spenrose

You have the dynamic correct but the values backwards. Cities are defined by their physical and social density and their long-term investments in physical capital. Uber lacks a sustainable business model. Transportation within a city features many "natural" monopolies (i.e. train tracks), and the promise of an emergent overarching monopoly: providing one-App entry into multi-modal transportation. If Uber can capture those monopoly rents it has a chance to survive as a permanent parasite. If it doesn't, it's doomed.

An attempt to capture the next few years of charging infrastructure: https://twitter.com/sampenrose/status/1152625267589840896

It is an abject failure of our amplification systems, notably HN and Stewart Brand's Long Now, that we are listening to two Baby Boomer musicians debate energy policy. Who cares what Eno and Byrne think about nuclear power?

Completely agree. Just think that day is far off for some programming labor markets in which HN readers sell their skills.

Very similar experience; after unemployment I took a job with a sociopath for six months, then went back to school for C.S. (I had a humanities B.A.). Lessons:

1. Human toll of layoffs can be extremely high (up to and including death) and hard to see.

2. Many coders are much more re-hirable than most non-coders, but many aren't.

3. In resilient labor markets such as modern SaaS and app coding, there is an experience<->hirability S-curve. Going from 0 to 1 year of saleable experience depends on the employer, from 1 to 3 years power shifts slowly towards labor, and after that you are in control. (These statements obviously don't apply to all software labor markets.)

4. Many coders get stuck in bad labor niches, many are really lousy at the sub-skill of getting a job, many don't realize how salable their skills are, and many can't tell which of those possibilities applies to them.

Interesting. The resurgence of "local idiot" nationalism has led to virulent anti-expertise in general and anti-environmentalism in particular in Poland, East Germany, arguably France ... what is denialism like in the Netherlands?

It's confined to the USA and Australia, where the combination of powerful Murdoch media and rural fossil fuel billionaires reign over the conservative parties. (Canada and UK get some spillover.) See above links to "Dark Money" and "Merchants of Doubt" for US. For Australia:

https://reneweconomy.com.au/australia-get-stupid-clean-energ...

"What’s more, Florida Power and Light has also applied to build two new reactors at the Turkey Point site—and the documents supporting that application rely on a projection of one foot by 2100. The NRC accepted that estimate after a year and a half of reviews that included hundreds of pages of scientific analysis, but there’s scant reason to believe the projection will hold. Over that same period, cities and counties in South Florida use a projection of between 2.6 to almost seven feet, based on the best available projections from climate authorities including NOAA and the United Nations Intergovernmental Panel on Climate Change (IPCC)."

No, I based on attending SparkConf between 2015 and 2017. You could probably assemble half of it just by reading summaries of Matei's keynotes.

"the first implementation was in Scala, which was chosen because of its algebraic data types and powerful pattern matching. This made writing the optimizer, which is the core part of the compiler, very easy. Our original optimizer was based on the design of Catalyst, which is Spark SQL’s extensible optimizer. We moved away from Scala because it was too difficult to embed a JVM-based language into other runtimes and languages."

There is an important "contemporary history of computing" article to write about the evolution of the Spark project from "let's build a distributed filesystem for MapReduce in Java because we read those early Google papers" to "SQL is the right model for working with data so DataFrames" to "meet data scientists where they are: Python (and R)" to "make machine learning easy" and now to "LLVM, but for crunching big numeric arrays".

As with any mass manufacturing technology, we climbed S-shaped learning curves. There is a substantial literature; search for "learning {curve|rate} {wind power|solar pv|renewable energy}"[1]. A simple explanation to the concept [2]. For solar, it's mostly silicon manufacturing techniques, a sort of distant cousin of the techniques used to manufacture your laptop's CPU. For wind it's that but also design (you want VERY tall towers geared for maximum efficiency at somewhat below median winds IIRC) and especially construction. The techniques for getting giant blades to site are not trivial; the techniques for building off-shore turbines are embodies in some of the world's more sophisticated engineering companies.

[1] https://scholar.google.com/scholar?q=learning+curve+analysis... [2] https://www.energytoday.net/economics-policy/credit-renewabl...

Pairs well with today's discussion of multi-stakeholder coordination on managing forest fires: https://news.ycombinator.com/item?id=20783166

Pairs well with today's discussion of multi-stakeholder coordination on a US military airplane: https://news.ycombinator.com/item?id=20792427

HN discussion yesterday comparing wind and nuclear: https://news.ycombinator.com/item?id=20724676

This thread has multiple links to nuclear professionals' on-the-record statements that cost is a problem:

https://twitter.com/sampenrose/status/1162772916125126656

The article says China's deployment plans are being derailed by cost and popular attitudes towards nuclear. The cost of new tech in turn is mostly driven by the chicken-and-egg problem of deployment. So you have to solve deployment, which means you have to change attitudes.

My tweestorm discussing this topic, with citations: https://twitter.com/sampenrose/status/1162772916125126656

"Officially China still sees nuclear power as a must-have. But unofficially, the technology is on a death watch. Experts, including some with links to the government, see China's nuclear sector succumbing to the same problems affecting the West: the technology is too expensive, and the public doesn't want it."

Wonderful leadership. Thank you, Stripe!

Identical for effective purposes. I don't make the same ~85 mile ride often enough to detect whatever small losses there may be.

“electrifying vehicles would reduce carbon more than completely decarbonizing the state electricity sector, pushing state emissions down 42 percent from 2018 levels by 2040 — not enough to hit the targets on its own, but a huge chunk. Second, electrifying vehicles saves consumers money by reducing the cost of transportation almost $600 a year on average.”

FWIW, my 2015 Leaf still goes 90 miles on a charge.

The chargers are cheap and getting much faster and will be deployed everywhere you can run an electric line: https://twitter.com/sampenrose/status/1152625267589840896

Best thing I have ever read on Python's history. Really fine work.

That halving will make variable renewables much more effective: https://www.vox.com/energy-and-environment/2019/8/9/20767886...

"A US energy grid run entirely on renewable energy (at least 95 percent of the time), leaning primarily on energy storage to provide grid flexibility, may be more realistic, and closer to hand, than conventional wisdom has it."

"for the same capital outlay, wind and solar projects will produce 3 to 4 times more useful energy at the wheels than oil will at $60 a barrel for diesel-powered vehicles.

For petrol cars, the ratio is even less favourable — the renewable investment will produce 6 to 7 times more energy. It is therefore increasingly difficult to argue that oil is the superior fuel from an economic standpoint, let alone when environmental issues are considered.

As electric vehicles proliferate, the long-term break-even oil price required for gasoline to remain competitive as a source of mobility could fall as low as $9 to $10 a barrel.

With nearly 40 per cent of current demand for oil coming from sources susceptible to easy electrification, oil companies should think very carefully about investing in new long-term projects that have break-even costs much above $20 a barrel.

This poses a major strategic problem for the oil industry"

Most of BART’s challenges stem from 180 software packages and 30 microprocessors that make up each Bombardier vehicle. Each is a complex symphony of systems that communicate over an Ethernet backbone. A bug in one line of code could shut the whole thing down, requiring a rigorous safety review process that may take months before engineers can reinstall the software."

There's an old joke about four-wheel drive on cars: instead of getting stuck right when you go off road, it ensures that you'll get pretty far into the woods before getting stuck. IHMO that metaphor applies here. Like the Web, Google, and even the WIMP interface before them, tabs were invented to manage information overload but "succeeded" only in pushing the problem out a layer:

"The most sophisticated and efficient users of Google today find that “trying to navigate the growing deluge of data [just within their scientific field] has become a second job.” The United States’ library of scientific papers for health and medicine adds new ones at a rate of two a minute, year round. Like Memex and the Web, Google has solved the problem of published work burying itself unread by pushing it out to a vastly wider layer, one we will perhaps never manage to tame. We have not escaped [Vannevar] Bush’s quandary:

Those who conscientiously attempt to keep abreast of current thought [find that] truly significant attainments become lost in the mass of the inconsequential. … publication has been extended far beyond our present ability to make real use of the record."

http://whatarecomputersfor.net/the-worlds-knowledge/

The Software/Data Carpentries focus intensely on these matters, and their founder Greg Wilson continues to work on them:

https://carpentries.org

http://third-bit.com