Video Of The Week: The Coming Disruption Of Transportation And Energy
My friend Steve suggested that I watch this video about the coming changes in energy and transportation and how profound (and rapid) they will be.
It’s long (an hour) but worth it.
Comments (Archived):
Thanks Fred and cued up for later. Have a great long weekend.
remarkable predictions. it’s the speed of this disruptive change that shocks me. if he’s right then it’s not only disruptive but also very destabilising to governments around the world. there could be trouble ahead. data as the new oil we’ve discussed previously, but if that becomes reality then we might begin to see more authoritarian nation state governance, and then a populist response.
remarkable predictions. if he’s right How can a prediction be remarkable before it’s even proven to be correct?Also even if it’s proven to be right (in other words we play this video in 20 years) aren’t we just cherry picking someone who said something vs. 10000 others that said different things? (In other words the lucky sperm).Warning: Ad hominem on Tony Seba’s credentials follows..And Seba? Sure ‘best selling’ book in the small niche Amazon category “Natural Gas Energy” with a total of 116 reviews:https://www.amazon.com/Clea…Seba’s claim to fame appears to be Printnation:https://tonyseba.com/biogra…Printnation failed and had to be bailed out. The site shut down and rescued by a small printing distributor, Pitman Company (used to deal with them back in the day):https://americanprinter.com…http://whattheythink.com/ne…He seems to certainly have succeeded at that ‘speaking and consulting’ thing though… Not seeing proof of the claim (bandied about on pages about him) that he is a ‘serial entrepreneur’ either.Lastly and ironically he failed to see the disruption the Internet would have on the industry that he started his business in. Printing.My point? His thoughts on all of this appear to be self serving (sure we all do this) about something that he stands to gain by promoting and from peripherally. I see this quite differently than, say, when Albert Wenger pushes UBI which seems to more pure and not for economic or personal gain. Or @williammougayar with the blockchain because he has real skin the game.
if he’s right you will only need to pause the video.yes, everyone has an angle, but the data extrapolations seem plausible. i am surprised though by the speed of adoption he’s suggesting. he doesn’t have much to say about the potential influence of government to regulate (put the brakes on) the acceleration.
See my post here athttp://avc.com/2017/09/vide…Net, for all of his predictions on transportation, he is extrapolating from Level 3 autonomy to Level 5 very quickly. IMHO Level 5 needs essentially full “general artificial intelligence”, and so far no one has even as much as a weak little hollow hint of a tiny clue how to do that.In my post I suggest a way to get what he wants in transportation with essentially just Level 3 — massively re-engineer the roads.For his stuff on energy, we’d all be much more comfortable with the backup of the coal mines, the coal trains, and the huge pile of coal out back, with the pipelines full of natural gas, and with the nuclear fuel rods instead of some hopefully just enough Li-ion batteries until the sun shines again! Or, no way do we want a blizzard to leave us without electric power. I omitted that point in my post, but I didn’t want to be TOO critical!Or just how big will the Li-ion batteries really, really, really have to be for the many crucial needs for electric power — hospitals, the Internet, traffic lights, water pumping stations, home heating in blizzards, the trains that depend on electric power, national security, etc.? Basically he’s suggesting massive DIY electric power!!!! Think about that for a few minutes and see if you are comfortable with it — I’m not.Or, if lots of people put in DIY electric power, then in normal weather the grid won’t get much revenue and will have to charge a big fee just for a connection even if it is not used, and nearly all their capacity will be for just emergency situations. So, we’d be paying the CAPEX for TWO electric power systems. I neglected to mention that point, also.DIY electric power? Naw!Sure, I like his S-curves a little — in my post you will see why!
To the man with a hammer everything looks like a nail.
I concur that your skepticism about the author was justified: I took the video too seriously. As I finally explained in my long post here, he didn’t make a good case, and his projections are not good.E.g., he started with two pictures of traffic in Manhattan 13 years apart from nearly all horses to nearly all cars. So, from that we’re supposed to admit that rapid change is easy? Nope, that picture is not evidence of a general pattern or a representative sample of anything general.Instead, each case of change, rapid or not, needs some careful justification some levels of detail below just two pictures.Once I was on an airplane flight, and the guy next to me was with IBM and had just come from a meeting at a university. Those days microelectronic line widths were in microns, and that seemed amazingly small. He then told me he had just seen demonstrated transistors in the nanometers. Gee, one nanometer is 10 Angstroms, and one Angstrom is about the diameter of an atom. 4500 Angstroms is, IIRC, about the wavelength of green light. So, he was talking nanometers, so few he was talking not so many atoms.So, a lot could have been extrapolated from that, really to the present.Actually, before 1985, IBM had in Fishkill, NY a cyclotron. Why? Because it runs electrons around in a circle quickly. Why? Because when electrons do that, they radiate X-rays, and IBM wanted the X-rays as the light source for microprocessor lithography. Well, those X-rays implied that before 1985 IBM was thinking transistor sizes of a few nanometers. Indeed, IIRC, without looking at the latest news from IEEE, current production microprocessor lithography is at about 14 nm. Ah, to heck with my memory; the Intel 14 nm is correct as athttps://www.intel.com/conte…But, IIRC, so far the light sources are still just in the ultraviolet (UV), maybe extreme UV, maybe on the way to soft X-rays. So, as of now, at 14 nm, microprocessor production is finally catching up with what IBM had in mind and believed in enough to build a cyclotron in 1985.So, really, 14 nm should not have been a big surprise since 1985.Now what I want to know is why Microsoft for Windows Server 2016 wants the computer’s main memory to implement error correcting coding (ECC) but the DDR3 ECC memory is now out of production in the whole industry, DDR4 is available, but Intel i3, i5, and i7 processors don’t support ECC, AMD claims only softly that their new Ryzen processors support ECC, and motherboard manufacturers Asus and Gigabyte for their motherboards for Ryzen processors claim support only for non-ECC memory.So, to get ECC memory, have to pay MUCH more for server processors that appear to be different only because of price and admitted ECC support.Last year, could get an AMD FX-8350 processor, 8 cores, 4.0 GHz clock, 140 Watts total dissipated power (TDP), with ECC for about $140, an Asus M5A97 motherboard for that processor and that supported ECC memory, and such memory from Kingston. Now all of a sudden, that level of performance costs several times as much.It’s about time for a good motherboard manufacturer to step forward and admit full support for AMD Ryzen processors with ECC memory and claim that Windows Server 2016 installs and runs just fine.
Argentina had the 2nd largest reserve of lithium. Anyone with insights into investing in Argentinan industrial sectors?
who has the largest?
I believe it’s Chile
Neither China or Argentina value private property rights. If you are embedded in the govt and in favor you can invest and make money
China has much larger lithium reserves than Argentina (more than 50% greater than Argentina), but Argentina produces more than China. China will likely increase production.
This video disrupted my morning and it was worth it. I think you undersold it, Fred. The speed of the coming disruption in energy and transportation is astonishing. Thank your friend Steve again for this. Definitely one of the best resource information pieces I’ve ever encountered on this subject.
Who is the Howard Hughes Senior of Lithium precious metals production?
Whenever autonomous driving gets the regulatory green light the dominoes will begin to fall. Interesting times ahead in the near future.
He’s assuming Level 5, and that’s not on the horizon.
We will also disrupt the heating market in the near future. By 2050, heat pumps will have replaced most of the furnaces and boilers in the country (about 110 million of them). That will dramatically reduce demand for oil, gas and propane while increasing demand for electricity. This “Beneficial Electrification” of transportation and heating will lead to “The Second Great Electrification” of our society. When we’re done, there will be essentially no more point-of-use combustion for energy. Oil and gas won’t be used for energy, they will be limited to use as feedstocks for the chemical industry — until we eventually replace that use of fossil fuels with recycling of carbon waste in sewage, etc.
So, perhaps Google (Alphabet) should have invested in heat pumps instead of Nest smart thermostats?Heat pumps have been touted as “just around the corner” for fifty years. Usually, the upfront cost and “strangeness” of the technology have kept it as a “future” product rather than something homeowners purchase at Home Depot. Also, if there is an electricity outage, you have no heat for the house!
> Also, if there is an electricity outage, you have no heat for the house!True also if have forced air heat — need electric power for the fans for the forced air. True also if have oil fired hot water baseboard heat — need electric power for the oil pump.For that problem, I have a portable generator, 6500 W. That’s enough for the essentials — pump in a water well, a computer, a few lights, one burner on an electric stove, and the pump or fan for the furnace.And, if had a heat pump that drew more than 6500 W, then could use just some electric space heaters for about 3000 W and keep one room warm.If had 0 F or some such outside, might keep the water pipes from freezing just by letting the water run slowly so that would be running 55 F water from the well or city through the pipes.Generally a gasoline powered generator for 6500 W could go a long way to supplying essential electric power for a day or so if other sources, e.g., the grid, own Li-ion battery, etc. went dead.Right a house with 100 A service at 230 V has 23,000 W, a lot more than 6500 W. But with the 100 A and 230 V, can run the oven and some burners in the electric stove, lots of lights, a flat iron, a microwave oven, the washer, and electric dryer, a dish washer, etc. all at once. Right, can’t do much to charge a Tesla!6500 W can be okay for essentials for a day or so, a week if could get out and get some more gasoline.
I know all this…I lived “off grid” years ago in the Sierra Nevada mountains. I had a 4000w gas-powered generator, with a requirement most people don’t need: 220v for the deep well water pump.My lighting and TV were 12v, and the deep-cycle battery I charged using a second set of heavy wire, hooked to my 12-v car battery, and charged by driving to and from work (10 miles each way).As a stand-by, I see Harbor Freight has a $99 CA-approved generator. Good enough for the usual 2-3 hour outage in CA (fridge, forced air fan, LED array) if you pay attention.
Yup, 100% absolutist thinking loses again! Instead of depending 100% on the grid or 100% on own Li-ion batteries, etc., have some local backup power, not enough really to run the house and really replace the grid but enough for essentials for a day or a few days.My 6500 W unit is on two nice wheels with rubber tires and has a handle can use to wheel it around. It has a very nice gasoline engine, overhead valves, generator, overload protection, fuel gauge, sockets for 115 V or 230 V. It’s usually intended for general contractors at construction sites. I got mine at Home Depot for ~$1000.I did make a mistake! I got several containers for gasoline, 6 gallons each, filled them, and put them on a wire rack shelf unit. That was enough gas to run the generator for a week or so. Alas, after a year or two, the containers were all empty from very slow evaporation! Yes, the containers were plastic!Another problem with storing gasoline can be getting some brittle, waxy flakes that can clog up nearly any fuel filter. So far I’ve seen that only for my lawn mower. I don’t know the real cause.
Google (Alphabet) *has* invested in heat pumps! Check out https://dandelionenergy.com/ which spun-out of Google X in July after several years of internal development and research. The process of eliminating fossil fuel furnaces in 100 million homes by 2050 has begun…
I read the site! Not in my area. No snow here in the Bay Area! I see they kicked off in July 2017. All brand new! Good luck Dandelion (dislike the name…how about “All-Earth Heating and Cooling”?)
Empty — delete.
Made some minor edits and then, due to some Disqus logic, had to repost as another post to retain the attached PNG file.
Lots of excellent quantifications of disruption with examples in the x100 to x120 flavor.
(1) Bottom LineHe is assuming:(I) Really cheap solar panels.(II) Really cheap Li-ion batteries.(III) What is called “Level 5” self-driving cars.(1.1) EnergyFor his topic “energy”, that is, the current electric grid, with assumptions (I) and (II), okay:(1.1.1) Grid Peak Load ApproachesToday maybe can replace current grid peak load generation approaches. To know, just finish the engineering and do the cost arithmetic.(1.1.2) Replace Much of the GridSoon we could have solar panels distributed close to where the electricity is needed, e.g., roof-top solar, backyard solar, factory roof solar, with a LOT of Li-ion battery storage, and replace much of the current electric grid.Actually, in practice, this can’t work: This would be nearly all users of the grid to change to do it yourself (DIY) electric power.Then, just how much Li-ion battery backup would be needed to handle all the emergencies of 1-5 or so times a year? How about a blizzard that blocks out nearly all the sun for the solar panels and lasts for five days?So, the electric utility would sell power mostly only in emergency situations. Then to pay for their costs, they would have to charge just for a connection.Then, really, as a country, we would be paying the CAPEX for our electric power twice, once for the DIY option and once more for the grid for emergency situations. Bummer.And, as long as we still have the grid, are we going to depend on LI-ion batteries for power during emergencies? Just how big would those Li-ion batteries have to be?We’d feel much more comfortable with the big supply of coal from the coal mines, on the coal trains, and in the big pile of coal out back of the generating station, the big supply of natural gas in the pipelines, and the fissionable materials in the nuclear plant fuel rods. Each of those can last for months.(1.1.3) BummerNet, I have to conclude that the energy part of the lecture is mostly nonsense. We still want the grid.(1.2) TransportationFor his topic “transportation”, with all of (I)-(III), his projections would say that personal cars would get replaced by taxis with no drivers. But that future needs (III), Level 5, and that’s not even on the horizon.(1.2.1) Re-engineer the RoadsThere is a possibly feasible technical approach to replacing personal cars with taxis with no drivers with essentially just current Level 3 self-driving technology — massively re-engineer the roads so that the taxis can be on essentially electronic tracks with none of the occasional things on the roads, say, police directing cars around an accident, that Level 3 can’t handle.(1.2.2) DieselHis “transportation” is too broad because he has nothing to replace most uses of Diesel in transportation, on land or water. And for his transportation, he didn’t include airplanes.Basically his “transportation” disruption is for a lot of self-driving taxis that need either Level 5 self-driving, not on the horizon, or massive re-engineering of the roads. So, his disruption of “transportation” is not promising.(1.2.3) Charging TimeIn the short term, he wants to project a lot of people owning all-electric cars. That’s not promising because the charging time is too long, especially in a house with a grid connection at just 100-200 A at 240 V.(1.2.4) BummerNet, his predicted changes to transportation are essentially just a bummer; won’t happen.(1.3) MotivationsHe seems strongly motivated by some emotions for solar panels and batteries and against electric utilities, carbon based fuels, and personal cars.His emotions are getting the better of his rationality.More details:Why do smart people at smart organizations consistently fail to anticipate or lead market disruptions? Because individually it is not in their interest to so lead.They would rather stay with the success they have than pursue risky success from disruption, especially if the disruption would disrupt the success they already have.For a CEO, even if he sees his existing business slowly dying, for him to pursue a disruption of that existing business would be seen by stockholders, etc. as bad for the existing business. And if the CEO does pursue such a disruption that fails, he will get fired. If the disruption disrupts his existing business but for a smaller business, he may get fired.Basically, nearly no one, stockholders, stock analysts, investors, BoD members, CEOs want anything to do with anything both new and significant. Nearly all such people see any such new things as too risky or as with likely big downsides and only a tiny chance of an upside.A fundamental problem here is the inability of relevant people to evaluate new projects accurately.There are means of such evaluation and lots of great examples, especially from applied science, the US DoD, and engineering, but the business community just ignores all such things.Cost reductions in batteries and photo-voltaic solar cells are close to irrelevant: Even if such devices were for free, for the main, currently envisioned applications, there wouldn’t be much change. Why?(A) For the batteries, for electric cars, the bottleneck is recharging time, especially at home on a typical home electrical service of 230 V at 100 A or 200 A.(B) For solar cells, the problem is storage. why? Because electric power is needed when the sun is not available so that the energy from solar cells must be stored.Yes, if could get BOTH batteries and solar cells dirt cheap, then could cover deserts with cheap solar cells, have the solar cells charge the cheap batteries, and have the batteries feed the electrical grid.Or, maybe could have solar cells on roof tops, in backyards, on factory roofs, etc., have them charge batteries, and have the batteries supply the electric power needed locally. Maybe.So, this situation would be the author’s “technology convergence”.How to know? Do the costing of the candidate engineered systems, and for given technologies we know very well how to do that quite accurately.Yes, accurate projection of when both battery and solar cell prices will be low enough to be attractive for feeding the grid is difficult.His technology adoption S-curve appears to be what I reinvented the second time I saved FedEx from going out of business. So, the BoD wanted some revenue projections.I said, let t denote time, y(t) be the revenue at time t, b be the full size of the market, and y(0) be the current revenue.Then at time t the rate of growth in y(t) will be from calculusy'(t) = d/dt y(t)Then from virality, at time t the rate of growth y'(t) will be directly proportional to both the number of current customers, proportional to y(t), and the number of customers not yet served, proportional to (b – y(t)).So, for some constant k, we have thaty'(t) = k y(t) (b – y(t))with y(0) and b both known. So that is an initial value problem for a first order, linear ordinary differential equation.Yes, easily there is a closed form solution, the S curve:y(t) = y(0) b e^(bkt) / ( y(0) (e^(bkt) – 1) + b )also known as the logistic curve.ForThe exponential is actually getting even more exponential these days. No, if just look at the algebra, how fast the S-curve grows is just from the constant of proportionality k; that clarifies the remark in the lecture.So, I reinvented the logistic curve as a model for viral growth and got that curve from a differential equation that made sense from viral growth. The logistic curve is old; there’s a chance my derivation from the differential equation and virality are new.Why did this little bit of calculus save FedEx from going out of business?The BoD had asked for some revenue projections. No one had anything beyond hopes, dreams, intentions, ambitions, etc. So, reluctantly I got involved and did the work above.The person assigned to develop the projections was SVP Planning Mike Basch. I showed Mike my derivations, and on a Friday he and I picked a reasonable value of k, and I used my HP calculator to do the arithmetic