Sunday, January 20, 2019

Recapping 2018 In Technical Appendices (With Some Cheating)

Tell me what you see.
In Gordon Dickson's knights in shining armour are the standard solution to bad dragons, because a charging knight concentrates all that momentum on the lance point, allowing it to penetrate the heavily-armoured, fire-breathing pests. It's probably more complicated than that, because there are many sequels, but I'll move on now to the fact that this is an homage to one of Lynn White Jr.'s many brainstorms, and a pretty good indication of the impact of White's approach to history of technology, which was basically reading medieval comics and telling everyone about what he saw. 

It turns out, Lynn tells us, that much technology, and therefore technological change, is too quotidian for anyone to have ever bothered to write a description and earnest account of its historical impact. Without unconventional sources, we're left to wake up one day in our fifties and say, "Duh, of course iron axes and early glass making are related through soda ash!" The knights and stirrups lead to modernity thing might not be one of his best, but we're not going to get anywhere without investigating these things.  

So history of technology turns out to be pretty invisible. Writing the technical appendices for this blog has brought Lynn White's insight home to me repeatedly, and it occurred to me that it was time for a summary recap. (Also, I am not going to have Sunday off, so I can't get the next postblogging installment done this week.)

Since this is the first time I've done this, and there were technical appendices before 2018 that could do with revisiting,  I am going to do that. Ideally, I'd go back to the beginning and do them all, but that seems like work, so I'm going to start with my favourite, "The Science of Cozy," posted on 26 October 2017. That's the cheating in the title.

A 1948 house in San Jose, California

By Angelsharum - Own work, CC BY-SA 3.0,

"The Science of Cozy" explores the forced draft furnace that probably sits in the basement of the building in which you read this, if you are reading it in North America.  There is a brief explanation over at Wikipedia, but it isn't rocket science. A fan forces air over the surface of a heating unit, usually a furnace burning natural gas, and then through ducts leading to the various rooms to be heated, where a boy reading a science fiction novel may be sitting on a vent like this in the company of an ill-tempered, heavily scarred, but basically adorable black and white tomcat. The technology was around long before WWII, but the Warm Air Research House of the University of Illinois's Department of Mechanical Engineering proved and validated the economics in good time for the postwar building boom. 

Being cold a lot and getting sick is
correlated, even subtracting hunger
Although radiative heating of various kinds works well, too, the advantage of heated air is that the human experience of comfort starts with the temperature of the skin, which is largely determined by the temperature of the air around it. Radiative heating from remote sources can easily get out of line with air temperature, leading to situations where you are pouring heat into a room and either overheating it to the point where windows have to be cracked open to the freezing Toronto night (just as well given that the window didn't actually close), or leaving the occupants still uncomfortable. This is not likely to happen in a small room, to be sure, but the reader will note that the large rooms and open plans of postwar architecture assume that we've mastered the "science of cozy." People are comfortable in the winter now, and that is not a small accomplishment in a continental climate. It has demonstrably changed domestic architecture and fashion, has probably made some contribution to the obesity epidemic, and makes a strong argument (in my heretical mind) for a place for producer gas distribution in our carbon-negative future.

You could make electricity with this out of methane just as easily
as from natural gas. 

On 15 December 2017, I posted a technical appendix that "segued" into another series. Leaving that aside, it was a discussion of the axial compressor. An axial compressor compresses gas by stacking stages of turbine blades on the axis of the tunnel through which the air is blown. It is key to most modern jet engines, and that December, I was particularly interested in responding to an economics blogger presenting the Concorde airliner as an example of "technological regression." The Concorde was powered by a turbojet that used an axial compressor, as, to be sure, most are. The Bristol/Rolls Royce Olympus has an important place in history by virtue of being one of the last turbojets used on an airliner. Turbofans are favoured now for most applications, as they are more thermally efficient at subsonic speeds.  That detour stranded the Olympus as an aero-engine, although it is still used as a power plant and marine engine. It is also interesting that an improved Olympus, a requirement for a Concorde II, had any such thing been feasible, would have emerged from another notable British aviation cancellation, the BAC TSR-2.  However, the axial compressor did not start out as  jet engine, but as a blower for base-metal refineries. If these cancellations are, indeed, evidence of "technological regression," it is interesting to set them against an era in which metal refining accelerated dramatically, the Early Iron Age.

Although Nineteenth Century axial compressors were used to refine lead and zinc, silver is often found in association. We know of the acceleration in metal refining at the beginning of the Iron Age from the lead content of ice cores. Economic historians associate this with the extraction of silver, and thus money circulation. And while that is interesting, for the purposes of a technical appendix, I can rest on the observation that getting silver out of complex ores is actually pretty challenging chemistry, by old time standards, and associated with the larger traditional "forest" chemical engineering industries of soda, potash and glass production.  What differentiates an era of technological progress, population growth, and state formation from one of technological regression, population stagnation, and state collapse? For this post, anyway, the acceleration of manufacturing across a wide range of industrial sectors. While that's a pretty anachronistic way of describing what was going on in the early Iron Age, I did start with The Dragon and the George. The cancellation of major aircraft development programmes is going to be a persistent theme in 2018's technological appendices and beyond, and just perhaps we can learn something from them.

On Thursday, 4 January 2018, I posted "When America Gets a Cold," which was a discussion of one of the early post-war joint British-American development projects, an attempt to develop an improved naval anti-aircraft gun and associated director. It was also a reference to the way that the post-1948 election recession affected the rest of the world, something that I think I can leave out of further coverage until lit is actually happening on the postblogging schedule.

The 3"/70 appeared on a small number of warships, none of which had occasion to fire their guns in anger.

I use that excuse to update the blog on the status of the Royal Navy in 1948, something that doesn't always come out clearly in contemporary news coverage, and which has been the subject of a classic of modern naval history, Eric Groves' From Vanguard to Trident. One of the reasons that the contemporary coverage doesn't always align with modern scholarship is that Groves has to skate over the details sometimes, and recovering the fine-grain details and following them in new directions is what these appendices is all about. That's my excuse for talking about ships that weren't intended to mount the 3"/70 in the appendix.

So, what's the new direction? The director, not the gun. For this appendix, the issue is the control engineering, not the aborted effort to computerise it. Definitely worth investing some time in, since control engineering, under the flavour-of-the-day label of "cybernetics," is going to be a very hot topic for a few years. Until just about the time that the character of Ant-Man premiered in Tales to Astonish in 1963, to be precise.

Continuing with talking around early computerisation,on 24 July 2018 (my birthday, because I am industrious!) I gave the world "Somewhat Germanium," in which I laid my markers down for an argument demythologising Silicon Valley. I mean, why take on small targets?

Plus, by this point it's a sitting duck
The core myth of Silicon Valley is that on 15 December 1947, John Bardeen, William Shockley and Walter Brattain were hanging around Muppet Labs when one of them said, "Hey, what if the semi-conducting effect seen in germanium is due to a depletion effect in the electron network in the surface of the crystal quantum-blah-blah logic gate?" And so they took a piece of pure germanium, prodded it with some soldering guns, and caused it to wake up, achieve the Singularity, become a General Artificial Intelligence, attempt to launch the world's atomic missiles at each other, discover that we didn't have atomic missiles yet, and settle in to wait for Elon Musk to invent them.

In reality, after a diversion through some early British computing projects, there is a pretty rich history of semi-conductor-based, solid-state devices in 1948, something that is covered in the Wikipedia article to some extent, but not nearly in the detail this particular historian of technology would like to see. Also, it's not nearly facetious enough. The theory for which the Bell Labs trio received the Nobel for, seems to  have been developed in connection with work on selenium rectifiers (devices for turning alternating current into direct current; there's one in your car), two decades before. Probably. And it was also developed at the industrial laboratory of Federal Telephone and Telegraph, an electronics company founded at Palo Alto in 1907. While I haven't managed to excavate a blood-gorged Stanford University from its burrow in American industry's flesh, I have at least pushed "Silicon Valley" back a good half-century. In the end, I suspect it is going to come back to the railroads and, in particular, the Navy.  (The linked enthusiast's site just says FTT's early days saw enthusiastic support from "local capitalists.")

