HN user

nmrm2

1,122 karma
Posts0
Comments534
View on HN
No posts found.

There are plenty of examples where private companies control water distribution without government assistance or oversight if you look past the USA. We don't need fanciful theoretical speculation to know what happens in practice.

(note: not continuing with this thread because holy fuck hivemind downvotes.)

The popular narrative that, unchecked by government, a business will eventually rise to absorb everything is not borne out by practice.

I get your point, but there are exceptions. In particular, utilities come to mind. All locations I've lived except 1 provided exactly one option for high-speed internet. I don't see that changing in most parts of the US any time soon. Ditto for water and electricity, depending on locality. Even temporary monopoly abuse in either of those two sectors can have devastating humanitarian impacts.

But at that point, what are you offering? (Genuinely curious. Capital? Taking on the risk? Insight/market research?) And is that worth the overhead that goes into your pocket?

Sharing economy type stuff makes complete sense when it's something you share often enough to want first dibs, but not often enough that renting it out from time to time is an inconvenience. But if you're renting it out full time then I'm not sure I understand how the incentives line up. Seems like if they notice you're doing well they could just swoop in and under-cut you next time the contract is up.

Sure.

But the people with the skill-set and interest in doing that are going to do so anyways; I don't see how anti-theft tech would convince that type of person to shell out an extra few hundred for a bike. They would just buy something and install it themselves.

I just don't see the market for integrated anti-theft. Seems more like a custom component type of thing.

Especially as the price approaches four digits, "what if this gets stolen?" has to be near the top of most potential buyers' list of concerns and probably chills sales to some degree.

If you're spending that much on a bike then you're cycling regularly (communter and/or every weekend type). You're not going to ride a shitty bike if you're riding that often. Bikes aren't like cars -- a Corolla will get you from point A to point B just as well as a Cadillac, but a doubling in the price point of a bike (esp. 500 to 1000 range!) has a massive impact.

Of course there are people who buy racing bikes and ride them a few times a year. But those people rare enough that it doesn't make sense to spend the money designing and stocking a product just for them. And anyways, the guy on the sales floor is probably going to be the most important factor in their purchasing decision.

...by which I mean the experiment itself without the "fair" elements

I don't think the author or anyone else disagrees... the article was specifically about science fairs, not science projects generally. It's right in the title. And reiterated in the portion of the text where the author juxtaposes their childhood experience with their child's experience.

This isn't someone being born with silver spoons in their mouth. This is regulation thoughtlessly applied with no thought about the costs.

It's both (which gets to the heart of what I consider the most compelling arguments for decreased regulation). Whenever there are bureaucratic barriers, knowing the right sorts of people works wonders. You might even say it's a privilege...

Well, you can repair the damage from plastic bags.

Maybe. Maybe not. Maybe you can, but the costs to the environment involved in doing so are prohibitive. In any case, I don't think it's totally obvious that this is actually possible ATM.

If your model contains infinities, it's automatically unhelpful; that takes you into Pascal's Wager territory

Yeah, well, reality > model. Unless we colonize space, destroying earth = infinity. So the implication of your statement is that pricing is a bad model for certain types of environmental damage. However...

would justify draconian measures against even trivial risks of too many plastic bags.

...the realistic impact of most types of irreversible environmental damage has no known bound, but is not infinite.

Difficult-to-bound unknowns also make it impossible to quantify damages a priori at the level of detail necessary for pricing. Again, reality > model, so accurate pricing is a bad mechanism.

I think there are reasonable arguments against plastic bag bans. But trying base policy on an accurate accounting of the cost of long-term or irreversible environmental damage is a fool's errand in many cases.

(Also, note that my point is that in general, pricing environmental externalities is weird. Plastic bags aren't even nearly the best example of this, but are illustrative of the most common solution -- don't price, disincentivize.)

Do you believe that, given a million dollars for every bag used, the government could clean up this supposedly irreversible damage?

You're missing the point.

If you charged a million dollars for every bag used then basically no one would use plastic bags. Or at least few enough people that the aggregate environmental impact would be negligable. As it turns out $5 or $10 would probably work as well as $1M.

The whole point is that it's totally impossible to come up with realistic estimates for something like plastic bag waste. So you set the prices high enough to disincentivize their use.

The purpose is the disincentive, not actually putting accurate prices on externalities. Confusing these two things is the source of the confusion in SilasX's original post.

You seem confused about the point of plastic bag taxes/bans. The purpose of plastic bag taxes isn't to "price-in" an externality; rather, the purpose is to fundamentally change consumer behavior.

More generally, the language of economics ("pricing-in externalities") doesn't make sense when discussing irreversible damage to the environment:

1. When you can't repair the damage, the idea of "pricing-in" externalities is just confused thinking. (You can't buy a new planet and not all environmental harm can be reversed by spending money. At least for now.)

2. For any consumer behavior that causes irrevesable damage to the environment, the cost is at once too large to quanify and also too small to notice. (Many forms of pollution have highly nonlinear effects on the environment that are often impossible to quantify, especially at a global scale. I have no idea how anyone would go about calculating the "cost" of a lifetimes' use of plastic bags or fossil fuels, for instance, especially when you have to add up the permanent loss of a resource to humanity for the rest of the time it's on the planet.)

Cheating was rampant at my American high school; teachers were either completely incompetent or were aware of the cheating and didn't care. Penalties were similar to what you describe.

That said, I wasn't in the Pacific Northwest, and my general impression is that people from that region are, on balance, more honest than in the region that I came from. But of course generalizations and grass is always greener and all that.

The former is seen as "a problem" and the latter isn't

That's not true. There are organized efforts to increase the number of men in (especially elementary school) teaching roles as well as efforts to increase the number of male nurses.

The point of my previous post is that your supposed trade-off is a false choice.

There's A LOT of middle ground between today's limitations on drone operators and a world where we expect to shoot down aircraft near crowded airports. If you try to sell that either/or to the public, you're going to lose. And rightly so.

FORTUNATELY, there's a lot the FAA could do to make the USA more drone-friendly while also ensuring the safety of conventional aircraft and critical infrastructure. Good drone policy, if ever enacted, will follow from this observation. It will not follow from a false choice between militarized anarchy and industry-crippling regulation.

(It'd also be neat if we could introduce technical solutions to the stupid/clueless person problem. For example, it'd be nice if the FAA funded the development of open-source, optional, and easy to disable software that warns the pilot (and only the pilot) before the drone enters a restricted area. It'd also be really nice if drone manufacturers voluntarily installed that software on capable drones. We're all better off if Johnny doesn't commit a felony and ground the next 30 minutes of take-offs while trying out his cool birthday present...)

It's really not clear to me that there's a trade-off between bird impact prevention and drone prevention (or anything else you mention) for two reasons.

First, effort allocation isn't really a compelling explanation -- most of the world just doesn't work in those types of zero-sum terms. Drones are sexy; just because government or industry is willing to fund research and development on safe drone-airplane regulations now doesn't mean it's willing to allocate those same resources to something else.

Second, even when resource allocation is strictly zero-sum, the sorts of calculations you suggest are mostly a matter of qualitative speculation about the future. Which means you're mostly just going through a number crunching process that "confirms" your prior beliefs.

that common sense "don't fly near airport runways" is sufficient

I think the major challenge that incidents like this one demonstrate is that it might no longer be sufficient to assume pilots have common sense.

The MVP / market analysis process is different for biotech, but still should happen. It's pretty fascinating!

Right. I think we're all violently agreeing.

Biotech companies do need MVPs and market analysis. My observation is just:

1) The idea of testing product-market fit by building a complete product and putting it out there in uncontrolled settings doesn't really work for a lot of biotech companies. Especially anything that's going to need regulators' approval.

2) Compared with IT startups, Biotech companies are more likely to fail due to having the wrong answer for "but can we make it work?" rather than "is there a market?". So the MVPs optimize for testing former question, which usually means controlled experiments in a lab. There are counter-examples on both sides, I'm sure, just in general.

not shooting them down seems like it could go much worse

Sure. But a sensible policy would also minimize the amount of weapons that are deployed in heavily trafficked airspace. That is why the sensible approach is to heavily regulate or even by-default prohibit UAV traffic anywhere near airports or common commercial aircraft routes.

In other words, parent post is a false dichotomy. It's possible to design and enforce regulations that 1) minimize danger and 2) enable aircraft controllers and the military to address rogue UAVs before they enter heavily trafficked civilian airspaces. Those regulations should go far beyond "shoot-to-kill near airports", but also don't have to be as encompassing as current bans.

Simply deploying weapons in commercial airspace as a complete alternative to legislation and regulation is insane.