and drew the curve.The next day Mike was traveling but was also the BoD meeting that wanted the projections. At about 8 AM, the BoD meeting started and the graph was presented. Our two representatives (adult supervision for FedEx) BoD Member General Dynamics asked how the curve had been determined. All the FedEx people at the meeting worked to find out.At noon I got a call in my office from SVP Roger Frock who suspected I’d done the graph, confirmed that I had, and asked if I could come to the meeting.When I arrived, our General Dynamics people had given up on FedEx, returned to their rented rooms, packed their bags, gotten plane tickets back to Texas, and as a last chance for FedEx were standing in the hall, unhappy.I’d brought my HP calculator (right, HP could make a nice drama ad on this story).Roger picked a time, and I carefully punched my calculator and reproduced the point on the curve. Roger and I did that about four times, and then everyone was happy. The General Dynamics people stayed, and FedEx was saved. No one asked to see my calculus derivations!A guess is that the General Dynamics people saw a huge, even threatening, case of dysfunctional goal subordination at FedEx: That is, no one wanted to invite me to the BoD meeting, and, except for Roger Frock, they were willing to see FedEx die before they would let me into the meeting to explain my work.So, the General Dynamics people were not so much interested in the projections and the calculus as the goal subordination. Maybe.After that day, I did seem to be relatively less welcome at FedEx!The promised stock was over a year late. I decided to go get a Ph.D. and did. On my last day, FedEx founder, COB, CEO, Fred Smith had me come to his office with Mike Basch, then officially my manager, and said “You know if you stay you are in line for $500,000 in FedEx stock.” No, I didn’t know that, and Fred didn’t put it in writing. Basch had already told me “There is no money in the budget for you.”. So, I got my Ph.D.The attached graph shows solutions for the differential equations depending on the constant k selected.ForBusiness model innovation is every bit as disruptive as technology. Not really: Business model innovation is short on barriers to entry.For “autonomous cars”,’fully’ self-driving cars are “not even close”. from the head of the relevant research at Toyota and as athttps://techcrunch.com/2017…The Toyota guy is talking about “Level 5”, and so is the lecture.Level 5 is tough: No one but NO ONE has even a weak little hollow hint of a tiny clue about how to do Level 5. IMHO, Level 5 is essentially the same as general artificial intelligence, and that is not anywhere on the horizon.Currently what we have is Level 3; the lecture is extrapolating to Level 5 far too fast.If really want the results from what the lecture calls “A-EV”, that is, autonomous electric vehicles, then could just stick with essentially Level 3 as now but massively re-engineer the roads so that the cars could be fully “autonomous”. Basically put the cars on electronic tracks in the roads and otherwise carefully control the roads to keep away anything that needs actual intelligence.For Li-ion batteries as an energy source for electric utility peak loads instead of gas turbines fed by natural gas?Just do the cost arithmetic.IMHO the interesting parts of the lecture were the fall in price of https://uploads.disquscdn.c… Lidar, Li-ion batteries, and solar panels.
This hour-long video is *10x* more valuable than any hour-long news show I’ve seen in 2017…
The autonomous vehicle predictions make no mention of the capital costs and maintenance costs of roads. Currently in California, the gasoline tax is supposed to pay for CalTrans work, and people pay a little at a time every time they fill up. But the tax is woefully behind, CalTrans is “Billions” behind in road maintenance, and County tax bills cannot keep up with road maintenance either (if you apply for a building permit in a rural area, you get socked with a huge road-related assessment.) With fewer vehicles, mostly electric, the cost of roads per traveler is going to get onerous, it appears!
I’ve been developing utility-scale solar farms for 6 years (280 MW and counting) and he is under-estimating the solar potential of the grid. We are operating at cost decline rates and scale increases that outsiders can’t believe. If anyone wants to really understand what is happening on the ground in the US, contact me.
Ah, inhttp://avc.com/2017/09/vide… I took Seba too literally, too much at face value, and was being too nice. My guess is that Seba is pushing propaganda for “renewable energy.”
Good take on big disruption and the fact that it’s usually multiple factors that lead to the tipping point.Where Tony oversimplifies this is the grid model. We’re not going to be cutting cords any time soon (99.99% reliability is very expensive using solar + battery – needs generator backup to be affordable) and the grid is usually a local monopoly. They will have a big say in the speed of this disruption.Cheers.Dave P.
That was amazingly worth it. Thank you. Shared with everyone I know.