Existential risk and growth

Here is the abstract of a new paper by Leopold Aschenbrenner:

Technological innovation can create or mitigate risks of catastrophes—such as nuclear war, extreme climate change, or powerful artificial intelligence run amok—that could imperil human civilization. What is the relationship between economic growth and these existential risks? In a model of endogenous and directed technical change, with moderate parameters, existential risk follows a Kuznets-style inverted Ushape. This suggests we could be living in a unique “time of perils,” having developed technologies advanced enough to threaten our permanent destruction, but not having grown wealthy enough yet to be willing to spend much on safety. Accelerating growth during this “time of perils” initially increases risk, but improves the chances of humanity’s survival in the long run. Conversely, even short-term stagnation could substantially curtail the future of humanity. Nevertheless, if the scale effect of existential risk is large and the returns to research diminish rapidly, it may be impossible to avert an eventual existential catastrophe.

Bravo!  44 pp. of brilliant text, another 40 pp. of proofs and derivations, and rumor has it that Leopold is only 17 years old, give or take.

If you happen to know Leopold, please do ask him to drop me a line.

For the pointer I thank Pablo Stafforini.

Comments

Hail Leopold Aschenbrenner!

And hail Pablo Stafforini too!

(Oddly, one can translate 'Aschenbrenner' as 'ash burner,' or someone burning plant material for potash. Which leads to this, from the potash wikipedia article - 'The first U.S. patent of any kind was issued in 1790 to Samuel Hopkins for an improvement "in the making of Pot ash and Pearl ash by a new Apparatus and Process".')

"In a model ... with moderate parameters": well, thank goodness they are moderate. We wouldn't want any sort of extremist parameters around here. No antifa or Literally Hitler for us, no sirree.

Hm, maybe that sounds a bit snide. So let me be more positive. Sonny, you have to tell us, in the Abstract, how many adjustable parameters your model has.

I liked you better when you were snide. "Sonny" has, while still a teenager, produced a paper that is better than anything I (and probably you) have ever written, or ever will write.

I've written some awfully good stuff though I've no idea how you would compare it to Sonny's. But inevitably I have far more experience of mathematical modelling than he has so I know that always, for the reader, a crucial question is "how many adjustable parameters does your model have?"

That's because the reader wants some feel for how easy it is for Sonny to tune his model to give the results he desires. It's only a rough feel, but even a rough feel should be better than no feel.

I was going to say that I have literally never seen a paper that announces the number of "adjustable parameters" in its abstract, but then I second-guessed myself, because maybe I'm wrong...

Nevertheless, using JSTOR's search feature and looking for "adjustable parameters" in the abstract [(ab:("adjustable parameters"))] yields exactly 12 hits. Removing the quotes increases this to 20.

"having developed technologies advanced enough to threaten our permanent destruction, but not having grown wealthy enough yet to be willing to spend much on safety."

So far, humanity isn't close to having the technology to destroy itself. Of course, A.I. could somehow change that but climate change and nukes? You don't need 40 page of proofs to see this is not possible.

"Not possible" to destroy all of humanity? Most likely. But shoot itself in the leg and postpone any progress for a veeeery long time or indefinitely - pretty possible.

Imagine us nuking 50-60% of inhabited land and losing 80-90% of people that way. Do we as species survive? Yeah. Does our current civilization? Not likely.

Or imagine that due to climate change most of Africa becomes inhospitable. Poor sods will mass migrate. And history shows that mass exodus brings a lot of problems (wars, famine, hunger, pestilence.... you know... the usual). Just recently Europe had a few million of refugees come and sow panic in every politician, who did not know what to do with this problem. Now imagine a few hundred millions.

Have to note that just because something can be imagined does not necessarily imply that it is even remotely probable.

Just recently on this cite there has been a lengthy discussion (what-is-the-probability-of-a-nuclear-war.html) about exactly how probable nuclear exchange is currently. According to that research it is about 0,4% every year. The methods used in the research are somewhat questionable and are worth discussing, nonetheless, many researchers think that nuclear exchange is not that improbable. Let's just say they are pretty high for something that might mean destruction of current civilization. Certainly one of the most pressing and realistic Armageddon options, and we are the ones responsible for that.

Nuclear war maybe. What odds do you want to put on Africa becoming uninhabitable?