Legislating it will only really stop people who don't intend to fly small aircraft into a no-fly-zone anyway.

I think you under-estimate the size of the group of people who do dangerous shit without malicious intent. Look at the list of incidents at the end of the article, for instance, and guess at how many of those were intentional.

Furthermore, even in the case of malicious intent, crafting regulation that clearly and unambiguously sets expectations can allow regulators and the military to deploy counter-measures long before doing so creates an unacceptable risk to surrounding civilians and infrastructure.

I thought one of the key ideas of a MVP is that it's something you put in the hands of customers...

Internet-type startups, at their core, are about marketing more than they are about technology. Getting the technology right is absolutely no guarantee of success, but you can hack together some PHP and -- if you get the marketing right -- make a shitload of money.

Psychology is kinda hard to do rigorously, so you test marketing the only place you really can -- in the market.

Conversely, the question "does this material prevent infections?" can actually be studied, to a significant extent, in controlled lab experiments. You don't need the marketplace to answer that question.

And for many cases -- such as the OP's -- it's obvious that if the answer is "yes" then you have a valuable product. Period. End of story. Infection-resistant materials aren't a fucking social network. If they work, they are obviously and immediately useful. You'll need to get the business side right, but it's really less difficult than the science itself because hospitals are begging for products to buy that address this problem.

So the question is more "can we solve this totally obviously extant and well-defined and obviously significant problem" rather than "have we identified a problem whose solution will let us win some pocket change of a big number of consumers?"

Figuring out if the science works out IS the MVP test. If you have a product that works your customers will be lining up.

From the article:

The trucks did not travel in platoon for the entire journey — only on motorways when traffic conditions were "normal" — and each vehicle, even those following the lead truck, had a human driver on hand.

Also, the convoy formed a network and explicitly communicated with the goal of not just driving autonomously but also recognizing cost savings:

a Wi-Fi connection keeping their braking and acceleration (but not steering) in sync.

So quite a bit different from just lane keeping.

It might even just be the case of a vocal minority of people who think everyone else doesn't work as hard as them.

I worked in a grocery store (in the US) back in the day. People complained about the smallest stuff -- that I wasn't as conversational as they wanted, or that I wouldn't process their coupons in the order that they expected, or even in one case that I wouldn't help them steal from the store.

Maybe 1 in 10 complaints I received was even remotely close to something a sane human would be even mildly inconvienanced by (and even for those I can't imagine ever caring enough to spend my free time complaining to a grocery store manager). It wasn't just me or just that store; management had an entire formal process passed down from the corporate office for figuring out which complaints should contribute to evaluations and which were spurious. I never had less than 10 complaints per quarter and only two of those were mentioned in my evaluations over a several year period. Other cashiers had similar numbers. Interestingly, friends who worked at non-grocery-stores encountered these sorts of things far less often.

I'll never understand the psychological reasons for why people behave like this. It's probably the same reason people troll. But I do wonder whether it's a US-specific phenomenon wrt grocery stores specifically.

The app was used at 100 airports. If you include the cost of the iPads then it's like $7000 for the actual app. Still high for a couple hours of work, but you figure most of that is overhead inherent in contract-style work. I never charge less than 1k for a project because it's not even worth the time it takes to place a bid and manage the paperwork, and I don't have to put up with government BS.

The ironic thing is that the person who made the FOIA request is complaining about lack of transparency when transparency is a perfect example of the sorts of perverse incentives that cause government projects have insane amounts of overhead. One of those spend a dollar to prevent a penny of fraud things. I don't really have any solutions, but there's no free cake.

'course the elephant in the room is that TSA itself is a waste of money even without expensive software contracts.

I'm not going to actually respond to a decade of writing in one comment.

My comment was not meant as an indpendent discrediting of MM.

My comment was meant as a reading guide for someone who wants to read MM in good faith without wasting their time "down in the (irrelevant) weeds". A guide to the systematic flaws in MM's reasoning and writing style.

The process really is common sense:

1. Figure out what MM is actually saying, the evidence he provides, and then -- instead of engaging with that evidence -- first look for the obvious counter-evidence.

2. Beware of implicit theses. If the post doesn't directly and unambiguously address some point of interest -- but instead of just kind of rhetorically implies a position on that point of interest -- then ignore it.

It's not that MM is wrong "because common sense". It's that he's wrong, and the way of demonstrating he's wrong is by using common sense reasoning techniques instead of engaging with his inane details.