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

This is how Uber matches trips[0]. Optimising not for the fastest pick up for any one rider, but to reduce the total time of arrival in the whole network.

[0] http://blogs.cornell.edu/info2040/2019/10/23/uber-ride-shari...


In New Zealand, our broadband market is structurally separated.

The the access provider for both the copper and fibre network (right to the customer premises) is a regulated wholesale only provider, which is required to provide open access to its network on a non-discriminatory basis between all access seekers (retail telecommunications companies).[0]

This almost entirely resolves the net neutrality debate in New Zealand.

[0] This is a slight simplification: Chorus is the network owner for the copper network, and about 70% of the fibre network. The other 30% of fibre network is also built, owned and operated by wholesale only providers on a comparable basis.


How well do you feel that this has worked? I was under the impression that telecom prices in New Zealand were pretty painfully expensive?


NZ used to have expensive and slow internet (we don't have much TV cable here, so it was all ADSL), but it's gotten a lot better in the last few years. Currently I have 1000mbit down / 500mbit up fiber with unlimited data for $130nz/month (90usd).

The model does work pretty well.


I have 1000/500 UFB to my house for ~$100USD/month. Unlimited data :)


It's a little island in the middle of the pacific everything except maybe sheep is going to be expensive


Two islands, actually. More than 30 if you count all the outlying ones.


Electric service works that way in Texas.


The US has basically that model for many other natural monopolies, and it works alright.

I think we could do with a bit more government investment in infrastructure, because government is really the only entity that can drop a trillion dollars, then wait 40-50 years for its returns. (Well, them and pension funds, but neither has exactly shown great financial prudence lately.)

I just... Have a hard time articulating why infrastructure is important, because it's just so glaringly obvious from my perspective that it is. I mean, we see infrastructure everywhere in nature -- from the layout of bacteria mats to ant colonies. The idea that our society can be vibrant without having to perform those basic functions is absurd.

So it makes sense that the lives which depend on the services should control the final delivery network when it's not possible to deploy more than one or two -- eg, FTTH should be utility, cellphones should be a market; roads should be (largely) public, etc.

If the US's power is its people, then infrastructure is what allows us to maintain and focus that power.


> Imagine being a chef in 2015 and discovering that there are thousands of vegetable species you’ve never cooked with. It’s like discovering corn, arugula, tomatoes, and lettuce for the first time.

This struck me as particularly interesting. Corn, arugula, tomatoes and lettuce are all a product of selective breeding. To what extent has this been applied to "sea vegetables" and what might the possibilities look like?


The Dutch company Hortimare is working at exactly this.


Not an aeronautical engineer, but understand that the cylindrical shape used in today's aircraft design is very good at containing the stresses of a pressurised cabin.

However, the design does mean significant pressure stress on the front and rear of the cylinder. This must be managed with extensive (heavy) structures.

A doughnut design, in theory at least, solves for this by removing "the ends" of the cylinder.


It looks like 2 exit doors.

On that note, how would this affect the all important question of how to most efficiently board an aircraft?


Interesting from the FT [1]:

Airbus said although the design was worthwhile enough to protect with a patent — like about 6,000 other ideas its engineers devise every year — it was not an immediate prospect for shuttling passengers from Heathrow to JFK. “This is not something that’s currently under active development,” Airbus said.

[1] http://www.ft.com/intl/cms/s/0/ba80c518-6492-11e4-b219-00144...


One surprising thing I learned is that large companies like to mass-produce patents, even ones they have no intention of ever implementing or suing over. Why? Because they provide a MAD defense against patent suits. If Boeing ever sued Airbus for violating patents, Airbus could countersue and, odds are, each company is violating some patent of the other. Also, if each company has 10,000 patents applying to aerospace, it's going to take each company's legal team a lot of time to sort through them and figure out which ones they can countersue over. Time = money; a patent attorney's time especially so. So coming up with as many patents as possible makes it expensive (often prohibitively so) to sue over patents.

Your natural reaction might be to think this is an outrageous waste of people's time, and you'd be right. But if you don't do it, your competitors will and you will be vulnerable to patent suits that, because of how this works, are essentially impossible for you to avoid making yourself vulnerable to.


Make patents 100x more expensive and see if they still think that.