As a joke, I would say: 100%, it's barely inhabitable right now.

As a real answer: as always, it depends very much on state actions of the nations of Africa and all of the world. Africans themselves can try to revert desertification.

Probably it will become somewhat less hospitable, but the level of changes can vary a lot and therefore, the consequences can vary a lot as well.

"Imagine us nuking 50-60% of inhabited land and losing 80-90% of people that way. Do we as species survive? Yeah. Does our current civilization? Not likely."

It isn't possible to kill more than 50% of the people at this point since so many still live away from cities.

Of course something very similar to our current civilization(s) would survive. What are the remaining 5 or 6 billion people going to do? A lot of cities and far more towns would remain and the partially destroyed ones would rebuild as Hiroshimi and Nagasaki did.

It is science fiction to think that after a large nuclear war that people would not remember how to make roads, houses, computers, reality TV shows, etc.

"It is science fiction to think that after a large nuclear war that people would not remember how to make roads, houses, computers, reality TV shows, etc."

+1, most dystopian futures are high implausible escapism. I stopped watching The Walking Dead at the end of season 3, because at that point it became implausible that the humans hadn't started killing the zombies in mass and rebuilding civilization.

After the Romans left England the remaining natives did not know how to build buildings and bridges as the Romans did.

Shh - actually pointing out well documented examples from history is not as satisfying as talking about how a zombie series became unrealistic after a couple of years.

Yet another one of your passive aggressive posts that contributes nothing to the argument, other than expressing your apparent displeasure that someone is making a rational point that you disagree with.

'that someone is making a rational point that you disagree with.'

The point being that zombie shows are realistic, up to a point? You are right, I disagree completely with the idea that zombie shows are realistic in any way, shape, or form.

And if you thought the point was about rebuilding civilization, the Roman experience in Britain is one of the better documented examples of how the ability to build things like roads can collapse for centuries, not merely a couple of decades.

So then this is really about you just having poor reading comprehension.

My comment: "most dystopian futures are high implausible escapism"

See that's the point where I indicate, pretty clearly, that zombie shows are not realistic.

Your rebuttal: "I disagree completely with the idea that zombie shows are realistic in any way, shape, or form."

This is the part where it's clear you didn't understand my first comment.

And here I was, thinking that since we were talking about rebuilding civilization, this was the relevant quote - 'I stopped watching The Walking Dead at the end of season 3, because at that point it became implausible that the humans hadn't started killing the zombies in mass and rebuilding civilization.'

My mistake, obviously, in believing that one could read that sentence to mean that you believed, in contrast to the third season, the other two seasons were more plausible while you watched them.

So, you ignored the first sentence. Then took the second sentence out of context and thus came to a flawed conclusion.

Ok, duly noted.

"After the Romans left England the remaining natives did not know how to build buildings and bridges as the Romans did."

Sure and if all the literate people were to leave Earth, the survivors would have a rough time of it. But knowledge is widespread and well documented today. A third of Americans have college degrees and at least 2/3rds are literate enough to follow a Chilton's guide to repair a car.

What was the literacy rate in England at the time? 1% or so? The ability to read and write makes a huge difference.

'But knowledge is widespread and well documented today.'

As was the knowledge of how to build Roman roads throughout the Roman Empire, over centuries.

'are literate enough to follow a Chilton's guide to repair a car'

Ever actually machined any parts to fix a modern car? That would be the easy part, compared to dealing with how the car's motor is reliant on IC components that you cannot simply conjure into being with literacy and a college degree. A 1972 VW bus? Yep, you probably can keep that thing running for decades in a maintenance sense (that liquid fossil fuel infrastructure is probably one of the highest priority targets in a nuclear exchange is another subject - gasoline has a stable shelf life of something like a year, and the way it gums up is a real pain to deal with - a carb is not that hard to clean, but a fuel injection system is a larger challenge), particularly if you have access to a machine shop.

The same broad perspective applies to the ability to build Roman roads, though in that case, it was not literacy that was essential, it was a functional administrative state able to direct resources.

To be honest, it is pure fantasy to think you can keep basically all automobiles built in the last couple of decades running after the IC chips are unavailable - which, for cars built more than 20 years ago is already happening. Interestingly, car collectors are fully aware of this fact, and much prefer acquiring vehicles they can keep running for decades into the future - ones that have zero IC technology anywhere.

To be even more concrete about losing knowledge - the U.S. was able to build battleships with 16" guns in armored turrets. After the accident in 1989 in a turret of the Iowa, it turns out, even with literacy, we are no longer able to actually construct such massive armored structures, nor even effectively repair them. It is not impossible, of course - after all, we have basically all the plans, and it is old technology. However, we saw absolutely no reason to retain that ability, and absolutely no reason to reconstruct that ability five decades after the Iowa was launched.

Welcome to road building after the collapse of Roman administration in Britannia. No one saw the point.

And as an honest question - you ever kept a vehicle running years after most parts for it were no longer available?

It's not possible just by the nuclear blasts. But it is quite possible by following nuclear fallout and collapse of all industries.

"It is science fiction to think that after a large nuclear war that people would not remember how to make roads, houses, computers, reality TV shows, etc."

Well, just remembering the idea of how to make computers does not help you make fabs for integrated microchips. It requires a lot of specialists and very precise instruments, currently located in a very few locations over the globe.

So if you think, that producing basic lamp or transistor-based computers is not a huge step-back for progress (it is at least 50 year-old tech).

Also just knowing something does not make it viable. If you travel to middle ages, just knowing modern chemistry does not help you a lot, when you first need to go through metal-working to produce lab equipment and there is no industry, nor there any need to support such things, when you try to get enough food to support just existing people.

Do you say that putting us back a hundred years and irradiating 50% of territory is not a huge stepback for humanity?

"It's not possible just by the nuclear blasts. But it is quite possible by following nuclear fallout and collapse of all industries."

This is what many assume but fallout isn't dangerous outside an area pretty close to a blast. 50% of land would not be "irradiated" - not even close. Nor would there be a 100 year technological setback. Any setback would be quite short and probably close to 5 or 10 years. I think you are forgetting that billions would still be around with maybe tens of millions or hundreds of millions of skilled workers spending on how you define "skilled."

'but fallout isn't dangerous outside an area pretty close to a blast.'

Air burst or ground burst? That is a relevant question, as is time scale. Basically, it is reasonable to say that fall out is a problem for months after a nuclear blast, and that the weather will play a major role in what areas are and are not dangerous. Ironically, the area close to a blast tends to have minimal fallout, as the fallout (the term makes the concept clear) is carried high by the blast itself.

And knowledge that people made CPUs in fabs does not magically grant you a clean room or restored supply chains of the necessary materials - not to mention reliable power supplies.

"Air burst or ground burst? That is a relevant question, as is time scale. Basically, it is reasonable to say that fall out is a problem for months after a nuclear blast, "

Air or ground burst is irrelevant apart from those living very near a city. Fallout would not be a problem after about a week to say nothing of months later.

You are also forgetting that billions of people and computers would still be around after a severe nuclear war.

'Air or ground burst is irrelevant apart from those living very near a city.'

Um, no. Here is a bit of text from someone trying to actually model fall out effects - 'But weapons that are designed to destroy command bunkers, or missiles in silos, are the worst for the surrounding civilian populations. This is because such weapons are designed to penetrate the ground, and the fireballs necessarily come into contact with the dirt and debris. As a result, they kick up the worst sort of fallout that can stretch many hundreds of miles downwind.' http://blog.nuclearsecrecy.com/2013/08/02/what-the-nukemap-taught-me-about-fallout/

Nuclear targeting is always extremely secret, but it is fair to assume that all of the ports where the U.S. Navy has nuclear capable ships or subs will also get a groundburst or two (based on the fairly reasonable assumption that such facilities have hardened bunkers). And that the U.S. would return the favor to its opponent, obviously.

Airbust? Deadly, but clean. Groundburst? Deadly, and extremely dirty. A ground burst at the Pentagon could easily cause significant problems from fallout in Richmond, or Baltimore (not both, however - it really depends on the weather).

"Do you say that putting us back a hundred years and irradiating 50% of territory is not a huge stepback for humanity?"

Technology would not fall back 100 years. There are literally thousands of manufacturing facilities scattered across the US in areas with lower wages.

We might stop making state of the art integrated microchips for a decade or two, but the technology to make low end chips and boards is much more common. The technology to make a 30 year old VCR isn't that impressive by modern standards.

Civilization would probably drop back to 0 progress for a decade or two, then play catch up to where we are today for another decade.

Call it a net loss of 30 years. It's nothing compared to the Dark Ages.

It's probably comparable to the Black Plague which killed 30-60% of the European population during the 14th century.

"Or imagine that due to climate change most of Africa becomes inhospitable. "

Africa had an enormously diverse and rich biome 4 million years ago when Australopithecus roamed the Ethiopian plains. The global temperatures at the time were at least 2C hotter than today. During the Eocene period the Earth's climate was 14C degree hotter than modern times. Climate change will not make the Earth inhospitable.

https://en.wikipedia.org/wiki/Geologic_temperature_record

Basically, we don't know that. The result will depend a lot on humidity and desertification. Right now Sahara is big and is getting considerably bigger each year. It is possible to revert those changes, but that does not mean it is easy to do even now.

I am also a proponent to hotter Earth, and I do hope that higher temperatures will make more hospitable climate for trees. But that also depends on whether we give trees time to grow and not just cut them down.

How much of the Eocene biome can a human metabolize? Our agriculture is highly dependent on the climate of the last 5,000 years. How much research is going into adapting that agriculture to the probable climate of the future? Wouldn't that be the safety research the paper explores?

'humanity isn't close to having the technology to destroy itself'

Well, in a strictly precise sense, sure. We are now able to engineer viruses, but some people would likely survive even an extremely well designed virus. And even after a massive nuclear exchange of the type possible in 1985 (humanity has not been been able to keep its nuclear weapons stockpiles from decaying), even if civilization no longer existed, sure, there would be easily millions (quite likely even hundreds of million) people left a decade after that exchange. The living envying the dead in the months following such an exchange is quite imaginable, however.

You will still have billions of people living far from cities that were nuked.

You may want to think about how many people are currently reliant on food exports from ports that will likely not be functioning after a major nuclear exchange. You may further want to think about epidemics after the collapse of sanitary systems and availability of medical supplies reliant on products exported from ports that are not likely to be functioning. And of course, the collapse of the current liquid fossil fuel industry is pretty much a given (no major global nuclear war is going to spare vast reaches of the infrastructure that supports the ability to fight a modern war).

In other words, think starvation and plagues throughout much of the world after the destruction of the systems that currently exist to keep such in check. The time frame of months was intentional - after enough people die, the remaining food supplies will then be adequate.

Of course, this is all extremely dependent on what sort of nuclear war we are talking about - but where do you think the Israelis are likely to bomb if a major nuclear war between the US and Russia begins? And would India or Pakistan decide that such a conflict leads to such instability that there is no reason for restraint in the absence of American guarantees? The cascade can continue, until it makes sense to someone to take out resources that an enemy can use to rebuild (though the better dead than red ideological framework has pretty much faded into history).

Basically, a nightmare scenario is just that, but at the same time, some of the people I grew up with were directly responsible for looking at such nightmare scenarios - or to carry them out, if so ordered.

clockwork actually makes a good point here. If the damage were bad enough to collapse the oil industry distribution system, then modern agriculture would lose a harvest season and food deliveries would be interrupted.

That being said, it's still seems highly improbably that there wouldn't still be a population numbering in the billions.

No, he's not making a good point. The oil distribution would shift, prices may go up but there would be no collapse.

"In other words, think starvation and plagues throughout much of the world after the destruction of the systems that currently exist to keep such in check"

Not throughout much of the world at all. Again, the types of food would change as well as where people would get it from but there is no reason to think there would be mass starvation on any continent.

The worlds poplation might go back to where it was in 1990 with 5 billion surviving and that is a worst case scenario. I doubt a full scale nuclear war would kill more than 500 million to 1 billion people but maybe 2 billion. It isn't as if 100,000 nukes are available.

'The oil distribution would shift, prices may go up but there would be no collapse.'

Um, let us just assume that between the U.S., the Russians, and the Israelis, the three manage to wipe out basically all oil transportation and refining infrastructure in North America, Europe and Russia, and the Middle East. Where are you replacing that shortfall fall from? Between North America, Russia, and the Middle East, you are talking about roughly 2/3 of all oil produced currently. And a higher amount of the oil actually refined.

And do keep in mind that the real point is refineries, which are probably no. 2 or no. 3 on the nuclear targeting list (depends on whether you see command and control separately from the silos). There will not be a market involving the areas involved in that nuclear war where prices shift after 2/3 of the world's crude is no longer being refined. Along with the fact that one can safely assume that every port which can take crude tankers tends to be exactly one where one finds refineries.

I recognize that the standard planning of WWIII is no longer really all that commonly discussed anymore, but for those responsible for planning it back in the 1970s, destroying oil infrastructure was just behind crippling nuclear infrastructure. After all, who worries about aircraft or tanks that have no fuel?

Starving your opponent is just one of those unintended side effects, of course. Much like hitting a nuclear sub base - the death of the surrounding port city is an unintended consequence, though fully predictable.

And this is not a dystopian scenario - this is the sort of planning that was ongoing among the people I grew up with, and one reasonably assumes is still ongoing today. One can safely assume that there will not be a single working refinery anywhere that the Russians can use 24 hours after the start of a nuclear war with the U.S. And one should safely assume that the Russians will return the favor. The Israelis are simply thrown in to the mix because what would they have to lose by utterly crippling their Middle Eastern opponents?

I think you might be underestimating the amount of time and effort spent planning for full out nuclear war over the last couple of generations.

I think all this back and forth is highly contingent on the scenario.

A 1985 full nuclear exchange between the US and the Soviet Union would have been about 20-30,000 nuclear weapons.

Today, the US and Russian combined have under 1,000 ICBMs.

Obviously a factor of 20x makes a difference.

Hey, it's a model and the he is honest enough to say in chapter 4 that he did not find data to parametrize the model. But that doesn't make the conceptual framework less interesting.

@Todd, nuclear weapons may have consequences beyond the immediate blast and radiation. It is not that improbable that a nuclear exchange is followed by a world war fought with whatever resources are left. Hiroshima and Nagasaki are exceptional cases because the bomb ended the war and resources where available to rebuild. If a nuclear war is just the start of a war, consequences will be much different. The billions of people far away from the nuked cities on day 1 of the war may still suffer the consequences such as famine or genocide.

What we know from WW1 and WW2 is that war-related famine and disease are quite bad for civilians. Around 5 million in WW1 and 20-25 million in WW2. The war casualties of the Spanish Conquest of America are in the range of thousands while famine and smallpox killed millions. Presumably, nuclear weapons are more disruptive of farming, fresh water sources , sanitation infrastructure and supply chains than older weapons.

Nuclear war, as any other war, may be enough to make nation-states collapse and the make it hard to protect civilians from genocide. It was not caused by war, but power vacuum may have triggered the Rwandan genocide. Genocides before the word was coined : Turkey -> Armenians during WW1, Mongols -> rest of Asia during their conquests 800 years ago. College degrees and being able to repair a car may be useless if there's and ethnic cleansing going on.

'If a nuclear war is just the start of a war, consequences will be much different. '

Yes, this is all extremely dependent on the scenario. For example, it is extremely dependent on timing - a nuclear war that starts just before grain planting season in the northern hemisphere will have different effects than one that occurs a couple of months after that grain harvested.

I'm talking about the worst case scenario. On a large scale, the above timing aspects just don't amount to much. That was true of the Black Plague as well. Different circumstances may have led to 1/4 of Europe dying or 1/3 - historically, it doesn't matter. A nuclear war in 2045 would have far, far less famine than what happened in 1945.

For further discussion, see the Effective Altruism Forum thread where the paper was originally posted:

https://forum.effectivealtruism.org/posts/xh37hSqw287ufDbQ7/existential-risk-and-economic-growth-1

Suspect paper optimistic. "Safety" spending confronts a collective action problem, or an enforcement problem if a single authority is doing it.

Watch the drone wars in the middle east, there may be some clues.

At 17, he's already smarter than Dick The Butcher and EdR!

That's nothing to be proud of, Russ

The model is interesting, it helps to think about long term risk in simple terms. The exponents epsilon and beta control how much the expenses in consumption increase risk and how much the expenses in safety reduce risk.

It gets more interesting when he tries to parametrize the exponents. I respectfully disagree with this assertion:

"Our sparse evidence on the parameter values indicates that ε β world may well the one we live in. At the same time, the ε ≫ β case appears surprisingly possible."

I think there are other existential risks beyond nuclear war, climate change or AI. For older civilizations, a prolonged dry season meant existential risk. It happened to the Mayans, long drought periods made food scarce and probably fresh water sources got polluted too, which lead to political instability and wars. A civilization rich enough to make astronomical measurements and develop a written language collapsed, and people went back to hunter-gatherers who could not read what their ancestors wrote.

Water scarcity was an existential risk to past civilizations. I think it is not an existential risk anymore because a region suffering a drought can get food from the other side of the globe and water in a ship. Improvements in water resources engineering and being open to trade and coexist with others have minimized the existential risk of a drought...all this while consumption has increased. This would suggest a Beta larger than Epsilon for the drought existential risk.

Perhaps every type existential risk has its unique Beta coefficient. Resources spent on minimizing nuclear war may have a great return (large Beta) while resources spent on trying to avert extreme climate change are wasted (very small Beta).

Of course, the world's largest religion is based on the coming apocalypse. Does that suggest humans are pessimists? Not necessarily. Last week Cowen referred to a new book in which the author makes the case for universal salvation - it's not a new case (several church fathers held that view) but it's rarely made today, most evangelicals believing the rest of us are (deservedly) doomed. But Leopold's point about existential risk discouraging the kind of investment/innovation that could avoid the risk is well taken. Today, many investors choose not to invest in productive capital because, it is assumed, they don't have confidence that there will be sufficient demand for the additional supply. So instead of productive capital, they choose so-called safe assets or speculate in rising asset prices (including in the latest nonsense in Silicon Valley). The mind of the investor class matters. Fortunately, not everyone follows the herd, but as wealth becomes increasingly concentrated in the inherited class, and preservation of capital the primary objective, will investment in productive capital dwindle.

To me, real estate is the "safe" asset of choice that should be a cause for concern. Look no further than the Kushners, the Trumps, or the latest arrivals from China or Russia. This is where they put their money. Our system "rewards" this in the form of rising housing prices in all of the Anglo-sphere but the natives feel priced out on something they are forced to either rent or buy. Housing is unfortunately both an asset and an essential. There's no way to build wealth if your top expense climbs faster than your pay. This is classic rent-seeking. Say what you want about SV but you seem to admire Thiel who has said that venture capital that goes to startups end up in the hands of landlords.

https://www.sfgate.com/expensive-san-francisco/article/peter-thiel-silicon-valley-capital-landlords-12759450.php

I have commented many times that we have chosen rising asset prices as the path to prosperity, not investment in productive capital. But I have defended the Fed, not because the Fed has facilitated the choice with its interest rate policy, but because the Fed has been faced with choosing between the lesser of two evils: rising asset prices or a collapsing economy. Our monetarist friends believe monetary policy can deliver us from our predicament. No it can't. It can only dig us deeper and cause an ever greater calamity.

"This suggests... having developed technologies advanced enough to threaten our permanent destruction, but not having grown wealthy enough yet to be willing to spend much on safety."
So, we round up all the billionaires & confiscate all the $$$. Probably only a small start in spending on safety but it might help until we get smart enough to learn how to co-exist......

I think we should boycott Apple.

This thesis seems empirically unsound. The Soviet Union had weapons sufficient to destroy civilization during a long period of stagnation and relative poverty, but never used them. I think the author is overlooking the fact that institutions are comprised of individuals, and the incentives of those individual political leaders in any political system will always be to maintain or increase their personal privileges. Doomsday is generally incompatible with that goal.

Conflating nuclear war and climate change? Really? Goodbye, dope.

It doesn't look like he conflated them.

The quote is: "catastrophes—such as nuclear war, extreme climate change, or powerful artificial intelligence run amok—that could imperil human civilization."

I assume the phrase "Extreme" climate change implies worse than any projections today. A 14C degree change would bring the world to around the hottest it's been sin the last half billion years. That would be a catastrophe. Though still probably not as severe as a 1985 full out nuclear war.

Heh, I was just about interested in the same very topic given you wrote the book but I don't know 5% of the math needed to model this thing. Although i'm 14 years older than him.

Oh, I feel you, man

Note that the paper is mostly replicating Jones (2016), so it's not very original.

Just because someone can LaTeX some equations from Acemoglu's textbook doesn't mean they can write an economics paper or that they should circulate some nonsense as one.

If the person can actually do economics, they will get a PhD or submit this paper somewhere. Then it's worth reading.

Comments for this post are closed