May, 1948 was a pretty rich month for technological appendices. On 30 July, I posted "Mr. Smith Goes to Ground," which was about the Smiths' Electric Autopilot, the commercial version of the RAF service autopilot, which received a publicity drive in the spring of 1948 thanks to the appearance of the SEP2 (Service Mark 10), and strong competition from Sperry. (Honeywell has gone dark, although it has definitely not left the field.)  I have a fairly long introduction to the history of British autopilots at the link, mainly involving, again, the control engineering issues. There's a great deal more here to say. The aircraft autopilot is pretty much the paradigm case of the "automation is going to take yer jerbs" panic that has been depressing wages this past decade. If a plane can land by itself, why can't a truck drive itself, ask thumb-sucking future-prognosticators? While I have very good answers to this question, I would point out here that the truck drivers who deliver to retail outlets have to swamp their own trailers, and an automated truck would need a swamper. While it is hard to see trucks driving themselves into loading bays, it is very, very easy to see a truck cab outfitted with an autopilot that doesn't work, and a minimum wage swamper doing the actual driving, at a $10/hour saving on labour.

On 23 August, I posted a "half-and-half" Technological Appendix "From the Oder to the Spree," that I can't very well ignore, because it went into the issue of control engineering, via the seminal textbook on the subject, reviewed in the June numbers of the engineering press. The title is inspired by the forbidden first verse of the Deutschlandlied [relevant version] that traces the German Fatherland's borders on various rivers that are no longer its borders, changing the watercourses in question to the rivers of swampy Berlin. And, of course, it is the Berlin Airlift I had in mind --the big event of 1948, even above the Marshall Plan and Truman's victory.

On 15 September, "Berlin by Night," I talked about aircraft radar. An apparently obvious development of air intercept and ground mapping radar, the navigational radar proved to be an enormous challenge, mainly because of difficulties processing signals. As far as the coming of modern times goes, this is the most obviously important of all the Technological Appendices this year, because it was the need for better signal processing in avionics that led directly to the transistor, and, via the independent development of the printed circuit, the modern computer age. The relationship between landing enough coal in Berlin and dropping atom bombs out of jet planes is a fascinating one.

Speaking of planes landing on the ground, and not in it, on 29 September 1948, I took on runway lighting. This is a super-fun subject, since it starts out with FIDO, one of the goofier wartime technological wonders. There were plenty of runway lighting schemes before the RAF came up with the idea of lining selected airfields with rows of giant Bunsen burners, but they didn't get nearly as much press as the copy-cat American installation at the airport of sleepy Arcata, a northern California college town of somewhat weird reputation and terrible weather.

The original idea behind FIDO was that the open flames would "burn off" fog. The effectiveness of this might or might not be debatable, but it turns out that a row of gigantic flame jets is pretty easy to line up a landing on, with the core concern being that the plane might not want to land between parallel rows of flamethrowers. (British Pathe.) The Arcata experimenters soon put the bunsen burners on platforms, providing pilots with a convenient alignment that wasn't nearly as threatening. And while the story is more complicated, there is a guiding light leading us to the modern era, when airports are lighted with standard schemes in which a pilot who keeps the lights lined up properly can be assured that the aircraft is coming in on the right line and the right altitude. High intensity lights aren't particularly high technology, but we live in a world where you're pretty much expected to fly or drive anywhere outside of your greater metropolitan transit collective, and that didn't happen without a glorious triumph over the inconveniences of night and weather.

I've decided that a Technological Preface is as good as an Appendix, and on Saturday, 16 October, I posted one that explored the Groundnuts Scheme. The Groundnuts Scheme was an attempt to develop a largescale peanut-growing enterprise in a lightly populated region of central Tanzania. The outlines of a successful scheme are pretty obvious. The region has adequate but uncertain rainfall, so you need some kind of irrigation intervention. This exotic technology exists. It's called "dams," and the newfangled ones even generate electricity. Second, you need a way of getting the crops to market. A railway running to a deep sea port is the usual preference, although things start to get a bit tricky here, in that you need a bit of industrial infrastructure to make the trains run on time, or at all. The citizens of the port are also likely to demand general regional economic development, so that they're not bound to a single crop. Then you need some kind of marketing intervention to stabilise the export price. Since that sounds expensive, the Scheme ended up going with a narrow gauge pretend railway, instead, and it never amounted to anything.

The Groundnuts Scheme of course failed, which is, in itself, drives me a bit nuts. It was never about the peanuts, which have a limited market, but rather about the peanut fat, or "edible oil." By the time the world's agricultural economy righted itself in the mid-Fifties, there was simply no demand for that much more edible oil, so in retrospect the Scheme looks a bit daft. But the reason that Tanzania lost out on peanuts has a great deal to do with the fact that petroleum products took its place.

The Erie Canal did not start with buying everyone a mule.
Now don't go staring at me with pity in your eyes and thinking, "Oh, Erik's off on biofuels again." I have an argument. It's about industrial fabric crops, which peanuts are, also. In other words, Erik is off about rayon again. So we've lost going on  seventy years of Tanzanian regional development, and biofuel production largely because we set out to do the Groundnuts scheme on the cheap. This is the dark side of technological history, where we find purported inventions being advanced as excuses for avoiding the investments needed to carry a scheme through to completion. In this case, the "Shervick," a tractor based on a Sherman tank chassis, because tanks are big and powerful, so obviously a Shervick with a tree-pushing-over attachment would be able to clear central African acres for cultivation by the millions on the cheap, ensuring a quick return on the investment that can then be ploughed back into the Scheme for further, incremental improvements.

On 14 December, I had a mostly technological appendix, which mostly counts. In Hydraulic Despotism, I wanted to talk about a goofy postwar trend of making computers out of hydraulic parts for modeling economic systems, but "Hydraulic Despotism" was a dig at a fashion of the time of presenting assorted ancient and modern "empires" that dared to have state-operated hydraulic agricultures as being necessarily despotisms because the Guvmint could turn off the taps. This was particularly apropos, I thought, because one of our computational economists was Bill Phillips, he of the Phillips Curve, which has become a modern shibboleth of "neo-liberalism," by which I understand the claim that any time it looks like the employment rate is going to fall low enough that the worker will finally have some clout, central bankers should engineer a recession with a timely increase in interest rates. The possibility that the workers might resist in the only way left open to  them --a birth strike--  does not appear to have arrived on the policy agenda yet, and presumably won't until we misplace a more important country than Japan.

On 25th December, I found myself in the odd situation of being fully caught up with the blog, and wanting to mark the day, I posted "Santa's Sleigh," which took on the idea that the giant postwar airliners (mostly the Brabazon, but also the Saro Princess) failed because they were too  luxurious. My point was that they had more space per passenger than they really needed for the same reason that every other giant airliner down through the 747 did: You couldn't make the pressure hull smaller. I hope you liked my eclectic selection of Christmas carols, because I get really, really tired of the determinedly non-eclectic ones I hear at work. If I have a serious point, it is that the development of civil aviation is unthinkable without state support, take that neo-liberals.

The last technological appendix I worked on in 2018 was posted three weeks ago, so the dateline is 2019, but I'm already cheating, so . . .   "Prestressed Concrete" is about --No, I'm not going to tell you, guess!

Okay, no.

By Michael Schmahl - Own work, CC BY-SA 3.0,
Prestressed concrete is lego blocks of concrete with superior longevity and resistance to impact and shock, allowing the use of thinner sections and longer members. It is also easy to cast (and pretension) at one location, and then transport and install at other places. Since concrete is ugly, and, by itself, not a particularly desirable construction material, unless you happen to be building a fortress or dock in immediate pre-WWII Europe, lego blocks of prestressed concrete tend to be hidden. Either they are used in the kind of industrial deserts inhabited by high-capacity train lines, where pretty much the only humans with business in the area are the ones riding by at high speeds on the way to somewhere else, or they are buried. Perhaps they are buried underneath facades with better insulating capacity and a gentler touch, as in, say, a shopping mall. Or, perhaps, they are literally buried, as "rafts" of solid material floating on muddy soil with little flotation.

I started writing this when I realised that weird and atypical residential neighbourhood between 16th and 41st Avenue between MacKenzie and Arbutus, through which I cycle daily on my way too and from work, had to have been built on a swamp.
It turns out I was right. That's the next block over from the Valley Route, from the old days when it was "Asthma Flats," or "Consumption Hollow," a neighbourhood of cheap homes built on a 1920s clearcut, under the benevolent eye of the rich people on the other side of the railway tracks. With the Arbutus shopping mall and Arbutus Club as  honorary members of the eastern, uphill, Shaughnessy side of the line. I'm not sure how many actual poor people died of lung diseases in the Flats before the gradual encroachment of postwar wealth drained the swamp. Probably not many, given the decision to build Trafalgar School right right in the middle of it in 1946. It remains the case that the area was conquered rather late, and that prestressed concrete house rafts, I suspect, had a great deal to do with it. Or maybe it was lots of drainage culverts. That would have been prestressed concrete,too!

I appended a discussion of an aborted new standard British infantry rifle, the EM-2 discussion, because I am fascinated with the way that human society understands and measures the landscape. Often, it takes existential struggle to really dig and appreciate the difference between solid ground and mud. In this case, the British and Canadian armies had had a  hard fight in the mud of the Netherlands in 1944/45, and I was alerted to the role of prestressed concrete in rebuilding those landscapes by a 1948 series in Engineering. There are a number of ways that you can see the EM-2 as a belated realignment of infantry tactics with the realities of our modern, post-concrete landscape, but I continue to fumble and digress in my attempts to articulate them, and won't further trouble the reader with my self-indulgence.

So that's a year, more rather than less, in Technological Appendices. I'm bookending two "hidden" domestic technologies here quite deliberately. Forced air heating and prestressed concrete affect the way we live every day, and yet we hardly ever think about them. Climate change means that we are going to have to change the way we live in very profound ways in a very short time, so an inquiry into how we live might well be in order. Appendices are not where you make arguments, I'm told, so I won't BIOFUELS be making any BIOFUELS arguments here BIOFUELS.  Also maybe BIOFUELS trying to get equal time for BIOFUELS boring old civil engineering BIOFUELS alongside the Silicon Valley version of the last BIOFUELS history of the last seventy years. 


  1. So, the BIOFUELS.

    Do not want.

    Photosynthesis is (at best) 5% efficient.
    Conversion of plant matter into fuel is (in a happy future of implausible tech) 50% efficient. (It's tough to find a critter that's more than about 30% efficient in general digestion terms.) This process will outgas greenhouse gases.

    The resulting fuel will get burned in an engine; realistic ICE efficiency is 25%, very best attained is 43%.

    0.05 * 0.5 * 0.25 = 0.00625
    0.05 * 0.5 * 0.43 = 0.01075

    At best we're getting 1% of the sunlight per area turned into motive power.

    (That 43% efficiency is an immense marine diesel; diesel engines produce black particulate carbon which is a climate forcing and a health risk distinct from CO2.)

    Current tech solar PV can be stuck on windows as film at around 7%; bought readily at 15%; had for extra at 30% (meant for aircraft applications!) and exists for space applications in expensive multi-layered forms at 45% efficiency. Unlike combustion engines, which are at the end of their logistics curve, we can expect improvement in price/performance of solar PV for some time.

    This goes into and out of a battery at about 80% cycle efficiency; it goes into an electric motor at 95% efficiency.

    0.05 * 0.8 * 0.8 * 0.95 = 0.0304 (dingy solar film)
    0.15 * 0.8 * 0.8 * 0.95 = 0.0912 (common or garden panel)
    0.30 * 0.8 * 0.8 * 0.95 = 0.1824 (fancy PV panel)
    0.45 * 0.8 * 0.8 * 0.95 = 0.2736 (extra-fancy space PV)

    Even the "this cuts our cooling bill a little" dingy film is three times better than making biofuel in energy efficiency terms.

    Realistic numbers are better by an order of magnitude for purposes of moving your car.

    Arable is going to decrease sharpish during the time of angry weather; fats (in a zero-fossil-carbon-extraction economy) are going to need to be turned into many things, including lubricants. I expect we're going to see post-food recovery as feed stocks to plastics and lubricant synthesis.

    And it turns out if you just want jet fuel, direct synthesis -- crack atmospheric CO2 for the carbon, crack water for the hydrogen, commit chemistry while venting oxygen -- has a better overall efficiency than "grow and digest plants".

    This is without getting into the parts-count fundamental advantage of the electric drivetrain; you've several times correctly noted that the Napier fondness for sleeve values was an error. Needing values at all is the same error once there's an alternative.

  2. While I appreciate the argument that we aren't necessarily at the same point in the yield curve for solar-electric as for internal combustion, I find it very hard to see how chemical>electrical>mechanical is ever going to beat chemical>mechanical. For cars, yes. Electrical has significant advantages in city driving, and most future driving will be city driving. A regenerative engine -- I've never seen anyone talking about regenerative braking in electrical autos, but surely there's a wild-eyed schemer out there somewhere-- would be even better. Electrical rail grids promise decarbonised long distance bulk movement of goods. Yes.

    But for bulk energy storage, key to aviation, marine applications, and heating cities, I'm not persuaded . . . Also, isn't the climate forcing effect of carbon soot in the cooling direction? I'm no fan of climate engineering, or, for that matter, smelly old diesels, but biofuels is something we can actually do --it's not pie in the sky by and by.

  3. The Chevy Bolt has a regenerative braking mechanism. So it's in production for automobiles, rather than wild-eyed scheming.

    Smog moves the heating up; it gives lower ground readings and a hotter atmosphere. "Black particulate carbon" is a nasty health risk and increases anthropogenic climate forcing at ground level. Heating cities works just fine with heat pumps, which are much more efficient, don't atmosphere dump anything (air quality in cities is important! look at Beijing or Mexico City), work to cool as well as heat, and are already quietly taking over in new construction. Better efficiency of building design can do a lot, too; building to the standard sheet material size isn't optimizing anything except initial capital cost.

    Aviation does work electrically. People are looking at pure-electric short-haul stuff; hypothetical graphene batteries go a long way, there. But even if we do need to synthesize jet fuel, direct synthesis works better than biofuels. The US Navy has (unsurprisingly!) put a fair bit of research into this.

    Marine applications, indeed any pumpable-fuel application like combine harvesters and trains and heavy construction machinery, are candidates for ammonia fuel cells; easier to synthesize than hydrocarbons, better candidate for useful proton storage (which is what all the fuel cells run on) than any hydrocarbon, NH3 lets you use alkaline fuel cells with a ~65% theoretical efficiency that's functionally twice even your expected marine diesel so you're ahead on system mass/volume requirements.

    The really killer problem with biofuels, though, is still the "decreased arable"; we're not likely going to be in a position to divert food into fuel. (I am not real hopeful of keeping whisky.)

    1. If electricity isn't a problem, then neither is arable. Giant greenhouses might be expensive infrastructure, but they're also great for load-levelling.

      Direct synthesis of hydrocarbons from atmospheric CO2 would be great. My issue, however, is that we've already got it! Do not scorn the chlorophyll molecule. What it lacks in efficiency, it makes up for in infrastructure costs. (I wildly suppose.)

      I will, however, grant the slow takeover by the heat pump. Sigh. I miss my air vents.

    2. Alas... greenhouses are more of a problem than a fix.

      Leaving aside the "fossil carbon plastic sheet skin" aspect (no such thing as a glass greenhouse in commercial use these days), current greenhouses work by outsourcing a lot of ecological services; for one small example, people capture wild bees (very specific wild bees!) for use in tomato greenhouses as pollinators. This puts excessive stress on the breeding bee populations, which are already pretty stressed from other factors. There's a lot more land area than the greenhouse required per productive greenhouse, and (particularly around inputs like compost) there's a really sharp upper limit on greenhouse area and productivity.

      Can someone use milpa techniques in a greenhouse and increase soil quality while demonstrating high productivity? Very probably, but it's not known art, it's fussy artisanal work (meaning "long apprenticeship", so slow expansion of capacity), and it's going to be priced out of the market right up until the point we desperately needed to have started ten years ago.

      Direct synthesis is something we already have, or at least DARPA does. This is one of the things the GOP congress has been blocking as a response to climate change. It'd be pretty darn straightforward to productize. ("Any sunny seashore" which raises the amusing prospect of the Middle East continuing to export hydrocarbon fuels, if in greatly reduced volume.)

      (You have seen the analysis of ethanol subsidies on corn and agave production? It doesn't seem to do anything good and it doesn't look like anyone would without the subsidy.)

      Heat pumps can have air vents! This (old, rented) house has both air vents and a heat pump, and they're connected! I mean, I'd take a hypocaust setup over air vents but there's nothing incompatible between air vents and heat pumps.

  4. The hypocaust: Truly a low tech solution to the problem of indoors, artisanal fungus production. And, yes, we can (and should) have convection heating, however the calories get to the housing unit. The problem is the designs that put in radiant heating and call it a day. More than enough calories are going into the room, so who cares if the windows behind the radiators are cracked open to a -20 Toronto night. (Russian district heating is apparently a sight to behold, with stray cats draped over aboveground ducts melting their way through snowy parks.)

    Now, I'm not against brute force CO2 cracking. It's an exciting future for when we have mega-amps to burn. Deep electrification is the solution, the problem being how we're going to get there. To a paleolithic mind like myself, that's not a problem: nukes everywhere. But I'm just back from a wearing Crooked Timber threat where any number of people seek to prove that we can never build another nuclear reactor, because, like going to the Moon, SSTs or full employment and positive population growth, it is a thing that used to be possible in the long ago, but which we will never be able to do again.*

    Against that, the enthusiasts, tasked with displaying optimism, talk about the bright future of renewables, which, curiously, never means renewables we can do now (hydro, biomass), but always renewables just at the cusp of happening. Hydro is always maxed out, which isn't even vaguely true. We have any number of big hydro projects waiting to happen, all being opposed by ad hoc environmental concerns mobilised to suit the case --a pattern so systematic that I do not think I am being entirely paranoid in seeing the influence of lobbying. Biofuels, too, are happening, the problem being that we're seeing them spread as soya plantations and palm oil in tropical forest land rather than coming home to rural Canada to intensify agricultural production, for the obvious reason that it doesn't pay.

    I can't help but add that, in said Crooked Timber thread, the ever-so-serious poster seriously proposes tidal power for the dense power generation of the future. Sure, sure, not impractical, although presenting some difficulties, but move from the pie in the sky to the Severn, and suddenly what is just on the horizon becomes utterly impossible for ever.

    Claims for efficiency of solar, and of imminent breakthroughs in storage, are just that, claims. In spite of talk of the novelty of solar power, and projections of heroic curves of improved efficiency into the future, we have seen on this very blog that the "solar house" is more than seventy years old, and while I grant that a lithium house battery is better storage than a tank of mildly corrosive industrial detergent, it's still inferior storage to an oil tank, and always will be, because you can store more energy per unit weight in a carbon-carbon bond than in any conceivable current-producing ionic separation. Solar's problems of periodicity and variability are inescapable. So are nuclear's problems of manoevrability, of course. Nothing is going to beat natural gas there without subsidies, which is the real issue, and the real story of the failure of "Certain Grand Schemes." They're known as Grand Schemes because they failed.

    Hmm. I don't think I've ever done a "Gathering the Bones" on the Erie Canal . . .

  5. *This comes up at work quite often, so I'm going to use this asterisk as an occasion to share solutions, not problems: Want your problems to go away when doing anything is impossible because reasons? Just go lie down in the highway, and they'll be gone in a flash.

    If that's not your solution, it might be time to ever so gently contemplate saying, "Fuck reasons."