Another Monday Message Board. Post comments on any topic. Civil discussion and no coarse language please. Side discussions and idees fixes to the sandpits, please.
If you would like to receive my (hopefully) regular email news, please sign up using the following link
You can also follow me on Twitter @JohnQuiggin, at my Facebook public page and at my Economics in Two Lessons page
John Quiggin 
Data. Not new gold or oil… our dynamics held by “Uber’s massive—and proprietary—dataset”.
The gfc and then gdc. Great Data Crisis. Article is short and sweet (actual examples and suggestions).
This stood out like the proverbial;
“Congress could apply this insight to industries accumulating vast quantities of data about daily life. At an artificial-intelligence conference in September, an Uber representative suggested that anyone looking to do cutting-edge research on autonomous-vehicles would have to come work for the ride-sharing company in order to access Uber’s massive—and proprietary—dataset of vehicle trips. Opening access to that data would speed up innovation and spur growth.”
Government Can Do More to Support Science and InnovationFocus on infrastructure, information sharing, and separating scientific advisors from regulators.
By Paul Romer
https://www.wsj.com/articles/government-can-do-more-to-support-science-and-innovation-1542758326?mod=e2two
Uber, a company built on misrepresentation and exploitation, is talking b/a as usual. The problem of autonomous vehicles has nothing to do with statistical patterns of use, which is all Uber can offer. It’s about handling edge situations safely with very high reliability: high-speed mergesnb, cyclists and drunks coming out of nowhere, emergency police signs.
Rooftop solar alone has provided up to 63% of total electricity consumption in South Australia at times around the middle of the day. With Labor returned in Victoria their solar subsidy will remain in place and with solar costs still trending down installation rates are likely to remain strong. With nothing to replace the Large-scale Renewable Energy Target that will mostly end in 2020 and a emissions reduction target from the Coalition that will require them to fight renewable energy to achieve, rooftop solar is probably the strongest force driving decarbonization we have at the moment.
I agree. Wind power is very useful too at certain times and in certain places. There is no doubt that Australia can get all its primary energy for electrical power from solar, wind and a few other renewable sources. It should aim to do so 100% by 2030, just 12 years away.
Primary energy for transport and industrial processes (in some cases) is a bit trickier but I believe we can get there too (100%), although not by 2030.
In terms of needful progress we are way, way behind the eight ball. Dangerous climate change has arrived. It is here right now in 2018. It is not something still off in the indefinite future The super-fires, super-storms, super-droughts, super-deluges, super-heatwaves etc, are happening right now in their earlier manifestations. Expect all this to get a lot worse.
Only when people realize that this is all real and that they rightly should be terrified will they act quickly and decisively to change our energy and production system. Let’s hope we are not too late. It’s touch and go now.
Soon enough, people espousing climate denialism will need to duck for cover or face a public pillorying of the first order. That’s the one bit of schadenfreude I will get out of this.
Who would have thought, Queensland once sold on the sunshine and tropical, now suddenly has government dire warning of heat wave? In November! Slip, slop, skedaddle.
Greece’s Varoufakis to run in European election – in Germany
https://www.reuters.com/article/us-eu-election-germany/greeces-varoufakis-to-run-in-european-election-in-germany-idUSKCN1NU0QA
Coal update. China has just suspended thermal coal imports for the rest of the year (****endcoal.org/2018/11/coalwire-255-november-22-2018) because stocks are piling up in coastal power stations. This is just an anecdatum, but it comforts the theory that Chinese coal is still heading down, and is not helpful to the gloomy alternative that the rise in Chinese coal burn in 2017 and early 2018 marked a reversal of the good-news trend. It is particularly unhelpful to Australian coal exporters.
Carbon sequestration update. A post at Climate Home, spun off from a journal article, suggests that policies for carbon removal are beginning to gain traction in the climate policy community. About time too. (****climatechangenews.com/2018/11/27/eu-target-carbon-dioxide-removal-part-net-zero-emissions-strategy) Thankfully, the post is technology-agnostic. It’s far too early to pick winners, as the fiasco of coal-plant capture shows. What we need is serious pilot projects (on a thousand-tonne scale) of different approaches.
Australia has apparently spent $1.3 billion on Carbon Capture and Storage for coal generation. That’s around two thirds the initial estimated cost of the giant Snowy 2.0 pumped hydro scheme. I’m not sure Snowy 2.0 makes sense given the cost of other options including multiple smaller scale pumped hydro schemes, but at least we would have something that is technically useful to have even if it would never really pay for itself. That’s better than the money pit CCS has turned out to be.
We definitely should remove CO2 from the atmosphere and sequester it, but we should – except perhaps for seasonal and emergency use — be looking at closing down a coal power station each year.
Ronald
the $2 billion cost of Snowy 2.0 is a number that Turnbull pulled out of his backside. Snowy itself says the estimated capital cost is between $3.8 billion and $4.5 billion and that estimate “will be further refined as the project moves to FID” (read: the cost will go to the moon).
Smith9, yes I really don’t see how Snowy 2.0 makes sense given that small scale pumped hydro is coming in at $1 a watt or less. Snowy 2 will have a huge energy storage capacity but at the current cost of solar and wind that’s not very valuable. Power output is where the money is and we need that all over the map and not concentrated in one location. I’m surprised I haven’t heard of anyone looking into increasing the power output of existing hydro power dams.
The Snowy 2.0 cost estimates also exclude the cost of building all the new transmission lines to get the electricity to where the people are. This will also be in the many billions.
Smith9: Australia will very probably need new transmission regardless of the chosen option for firming of renewables. See Blakers. If you choose many smaller off-river pumped hydro schemes over the Snowy megaproject, they are still in the hills a long way from the coastal cities. The same goes if you pick HVDC lines to Western Australia to balance wind over weather cycles, or big CSP plants with lots of hot salt storage. Grid batteries can be more local, but they are far too expensive on horizons over a few days. V2G and P2G at scale are untried.
BTW, advert for a wonkish blog post challenging last week’s US NCA expert report saying that the woorls’s carbon emissions are still on the horrific RCP 8.5 track: ****samefacts.com/2018/11/energy-the-environment/climate-alarmism/
James, if Australia ends up with one in 10 homes having battery storage by 2025 as Labor has set as a target it will be difficult for new transmission lines to pay for themselves. But even if they lose money big time they should, in the short term, help close down coal power stations.
Snowy Hydro transmission lines are already full up when electricity prices are high so Snowy 2 will need a lot of additional transmission capacity. But smaller schemes can be located right next to population centres. And there is no large city in Australia that isn’t near hills on account of how it’s all about the rain shadows. For example, Cultana pumped hydro will only be 360 km from Adelaide. Also, there is existing transmission capacity nearby because it’s near where the state’s coal power stations used to be. Sometimes that transmission capacity is full up with wind power, but that’s okay because when that happens the pumped storage isn’t needed.
@james w
My commet @1above was off the cuff. I certainly didn’t explain myself well. Your rejection though, of a core (private) dataset imho, for statistical purposes of defining the edges is unwarranted.
WE should own the data, was the point I was trying to make.
James you said; “The problem of autonomous vehicles has nothing to do with statistical patterns of use, which is all Uber can offer. It’s about handling edge situations safely with very high reliability: high-speed mergesnb, cyclists and drunks coming out of nowhere, emergency police signs.”
Ubers data provides some of the centre to define your edge effects. I love edge effects too. And agree they will define handling safely the almost limitless edge effects. Where are “the edges”? How would you find them?
And about handling those edge effects, and approval of autonomous vehicles, i do not agree they will be defined at 5 sigma;
http://www.physics.org/article-questions.asp?id=103
… before approval.
Did anyone get the edge effects of a colt 45 nailed down? What about the recent medical implant debacle which saw Australia use a device obviously ‘statistically’ faulty? Can you provide an example please of products or services which covered all edge effects before approval? What is the opportunity cost /s of delaying approval?
Some juristictions are already patially ignoring your plea about lanes and drunks, and approve autonomous vehicle trials in the real world with insurance cover backed by that juristiction. Then we will just pay to deal with the edge effects via a dollar more insurance on the autonomous vehicles trips. And then collect real data (which hopefully is not lorded over by uber waymo zoox etc) and continually improve not hitting drunks and lane changers. Some wags have said the mayor of Palo Alto sugested living in youy autonomous vehicle is affordable as they won’t charge rents on roads. Couldn’t verify.
I have not really been able to find where we define edge effects to a level of ???? or sigma-five before releasing to the public. The site below does agree with you, yet I still think my point will see autonomous vehicles on the road a long way before you, and I, are satisfied. See point 5:
“5. Safe driving: vehicle can manage all reasonably expected situations by itself”
http://news.thatcham.org/pressreleases/insurers-set-out-criteria-for-what-drivers-should-expect-from-vehicles-described-as-automated-2251926
“all reasonably expected situations” sounds very unscientific and open to very broad definitions. We will have autonomous vehicles before all situations are managed.
And hopefully we will have laws covering data which situate this data in the commons, available to “us” and the market.
And an aussie and experts have this happening – today;
“”The Zoox vehicle makes its way through the suburbs with ease, politely waiting its turn at four-way stops and giving cyclists plenty of room. When a black delivery truck unexpectedly whips across two lanes of traffic, the Highlander stops and avoids a collision. A few minutes later, we get on the freeway, and the Toyota merges in a manner that could be described as ultra-safe mode. Instead of jamming the accelerator to outrun oncoming traffic, it hugs the edge of the on-ramp as it waits for an opening.
It’s in the city, though, where Zoox really shines. The screens inside the vehicle show an overwhelming amount of information, as the computer vision software keeps tracks of cars, people, stoplights, and road markers all at the same time. Unlike many self-driving cars, it glides to stops. At an intersection with a left turn, it allows oncoming traffic to pass and then waits for some slow pedestrians. Overall, the vehicle performs so well that you forget no one is driving.”
https://www.bloomberg.com/news/features/2018-07-17/robot-taxi-startup-zoox-has-800-million-and-a-wild-pitch
And this in nsw now:
https://www.transport.nsw.gov.au/news-and-events/media-releases/all-aboard-first-driverless-shuttle
I am no statistician or autonomous vehicle expert, so can someone please point out flaws. Yet I firmly believe sd cars autonomous vehicles will be on the road long before they, we, solve the trolley problem.
Rare earth mineral discovery set to make Australia a major player in electric vehicle supply chain
https://www.abc.net.au/news/rural/2018-11-30/rare-earth-mineral-find-to-boost-electric-vehicle-sector/10562460
“Currently 50 tonnes per annum are produced by the pilot mine at Browns Range, which will double in the next two years”
– At the 2011 market prices quoted in the article that little mine will be producing dysprosium valued over AUD300M pa!
There are politicians who want batteries built in Australia, but I’m not sure they understand what’s involved. After all, Australia is second largest producer of lead in the world and despite the vast majority of the weight of a lead-acid battery being lead, we export most of our lead and buy most of our lead-acid batteries from China. (China is the world’s largest producer of lead and so most of Australia’s export lead goes to other countries.)
Meanwhile lithium is the lightest solid element mined so it is easy to transport and a tiny fraction of the weight of a lithium battery is lithium while lead. Also, humans are useless at making lithium battery cells. It is something that has to be done by machine if you want a cost competitive/usable product. So not a lot of employment there. It’s mostly capital investment.
So if politicians want battery manufacturing they should raise the super annuation contribution rate to keep interest rates low in this country and not expect many people to be employed doing it.
Or they could give companies money to assemble components from overseas that include fully made lithium battery modules from China and pretend that is manufacturing batteries.