Yeah, we definitely don't want Airbus to come up with new ideas. Fuck those guys.


it will create a market where only the rich can get patents - precisely the opposite of what I presume you hope to achieve.


Keep patent costs the same, but reduce their lifetime to 5 years.


What problem would that solve?


People patenting ideas they have no serious intention of ever building. The maximum effort they'll make is stopping someone else from doing it.


Much better idea would be something similar to trademark: "If you don't defend it you lose it".

Similarly, if you're not utilizing it, then you probably don't need it.


This is why patents are "short". To offset the monopoly.

It's why you hear pharmaceutical companies whine patents are strong enough, they need them longer. Automobile makers, aeroplane makers, ship builders, and "industrial" companies that make machinery and equipment rarely complain, they are pretty much ok with their duration, they might like them a little shorter but could take it or leave it. And on the other end if the spectrum we have software companies who are forced into the game because of consumer electronics companies (for whom perhaps something between 5-10 years would be best) taking patents on "features" they wanted to make as software... And the software companies pretty much want 0-5 years because this stuff is a all mostly math, or irrelevant, or under better protection by copyright, etc.

Patent limits were devised as a one size fits all solution not taking any of this diversity into account.


The article is quite clear that it's a totally unreal hypothetical:

  Of course, once your investment had grown past a certain 
  point, it would be impossible to put into any stock
  without drastically affecting the stock’s price. In many
  cases, investing this volume of cash would be
  impossible [...] we also did not take into account
  trading fees nor changes in the composition of the S&P
  index. We also assumed that all companies allowed for 
  investment in fractional shares.

  Moreover, even if one made the impossibly optimistic 
  assumption that picking the best stock of the day is a 
  50-50 guess, there would be only a one-in-3.53 
  trevigintillion (3.53 x 1072) chance of matching these 
  results.
There's another humorous follow up on Quartz, positing that "[i]nvesting a day late every day would have returned 19.6% in 2013".[0]

[0] http://qz.com/158489/investing-a-day-late-every-day-would-ha...

Edit: spelling.


A lot of people here seem very upset that this isn't a recipe they can follow to riches. Clearly this is just an exercise illustrating wild growth, and an interesting article. No need for people to be so bitter.


19.6% is less than the S&P 500 index performance in 2013!


An economic analysis on a proposed minimum income model for New Zealand:[0]

  The [basis minimum income] scheme proposed by the Welfare
  Working Group is a significant policy change with large
  economic consequences. The scheme is fiscally very costly
  and would not necessarily achieve its main goal of 
  reducing poverty. The high personal tax rates required to
  fund the scheme are highly distortionary to the labour
  market and to savings and investment decisions, and would 
  be likely to induce a significant behavioural response. 
  This has damaging effects on the tax system and economic 
  growth.
Also of importance, from the same report:

  Although the Gini coefficient improves under all models,
  many beneficiaries (including the disabled, carers and 
  sole parents) currently receive more than $300 per week 
  and would be made financially worse off under a GMI 
  scheme.
At a high level, it seems that the economic reality of providing a basic minimum income to everyone requires a significant increase in taxation revenue. That might be possible with significantly higher GDP per capita, but I'm not certain we're there yet.

[0] http://igps.victoria.ac.nz/WelfareWorkingGroup/Downloads/Wor...

Edit: formatting.


That might be possible with significantly higher GDP per capita

That's actually a really interesting point. The feasibility of BI is going to be tied to GDP per capita, as well as cost of living. In a country where GDP-per-capita is several times cost of living, BI would be relatively easy, whereas a country where GDP-per-capita is barely above cost of living, BI would be very difficult (because where do you come up with the money?)

US GDP per capita is about $50k. I'm not sure what cost of living is, but that definitely isn't a whole lot higher than what is usually accepted as the threshold to living somewhat comfortably (somewhere around $40k)

Norway, on the other hand, has a USD GDP per capita of ~$100k, so unless cost of living is much higher, BI may be much more feasible there.


The linked numbers are brilliant. Lacking the "in my head" math skills I should probably have, I regularly whip out Excel to solve the kind of use cases you can imagine from the linked numbers in the video.


"Simply put, these are the best racing graphics I've seen on a PC!"


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

Search: