than the risk of unrecoverable fraud, although there's also the theoretical risk of some kind of fraudulent conveyance by the card holder.

Fraud: I steal someone's credit card and use it to buy crypto, then move that currency somewhere unrecoverable. The card company is unable to recover anything. (This can be mitigated by recovering from the exchange in some cases.)

Fraudulent conveyance: Holder maxes out all their credit cards buying crypto, hides it, then claims bankruptcy. Card companies can't recover anything. (Yes, you could do this with something like gold, but it would be a lot harder to hide and sell later on.)


I ride both ebikes and scooters, and while the ebikes are a better way to get somewhere fast (and I think, overall, safer) the scooters are generally a much more pleasant ride - you're standing up, you just kind of surf along, and you don't have a 30lb bike to deal with... they're magic-carpet-like.

The exercise benefits of both are... questionable. Even just cycling as a commuter on a regular bike isn't very intense exercise, and on an ebike it just seems like you're pretending to pedal. I can't imagine it has much health benefit aside from being better than sitting in a car.


At least in SF, they protect the bike lanes sometimes - no right on red, at least, when there's a protected bike lane.

Re filtering, is that technically against the rules? It's a single lane at that point, but in CA it's legal to share the lane if there's room. Seems like a cyclist or motorbike has the right to lanesplit and move up.


> Seems like a cyclist or motorbike has the right to lanesplit and move up.

Yes, but in this case you filter on the driver's left. You don't filter up in the 10cm between the curb and the car when the driver is signaling right; it's a dick move and eventually you're going to get hurt.

http://www.sfbike.org/news/bike-lanes-and-right-turns/


Fair enough, but honestly this seems like far less of a problem than the large fraction of drivers who don't know how to turn properly in the first place, and just make that right turn without having merged into the bike lane.

I commuted in SF for a year, and I'd guess 75% of cars don't know why the bike lane starts getting dashed before an intersection.


Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: