brrt

My main guesses:

* multiple dispatch * the meta-object protocol

And the thing is, these features are being used all the time, because the core language is built out of them. If you want to see how a language which is not built on these things runs (on the same vm implementation, I might add), I suggest you try nqp :-).

https://github.com/nqp/nqp

GP means useful to the 'host' of the meme, that is, to you or me, while the survival and propagation refer to the meme itself. For example, a particular catchy tune might be very successful (very popular i.e. high survival and propagation) but it isn't necessarily useful (to me).

They also assume a wide grid and overcapacity, which balances it out. And their storage capacity is minor, and they assume relatively cheap backup power. Like I said, there are plenty of things wrong with it, but it's nothing like what is wrong with the OP.

Actually, renewable energy is, when you actually run the numbers in a sensible way, pretty cost effective: http://www.sciencedirect.com/science/article/pii/S0378775312... - this finds that with 90% solar/wind and modest amounts of storage, electricity would be cheaper in 2030 than it is today, and that the cheapest option is actually a vast overcapacity. There are plenty of flaws with that article, but still less than this post.

We burn it! That is, nuclear burn it, not chemically. And what can't be burned can be bombarded with neutrons into radiological inertia. And what can't be neutralized in that sense can be molten into a ceramic, more specifically, a glass, and stored underground. Which is of course where it came from, when you think of it. Seriously, nuclear waste storage is not a problem. This page gives an excellent eplanation: http://www.phyast.pitt.edu/~blc/book/chapter11.html

Nuclear energy has lots of problems, and I personally think this article is not worth this place on HN, but waste is not really one of them, especially compared to the scale of chemical wastes from fossil fuels.

It requires a lot of energy to do this, obviously. I don't know the exact details, but my naive assumption is that they require hydrogen as an intermediate, which would also imply they need relatively pure water as a source. Altogether I just don't know why they decided to take this from the air rather than build a plant in any major port with a refinery installed, which would seem to be much cheaper.

Cool stuff, certainly. But why take C02 from the air, when there are large stationary sources of CO2 readily available, like any bloody thermal power plant? It... seems very wasteful to me.

EDIT: also, one tonne per day, very wow.

Actually, yes. Global atmospheric CO2 concentrations will (probably) stabilize if we reduce output to about 5% of the current level, by outflow to the deep ocean (which is a vast reservoir). But that is still very difficult to reach, especially with business as usual.

I admire your optimism :-) but I'm not 100% sure that is true. If there wasn't a political drive behind this, wouldn't it have been easier to burn coal for the next 100 years?

> The main problem is that the networks are usually owned by monopoly utilities with interests in power generation. How much do you think they want to upgrade if the net result is more competition? Hell, they'd probably pay not to have to upgrade.

This... is exactly what is happening :-). Utilities and network operators are trying out ways to give incentives to customers who can respond with negative load (i.e. turning the power down). It makes economic sense because the alternative is increasing production at immense speeds, which is quite costly and inefficient (often using diesel generators, and a lot of them). Aluminium smelters like the examples you give (and quite a few other businesses) are doing precisely this.

Also note that an aluminium smelter works by electrolysis, not by pyrolysis (which is what GP seems to suggest?), which is why they use so much electricity and why they can in fact turn it on and off at a whim.

I think I agree with at least some parts of that description. (I have no idea who Judith Curry is). However, it is important to note that the phrase 'wicked' problem refers policy problems rather than science problems. And it is useful to disambiguate between climate science and climate (change) policy. Sure, climate science is hard, but not much more so than say, geology or microbiology.

Climate change policy is on the other hand wickedly difficult (despite climate change itself being well-established), because there are so many practical uncertainties and so many different stakeholders, in which you could reasonably include unborn people and/or nonhuman ecosystems, which have a hard time lobbying for themselves. However, much like poverty (which is a hard policy problem, for much the same reasons), just acting like it doesn't exist, or denying the scientific evidence, is sure not to help.

There are no dumb questions :-), but no, this is not what will happen. For one thing, the ice that floats around in sea displaces exactly as much water as ice (uses the same 'water volume') as it will as water. (Almost, because the water volume per mass will actually increase with increasing temperatures, which is quite a significant effect). However, a large amount of ice is on land, where it doesn't contribute to ocean volume. If it were to melt that would be a major contributor to sea level rise.

Also, the average humidity is quite stable on earth, and any excess water that evaporates into the atmosphere quickly rains out again. Furthermore, the amount of water in the atmosphere is very small in comparison to the amount of water in the oceans. (See e.g. here: http://water.usgs.gov/edu/watercycleatmosphere.html). So it is very difficult to imagine that an increased humidity would have a significant effect on the sea level.

I think that is an unrealistically optimistic and friendly interpretation of Christensens' position. I agree that the idealistic aspect of disruption probably explains it's popularity (not to mention the opportunistic aspect). But Christensen cum suis have been propagating disruption as a fact for nearly twenty years, and it has had an enormous influence on the public debate.

And as usual with influential ideas, it's not actually true, and (I think) a very reasonable amount scepticism would have demonstrated that much earlier. As an academic, I really think the global community could have expected more honesty from Christensen.

Fascinating to see how well a distributed architecture with relatively low processing power coupled with abundant sensors works out. Very similar indeed to biological systems, and quite humbling with respect to our current 'big data' machine learning algorithms that require entire clusters. (Although these do very different things, of course).

I expect that as these robots would become more clever, more 'implicit' centralisation would enter the structure, much as they do in biological organisms. But I don't know how far this line of research has ever been followed.

It kind of looks valid to me, yes.

You are absolutely right, I oversimplified, but that has the advantage of giving us plenty of other experimental bodies to compare with (i.e. the solar system). For what it's worth, this oversimplification gives a pretty good average temperature of 15 deg celcius on earth.

Albedo is an important factor, but there is actually very much uncertainty about human influence on earths albedo, and at any rate the effect seems to be pretty small. Humans' effect on CO2 on the other hand is extremely well-established and much greater, which is why it's logical to focus on it.

Norway is not comparable to mainland EU, I'm not claiming that. But Norway does have an intensive oil and gas industry which emits copious amount of CO2. Because of the CO2 tax, they have implemented the worlds first (and only) commercial carbon-capture and storage (CCS) project (https://en.wikipedia.org/wiki/Sleipner_gas_field#Carbon_capt...).

As for China stealing European industry, that's happening already, and it's really nothing a little protectionism can't help. Protectionism is bad, but so is losing jobs, and worse is acting like you are powerless when really you aren't.

No, you are wrong. You are clearly posting out of ignorance.

The earth climate system is not impossibly complex. The global temperature is at it's core determined by an energy balance. Visible light from the sun enters, infrared radiation leaves. Some of that infrared radiation is blocked, then temperature has to rise before the balance is restored. Simple as that, and completely uncontroversial, because proven by spectrum analysis of our neighboring planets (and the moon).

Now I know that doesn't give you hard answers to questions policymakers want, like 'will New Orleans flood from increased tropical storms? Will the Middle East and India dry out?' We don't know that, and yes, that sucks.

But we do know the effect of increased amounts of greenhouse gases. (And, for what it's worth, we do also know that carbon dioxide taxes work much better than an emission trading system does, as proven by e.g. Norway vs. European Union, and as proven by the fact that industry and fossil fuel companies actually prefer it).

And the fact that the top comment on HN is about climate change denial rather than the actual article is excellent proof of the study's truth.

Actually... yes there is. At least there is a limit to the economic value of art. Given a finite human population with a finite lifespan (also implying a finite rate of reproduction), there is only so much attention that can be given to any form of art created. If a world would start creating art non-stop (which is pretty much what is happening now) the economic value of an average piece of art must drop to zero for the simple reason that people haven't the time to observe it. (Assuming that art can only be valuable economically if it is observed, which is kind of optimistic actually). Also pretty much the state in which we find ourselves now already.

A finite earth can support a finite population which can use finite resources to employ a finite economy. I'm not sure what's difficult about that concept; especially since a zero-growth economy is not a new concept (it's called 'maturity' when applied to an ecological system).

Oh. Well, if that's so, I'm not sure why they've chosen to use photovoltaics at all, because solar thermal should give you a much better cost-performance ratio, and still give you plenty waste heat to use as heat.

That's possible because they count both photovoltaic and thermal conversion efficiency. I'm not entirely clear on whether they want to use the heat as heat or always for electricity generation. They claim GaAs solar cells get 40% efficiency, which I'm ready to take at face value. 60% of the energy is then converted to heat. For a total of 80% conversion efficiency we'd need only 66% efficiency for the thermal-to-electricity conversion, which is well within theoretical reach, given they also claim >1000C achieved temperatures.

On the other hand, their setup claims to melt iron, at 1518C. Which would give you a pretty good carnot efficiency if you'd use it directly to fire a steam turbine.

Really cool work. The upshot of it is that SIMD optimizations can work within traced numpy calls. This is preferable to calling out to an optimized subroutine, because the first lets you keep your registers intact (and your trace) and the last does not.

I wonder what issues prevent them from tracing the happy-case list yet.

A large problem with current generation nuclear is that the easily obtainable uranium isotopes are in the process of running out (in no small part due to artificial limitations on e.g. reprocessing of fuel). We'll probably continue to have enough for some time to come, but a large-scale switch to nuclear energy is actually quite unlikely at this stage.

I don't mean to say that this is the primary reason for nuclear's demise - it isn't - but it is a real hurdle against increasing deployments.

Of course, there are very exciting new developments in nuclear technology that solve this - breeding, reprocessing, etc - but they are all (still) at an experimental stage. That's the catch-22 of nuclear: there's an existing, mature industry, which uses it's fuel unsustainably, and there are developments to sustainably use nuclear power, which is quite far from being an industry.

There's also, at least on a global-energy-system scale, the issue of efficiency. Solar thermal (I'm thinking of solar towers and the like) can heat a medium to about 300, maybe 400 C, and if I'm not mistaken using a conventional steam turbine that can give you a Carnot efficiency of ~25%, maybe 30%. Real generators engineered today are on the level of 90% efficiency, meaning that a solar thermal plant can - practically - convert ~20% of incident solar energy to electricity. Commercial PV will likely not reach that level of efficiency for years, and there are - probably - good ways to improve thermal efficiency beyond this 20%.

Of course, thermal generators required direct incidence (don't work well on diffuse light), but the long and short of it is that with thermal we (may) need about half the total area we would need for PV. (On the other hand, PV can practically uses space - rooftops - which is inaccessible for thermal.)

Nice blog, by the way :-)

This is calculated in energy return on energy invested (EROI). Solar is doing.. ok in that regard: http://rameznaam.com/2015/06/04/whats-the-eroi-of-solar/

The comparable number for fossil fuels (in this case, oil) is 'well to tank efficiency', which amounts to about 25% for oil, if i recall correctly. So 1/4th of energy contained in oil is already spent to bring it to market.

Notably, wind energy has a far more favorable EROI than solar does (currently), but wind power is a relatively limited resource.

You're setting up a strawman against a non-essential point in the article. This article was not about the difference between apples and candy.

Betz's law arises from the fact that you can't very well stop air dead after you've extracted energy from it - it has to keep on moving to make space for new air. (Otherwise pressure would build up indefinitely, a clearly non-physical situation). So it applies to all such devices, rotors or no. (It doesn't apply to fluids because fluids don't compress, but that is a completely different story).

Uhm, I think you're reading my post a bit uncharitably. I don't think I insulted you, for instance.

I never said that estimates were meaningless, but they are tools for planning with high intrinsic uncertainty, and translating that uncertainty to evaluations gives unfairness. People are very sensitive to (perceived) unfairness and the idea that they will be treated so will give rise to resentment, not to mention incentives to game the system and overestimate. Hence endless bickering about points.

If agile failed for any reason it was misunderstanding human psychology.

Personally I don't think human performance in creative professions is well suited to quantified analysis, especially when team efforts are involved. So I don't really have an answer there.

What? No. Glad I don't work for you. Nearly all developers I know estimate time (and storypoints) by some (often linear) function of volume and complexity. Volume estimation is usually not very difficult, but complexity is very tricky (and dynamic too). Estimating developer work hours by any proxy that depends on complexity will introduce a margin of uncertainty so large as fo render the number meaningless.

I'd think by and large you would evalute remote workers by theit work output, not by estimating how much time they've spent.