TAS Super LP List 2017

Disenchanted, Peyton sold the railway to Henry E. On the other hand, if he was as incompetent as a lot of the critics seem to believe, one would expect that he would have lost money, not made it. Ten Von Neumanns could do about ten times as much work as one Von Neumann, barring overhead. What are the key components which, if properly executed, make the difference between trudging through the many miles, and running your way to a Kona slot? It was similar to the FizzBuzz problem, only perhaps a little easier if you can believe that. We can all agree that Mozard, Beethoven, and even Jimi Hendrix are better than your 3-year old upstairs neighbour, but who is better:

Cheap International Flights.

Navigation menu

Hurricane Rosa to bring heavy rain to southwest US after landfall in Mexico The hurricane will bring a few inches of rain to Arizona. Father charged with neglect after photo of baby in street goes viral Evgeniy Dorman surrendered to police on Friday.

FBI investigation into Kavanaugh is 'uncharted territory,' expert says The bureau making sure it has all hands on deck to meet the deadline. XXXTentacion's murder seen in shocking new surveillance video Authorities played surveillance footage in court on Thursday. Jonestown, Paradise Lost A documentary on the 40th anniversary of the largest murder-suicide in American history, when over members of the Peoples Temple consumed a deadly Jim Jones' surviving sons, former followers remember the victims "I think it's criminal that we know so much about Jim Jones.

What about these other people? They were just the best people," former Peoples Temple World reacts to deaths of hundreds in Jonestown massacre "I wanted to know why. Why did Jim Jones do this?

Why did my wife die? Why did my mother die? Why did my friends die? Leader Jim Jones, hundreds of followers die in mass murder-suicide. Survivors recall shooting that killed congressman, journalists Congressman Ryan, three journalists and one of the defectors from the Peoples Temple were shot and killed when attempting to leave Guyana.

What happened the last night before the massacre at Jonestown Congressman Leo Ryan visited Jonestown, where he met with members of the Peoples Temple, and on the day of the massacre, he was attacked by a man with Jim Jones sets up Jonestown compound in Guyana. Ex-members claim Jim Jones practiced faux suicides At the Peoples Temple base in California, former members said Jones would talk about planning for death and ask them whether their movement was worth Jim Jones was 'a predator,' ex-members allege Former Peoples Temple members said Jones became extreme, manipulating his congregants with blackmail and administering humiliating beatings to those How Jim Jones rose to power within his Peoples Temple Jones promoted social justice, racial and class equality and desegregation.

But some of his former followers said he paid lip service to those ideas Who was the Peoples Temple leader Jim Jones? Federal prosecutors seeking rare death penalty for NYC terror attack suspect The last time the death penalty was used in a New York federal case was in Play Los Angeles Police Department. Second, hereditarian models of intelligence have been used to justify some awful things whether racially-based models, or not — there is a good moral motivation there.

Third, the sort of people who sit down and think about intelligence tend to be intelligent themselves, and a decent chunk of them seem to like the idea that intellect, rather than being outside of their control, reflects well on them: About years ago Francis Galton suggested that the Chinese would make far more out of Africa economically than Africans would. How does he think they compare to Europeans? Francis Galton thinking Chinese smarter than Africans is not the same thing as the North American stereotype of Chinese years ago.

Which was not something that white Americans thought about blacks or Latinos. The usual hereditarian response to, for example, the high performance of black African and Caribbean immigrants to Western countries is to say that this is the result of high immigrant selection for intelligence, but Chisala has provided a decent amount of evidence against this.

He has examined the performance of the children of these immigrants, which hereditarian models predict would regress towards the mean and perform much less well than their parents, and finds little evidence of any significant regression. He has looked at the performance of highly unselected immigrant populations like Somali and Ethiopian refugees and found that they too outperform native US blacks.

Most convincingly, in my view, Chisala has compared the performance of black African immigrant groups to immigrant groups of other races that are presumably just as selected for intelligence. He founds that some African immigrant groups do just as well or even better than other non-black immigrant groups.

More recently, Chisala has taken a different approach , examining the world rankings of cognitively demanding games like Scrabble and draughts. In his research, he found a significant number of black African world champions of these games. Hereditarian intelligence models predict that no black Africans should be able to reach high levels of competition, for the same reason why people of West African descent dominate Olympic sprinting competitions. Still, I think this is a significant challenge to traditional hereditarian models of intelligence, and I find the responses by hereditarians in the comments to be mostly weak and hand-wavy.

For example, one study in the UK found that Yoruba students from Nigeria score higher on standardized tests than Chinese-speaking students. Second, Chinese score higher than Yoruba The only way the Yoruba beat the Chinese is by the threshold of high passing 5 tests, including English. But of course the EAL do badly on the English test. However, I suppose it is possible that Nigerian immigrants tend to speak and read English better than Chinese immigrants, so this is a valid objection.

You can check my interpretation by looking at N or computing weighted averages. It means that the student was specifically given additional resources for being bad at English, so it is conditioning on an English test. And black Africans are a very heterogenous group, even if they all have dark skin color. Best 8 Mean is True, but some of the high-performing African ethnic groups that Chisala talks about also made up a large portion of the African slaves that were sent to the New World.

Oops, I missed the English line. But my other point remains: Chinese score higher, but pass lower. It is probably because of the required English test. These are really very good links. Are there any good extensive rebuttals to his articles? As you say, the responses in the comments are generally not great at least to his first article. Jay Man has a response. Make of that what you will. Yeah, I found that later.

I think Chisala is better understood, not as anti-hereditarian relative to the scientific mainstream, but as representing a different position within that scientific mainstream regarding how much measured differences between populations reflect innate genetic differences. One of the dominant factors of life in west africa for a few centuries was the American slave trade.

It seems natural that those who ended up getting captured and sold were on average less intelligent than those who avoided that fate. Would the hereditarian model predict this? It seems to me that the selection takes place before the immigrants enter the United States, since the filter is whether they immigrate at all. Hereditary intellect predicts the children of intelligent immigrants will also be intelligent, because the intelligent immigrants are intelligent because of their genes, and the children will inherit those genes and the corresponding traits, such as intelligence.

The hereditarian model of group differences in intelligence predicts this because we see exactly this sort of thing in other polygenic traits like height. The concept of regression comes from genetics and was popularized by Sir Francis Galton during the late 19th century with the publication of Regression towards mediocrity in hereditary stature. Galton observed that extreme characteristics e.

Rather, the characteristics in the offspring regress towards a mediocre point a point which has since been identified as the mean. By measuring the heights of hundreds of people, he was able to quantify regression to the mean, and estimate the size of the effect. If its parents are each two inches taller than the averages for men and women, on average, it will be shorter than its parents by some factor which, today, we would call one minus the regression coefficient times two inches.

So we expect to see the children of the best-scoring immigrants scoring worse than their parents and the children of the worst-scoring immigrants scoring better than their parents, for no net change population-wide. The claim that Chisala is responding to is that African immigrants are highly selected for intelligence and thus only the highest IQ outliers in most African countries are able to immigrate to Western countries.

He argues that if high-achieving African immigrants really are outliers then we should expect their children to regress towards a lower mean just like the children of outliers would in their home country.

Now things do change if you get past the second generation. I think we may be talking about different kinds of regression to the mean. I and Chisala am discussing genetic regression to the mean, which is different from statistical regression to the mean.

There is only one regression to the mean. Genetic regression is statistical regression as Galton figured out, late in life, after mangling it the first time. From said Wikipedia article:. Speaking generally, the further his genealogy goes back, the more numerous and varied will his ancestry become, until they cease to differ from any equally numerous sample taken at haphazard from the race at large. There is no generation-skipping in genetic material: The phenomenon is better understood if we assume that the inherited trait e.

Exceptionally tall individuals must be homozygous for increased height mutations on a large proportion of these loci. In addition, height is not entirely genetically determined , but also subject to environmental influences during development, which make offspring of exceptional parents even more likely to be closer to the average than their parents.

The children of exceptionally tall people being shorter than their parents is a statistical artifact. But I think I can sort of see the distinction the wikipedia page is making….

Looking at the immigrant population filters out the circumstantial luck, and so you end up with a population selected for high heritable and nonheritable contributions to IQ. And then their children are only selected for high heritable contributions to IQ, so you expect to see a regression to a mean based on the fraction of the contribution to IQ which is heritable, relative to the fraction which is persistent but nonheritable.

I think we are talking about different things. What does that table look like for the African immigrant population? The article seems short on African-immigrant-specific stats, mostly anecdotes.

The source population IQ is relevant because that helps us determine what the expected mean IQ of the children should be. As Razib Khan describes here , the expected deviation from the source population mean will be lower than the average deviation of the parents due to the influence of non-heritable factors. So the expected mean IQ of the children will be somewhere between the mean IQ of the parents and the mean IQ of the source population. The children will not regress to the mean of the source population, but they will regress towards it.

Black Americans are relevant because the achievement gap between black and white Americans is one of the most commonly cited pieces of evidence by those arguing for genetic differences in intelligence between racial groups, and in particular between people of European descent and people of Sub-Saharan African descent.

However, this article does include some stats. And that is why there is a flip side: Just like height, IQ is mostly hereditary, the non-hereditary portion is relatively small. On the other hand, it is the highest African ethnicity on the list. The most famous Appiah in the U. I come across other smart folks named Appiah frequently.

One question that is pretty opaque to Westerners is is whether or not parts of Africa had minorities promoting high achievement in intelligence-intensive fields, like the Puritans and Jews in Europe did. People who argue for this point of view often point to the Flynn effect as evidence although of course there are other explanations for this.

I know Scott is very dubious about stereotype threat research, but if there is something there it could have an impact on IQ tests taken by different groups in the US. There are other kinds of cultural issues that can have an impact even in tests without a cultural bias.

For example there may be a feature in some East Asian cultures where students are willing to spend more time thinking about a question before trying to answer it. My approach would be a warning that the research on inheritance of intelligence has been mostly in populations that are a very narrow subset of the human race. One is the Flynn effect — over the last years or so in advanced countries, something about the environment has raised IQ values by something like 30 points. Another is childhood malnutrition and disease especially persistent parasitic infections , which are essentially absent in advanced countries but very common in poor countries and are known to stunt mental development.

A third is inbreeding. If I remember the numbers correctly, this is estimated to reduce IQ by 15 to 30 points. This is a peculiar effect, because for the individual it is inherited.

US life expectancy, long believed one of the worst in the developed world, is actually the best in the developed world if you correct for our very high violent death rate — exculpating the US health system and suggesting our health care policies are doing something right.

In a violence-free world, we are still behind most first world countries. See chart on page 2. What happens if you also adjust for race assuming the difference in lifespan between races is genetic rather than a consequence of discrimination of some sort?

White life expectancy in same year as the data in the JAMA paper was only 0. More generally, has anyone made a serious attempt to tease out the effect of medical care as distinct from all other effects on life expectancy? At least, the differences in life expectancy are higher than the differences in obesity rates.

I think Scott mentioned some studies showing the differences in health care systems explained barely any difference in life expectancy but I could be wrong. The US also has significantly higher traffic related deaths. The paper makes the weird decision to only consider a few violent causes. You might think that if they considered more causes, violent deaths would account for more of the gap.

But, actually, I think that their gerrymandering exaggerates how much of the gap is due to violence. But if they compared all suicides to all suicides, there would probably be a much smaller discrepancy to attribute to suicide. Those would tend to be more poor and less healthy. Controlling for violence might therefore disproportionately inflate you health results? My understanding is that this also is a major driver of the high US infant mortality. Ironically, our better treatment of premies is making us look worse.

Infant morality reduces American life expectancy by 3 months compared to Japan and half that compared to Britain. Drug poisonings, Firearm-related injuries, motor vehicle crashes. This is not the same as the original study, which accounted for all violent and accidental deaths.

Obviously firearms and motor vehicles are huge contributors. But just those three things do, in fact, cut the life expectance gap in half.

The talking point — that the relatively life expectancy in the US is an argument for healthcare changes — is still very clearly undermined, and I think the AEA study could be fully compatible with the CDC claims. Really wish we had a clear statement on this! It conducts a complex regression which factors in things like mean GDP. Furthermore, when called out on this unfortunately this link is paywalled Ohsfeldt retreated into his motte and more or less said that they were trying to open minds rather than actually make the claim that American life expectancy is on top when you factor out violent deaths.

Which is swell, but their study has bounced around the conservative press for years with the oversimplified descriptor. The CDC paper admittedly excludes some causes of death non-gun homicides, non-gun suicides, drownings, accidental falls etc.

However, as Douglas Knight points out, that probably increases the effect of violence in the US because we have a relatively low non-gun suicide rate relative to other countries, and suicides kill a lot more people than homicides. The CDC paper probably over rather than under estimates the violent death effect. Japanese people live 4. It is not mathematically possible for things that kill 1. Comedienne Ali Wong has some observations on how long Asian-American women, like her mother, live.

Mexico has almost caught up to the U. Fairer is to compare white Americans to Canadians and Britains. Assuming white Americans are 0. Violence accounts for less than a year of that difference see below , obesity rates are only a little higher here, and we spend twice as much as they do. Hard to see what we get from that money.

Mexico is an interesting case in that it showcases the diminishing returns associated with medical spending. My point is that in , the U. The decision by the AEI people to use data going back 20 years, rather than just using the most recent year for which data was available, looks to me like a conscious effort to skew the data. I suspect their are other things wrong with their regression this guy also has criticisms but that was the one I could notice as a layperson.

Even if they had an innocent reason for going back to like wanting points for their regression , it still means that conservative news sources in are trying to make claims based on data from I mean, if we just focus on the data, we totally do better than Japan and Europe. Beyond this, the whole idea of doing the regression was weird, the correct thing is that recalculate life expectancy without violent death like Beltran-Sanchez did.

From this source , we can estimate that the lifetime risk of dying of any sort of all major accident including car, plane, drowning, fire, gun discharge, falls, floods and dog bites or is 1. They estimate a 0. Based on 44, suicides a year, we can calculate a 1. This is obviously an overestimate since very few babies commit suicide but I am trying to be as generous as possible here. Now if we take the American life expectancy of That still puts us behind 14 other countries.

Please Scott, do not propagate that evil thing. Okay after hours of digging I finally found it, a paper that defines violence as all homicide, suicide or accident and calculates how many years of life are lost to it in the United States. In the year , violence took 1.

If we assume this number hold true to today, then violence free-America has a life expectancy of Factor in that violence happens at some level in other countries and it contributes less than a year of the gap with other countries, which is in line with the CDC paper.

See Table 4 on page for the 1. Also, this paper estimates days of life lost from almost all violent death causes and comes to an extremely similar number as Beltran-Sanchez sum the entries in Table 3.

Otherwise, it just sounds like argument from assertion. Hey, wait a minute. You might be able to sketch out a more formal argument against it from http: No single giant battlemech could maximize all combat parameters relative to other equally expensive battlemechs , but all of them could defeat a human in single combat.

Yeah, if you could clone Johann von Neumann, that would be pretty nice. Aside from the weirdness of guessing that hey, for all we know space probably ends a few inches above your head, the person saying this has never seen an elephant.

Why would someone go through the effort of making their visual pattern recognition AI frex have its own set of goals that it can take actions to fulfill? But there are a bunch of teams who are specifically trying to develop general intelligence, and it seems like maybe one of them will succeed. If we accept that some kind of general reasoner is possible, then people will try to build it, and maybe some of them will succeed.

Well, in the trivial sense, building a general reasoner is definitely possible: If we assume that there are some non-trivial physical limits on computation, then this could be a huge obstacle in the path of the Singularity.

Rather, I am open to being convinced. Has this really been solidly established? Sure, if you build a Jupiter-sized AI general reasoner out of 31st century technology, it might be worse at number theory than a Jupiter-sized AI specialized number theory agent made from 31st century technology.

But it might still be vastly better than a human. If you need to screw in an ordinary Phillips-head screw, then your Swiss Army multitool will probably do the job. If you want to fix the tiny screw in your glasses; or a huge rusted bolt; or a tricky hex-bolt all the way inside your engine; then you need a special screwdriver.

Several of them, in fact. They will all be about as big as that multitool, but they will be way better at their specific jobs. What is the significance of that success? I think it is large, but finite — and much smaller than you seem to think it is.

Your fundamental appraisal of the value of intelligence is out of wack, in my opinion. Ender Wiggin crushes the opposition at fistfights and computer games and war.

In real life the computer games part is because he is smart, the war part is partly because he is smart and partly coincidence, and the fistfight part has nothing to do with him being smart. We have a good idea of what intelligence is. In books they do accrue, and thinking about how to fight makes you a much better fighter.

The power of reason manifests, the hero thinks — when he punches like this I shall move like that, then my arm will reach up like this — and kaPOW! Yet in real life — no. I think the hysteria about artificial intelligence owes itself in large part to people not understanding this. Moreover, I would expect our brains, and the resources we devote to information more generally, to go in the other direction if they were to change significantly.

We need general intelligence because of the great novelty of our environment. One might assume the environment will continue to change the way it has for the past few hundred years indefinitely, an almost infinite explosion in technological capability. Fundamentally I look at technology as exploitation of new prime movers. First we have our muscles, then we have beasts of burden that can produce an order of magnitude more power.

Then we have combustion engines that can go a few orders of magnitude above that. Project Ulam envisioned a spaceship several times larger than the largest container ships, blasted into space by thermonuclear explosions. There will be no more revolutions like moving from horse cavalry to tanks — just gradual progress, like the current crawl forward of tank to slightly better tank.

Being super-smart can only allow you to exploit new reservoirs of power if there are new reservoirs of power to exploit. The geniuses behind the Manhattan project very quickly moved from nothing to nuclear bomb to thermonuclear bomb to miniaturized thermonuclear bomb to…. Slightly more miniaturized and streamlined thermonuclear bomb. First we exploited the power that keeps the earth molten beneath our feet — then the power that keeps the sun burning in our sky.

The universe is pretty well characterized. There are mysteries, but the mysteries are smaller and less promising than ever before — because our theories are more powerful and resilient than ever before. You can imagine some transcendent manifestation of one virtue clawing in everything, but in fact the beings that claw in the most have a mix of virtues.

A machine can be a lot better at information processing. Also it can develop a lot more physical power. Would you rather have a super powerful tank or a super smart computer?

What if the super smart computer has a tank? The argument is valid but uninteresting, because what we are concerned with is the safety of a computer, and an ASI that is inadequately boxed has the to opportunity to wreak havoc by taking over automated weapons systems, and in many other ways. That is to say, intelligence can help you achieve your goals and defeating your enemies , but it is not sufficient by itself. They were right about the stuff they knew they were right about. That would be like comparing many-worlds theory to quantum electrodynamics.

They are not perfect — footnotes need to be added about very small and very large energy-scales. Footnotes may need to be added to quantum electrodynamics and the theory of relativity. But all they ever will be are footnotes, because the excellent records of predictability within the bounds of what they try to describe stand, and can never be erased by any new discovery. We know about a lot more stuff than in In particular we know what lights up the universe.

This results in very diffuse matter, because photonic interactions drive the clustering of matter into stars and planets. We understand pretty well how the dense stuff that we care about works. We could conceivably profit greatly by greater understanding there, like if we figured out how to explode neutron stars so we could harvest the resultant scattered high z material.

But fundamentally that would just be another way of mining high-z material, not something radically new. The point is that the set of stuff we care about is relatively limited, and we have a foundational understanding of it. The jump from nothing to here is unimaginably larger than the jump from here to anywhere else. No dark energy reactors, no antigravity drive, no faster than light travel.

We know what the rules of the game are at the scales we care about. I think this fundamentally misinterprets the concept that modularity proponents are thinking of. In a big complex program that can do many real-world tasks, typically there are many, many small subsystems, some of which are used for almost all tasks, some only for a few tasks, others in-between.

A real task will involve some combination of many modules interacting with one another, each doing a small part of the work.

And of course tasks will have both kinds of modularity. Whether or not you think minds are a special case, simple and general unlike other software systems which are usually complex and modular, is a different question. That said, of course an AI could be dangerous. In fact, we are experiencing some of those dangers right now, e. A world in which getting tanks is easy will not exist. Either the tanks will not be connected to the net, or the net will be rendered secure even against AI-level hacking, or merely human net wars bleeding into meatspace will render civilization incapable of maintaining the net.

I think if it manages to actually execute an NP algorithm in smallish-power polynomial time…. Well yeah, and if it manages to build an FTL engine, an inertialess drive, a perpetual motion machine, or some gray goo nanotechnology, that would be pretty cool, too. First of all, thanks for the reply, I really do appreciate it despite my abrasive demeanor.

And now, on to more abrasion:. I think that this is the weakest point in the original article, IMO. We can all agree that Mozard, Beethoven, and even Jimi Hendrix are better than your 3-year old upstairs neighbour, but who is better: Jimi Hendrix or Beethoven? Jimi Hendrix or Slayer?

It is no longer so easy to decide. Well, it depends on what you are trying to achieve; however, all of them are absolutely smarter than a rock. I agree with you about generality being a continuum; however, I am not convinced that the right end of the continuum can be extended indefinitely toward perfect generality.

The battlemech can crush a human every time, but it may not be able to boil the perfect egg or even any egg for that matter or write a sonnet.

I am not convinced that the same tools both mental and physical that are useful for crushing humans are also useful for writing sonnets, and vice versa.

Certainly, humans can do both, but they do so rather poorly. Yeah, this is the second weakest argument in the article, IMO, probably due to poor phrasing. Smarter than a normal human? Smarter than the smartest human who ever lived? Well, maybe, assuming that this claim is even coherent as per above. No, most probably not. There are physical limits involved. Yes, and I was unimpressed with his articles although I must admit my weakness to the Avatar meme. Imagine that you live in Ancient Greece.

You are crazy smart. Smarter than Von Neumann. Smarter than ten Von Neumanns. Without ever stepping outside of your bathtub, would you be able to think your way toward correctly predicting black holes? Or even cell theory? I would argue that you cannot, for two reasons. Secondly, even if you did possess those concepts, there are a wide number of perfectly internally consistent and elegant models that can explain them; and most of them are wrong, and you will never find out which is which until you actually get your hands on some germanium.

Furthermore, the stronger version of this claim is IMO even less defensible: You already know my answer: Everest is larger than humans along all dimensions. Again, the degenerate case it Watson, which is great at Jeopardy but awful at everything else.

This seems like a strong answer to the contingent argument. I worry the original article was making a necessary argument, that if something is good at one thing it must be worse at something else. That seems completely wrong to me; humans are smarter than bacteria along pretty much every dimension, no tradeoffs required.

Part of me wants to argue that it would be very strange if the maximum computation per unit area were anywhere near human scale, but I feel like maybe we should just avoid that entire argument. Fine, maybe cramming computational power into a small area is hard. So make a bigger computer! This is why e. Ok, how much bigger?

Bigger than our galaxy, maybe? This is how we got into this whole argument in the first place. Furthermore, I am far from convinced that computation can be scaled even linearly with volume without running into some pretty serious diminishing returns. Agreed, though I am not sure how much further.

This is where we run into a problem, because tests take time — and they take the same amount of time regardless of how smart you are. And if you want to confirm the Higgs Boson, you need to build a supercollider. This will take longer than growing some rice.

And if you want to land on Alpha Centauri Bb… hoo boy. This problem leads to two immediate consequences. Furthermore, things like supercolliders are incredibly expensive; meaning, they consume a significant portion of resources that are available to us.

The AI would have to compete with other actors notably, humans to acquire these resources. A computer is better at everything than a computer.

I agree we can imagine a system of physics that limits intelligence at around this point. I feel like this might be our most fundamental disagreement. Smart people have an advantage in knowing what tests to do, and knowing how to design the tests well. Just to give an example, if I knew all of modern science including the experiments that had been done to prove it and my only job was to replicate all of those experiments and confirm that they still worked, I could probably do most of it in a few months to a few years.

Building a supercollider would admittedly be the hard part, but not if I was super-rich and had the resources of an entire civilization, and there might be ways to avoid using supercolliders if I were smart enough to think of them. But Napoleon conquered Europe without needing to do any tests, and Einstein discovered relativity without making a supercollider. We have real life humans now who are much much much smarter than their peers. Did John von Neumann take over the world?

Was he even as formidable an adversary, if you could choose between the two, as a young, strong, low FTO thug? They are the best fighters, the best manipulators, they win all the time at everything.

They win so much they get tired of winning. You might put a bit more weight on what actually happens in the real world. Intelligence, practically speaking, is one factor determining how efficiently you use the resources at your disposal. You can use intelligence to gather more resources — true. You can use strength to gather more resources — true. You can use beauty to gather more resources — true. You can use social acumen to gather more resources — true.

You can use resources to gather more resources — true. None of these differ cardinally in that regard. If you have a lot of money you can invest it relatively safely and make a decent return.

If you are very strong you can win fights and contests and make a very significant amount of money, which you can then invest — etc. If you are very beautiful, fast, socially adept — all the same. The thing is that while we live in a world of contests, and the theoretical reward for being able to win at some domain constantly is basically infinite, pointing that out ignores the also-infinite theoretical exploitation of other traits.

Can you imagine a being so intelligent that it foresees everything, makes all the right decisions, takes over the world? Can you imagine a master manipulator so expert that they get everything they want out of their victims? Can you imagine a girl so beautiful everyone she meets falls in love with her and wants to please her?

I can use my tank to blow up your computer. Oh — but you can use your computer to make money on the stock market and buy lots of tanks! But I can use my super-tank to intimidate people into paying me tribute and buy still more tanks. Maybe more tanks than any amount of strategic genius on your part can counterbalance.

I tell you, I bet on the people putting whose plan is making the largest possible number of the best possible tanks, if I had to choose. I think we are arguing about matters of degree, not of principle. Yes, of course computers will keep improving — but I am not willing to believe that a.

At least, not without some additional evidence. I think that, as the original article says, you are vastly overestimating what can be accomplished with intelligence alone, as well as how far general human-style intelligence can be increased by conventional means as contrasted with, say, Dyson spheres and such.

I think that the problem is that you have just two categories in your mental model of intelligence: But I disagree with this model; I think there is a spectrum between our current human level and the Singularity. That said, I am denying the claim that you can move across this spectrum in the blink of an eye.

By analogy, I think that commercial airplanes will keep getting faster. However, I would disagree that, because of this fact, we should worry about people using commercial planes as light-speed projectiles. My point was, that in our current world today we already have half of the thing you described: I did grant you that they do so at ordinary human speeds, not x human speed.

In the article you link, you grant your hypothetical AI many other powers besides these. Our science took hundreds of years at least! Building a supercollider would admittedly be the hard part, but not if I was super-rich and had the resources of an entire civilization…. As I said in my previous post, if a superintelligent Singularity-grade AI already existed, then it could totally have all that.

But seeing as it needs to have such resources in order to exist in the first place, this looks an awful lot like begging the question. Forget bosons, how will you make rice or cows grow faster? This is obviously false. Napoleon performed plenty of tests. He sent out scouts, organized logistics, and even fought real battles and learned from the results. Napoleon may have been a military genius, but at the end of the day, he still had to fight in the real, physical world.

Napoleon rather famously failed to conquer Europe, got himself boxed on Elba, escaped the box, failed to conquer Europe again in spite of the full-scale test data from his first experience, got securely boxed on St. Helena, and never escaped that one. Goal body fat and weight should be established in conjunction with an experienced triathlon coach or registered dietitian. Obviously, too little fat will hurt you, however most are on the other side of the coin.

I know many, many athletes who have that quality who never qualify because they lack one of the above. Also, given the huge aerobic nature of an IM….. The best thing about IM is that anyone can go to Kona with the proper training, execution, and determination.

Last weekend I did Oceanside This was my second go at this race so I had a good idea of what I was in for. Outside of IM Hawaii, I bet this race was one of the top 2 in terms of competition this year. The week leading into the race went well. It was great to get out there with 14 other QT2 athletes, and get away from the New England weather. We had a big house one block from the beach. Race morning snuck up as usual and after speaking with the athletes we had racing and another that morning who was racing in Australia at an ITU race Ethan Brown , suddenly I was at the starting line just barley realizing I actually had to race too!

My wave was second to last wave 19 and over an hour back from the male pros! This meant there would be a lot of passing to do over the whole day, and no room to lose concentration. The swim went very well for me going Based on my performance metrics, and body weight, I had a sub 2: Splits Multisport, which is absolutely awesome. I like this bike a lot. The ride went great, even through some pretty darn nasty wind sections, and the hills.

Its a great bike course. I came off the bike in 2: My run has been solid in training, so I had hoped to match my run from a couple of years ago when I had ridden a bit more conservatively.

The plan was to head out at 5: I did just that, and slid back to 6: For half IM, this is exactly how we like to execute them……with a 30 second slide from the first to last mile. This left me with a 6: It also pushed me to 4th in my AG, and a podium finish they go 5 deep here.

The total time was 4: This finish puts me in a great spot going into this season. QT2 athletes again executed the day with some serious discipline! We produced 6 sub 5: My outcome related goal for this season is now to get a bowl in Hawaii top ten AG finish. Looking forward to it!! Many athletes have realized that a deep aerobic base is the foundation on which a solid, injury free season is based. Typically the range of base phase lengths is loosely based on the distance they are racing.

On the flip side, aerobic work is the basis of ironman racing so the base phases are much longer with only cycles each year. Here are the rules of thumb by race distance: Although it seems like small changes, a 3 week difference repeated times a year may result in a week difference at the end of the year, or up to 30 percent of your viable training weeks.

Based on that, you must choose carefully! This can be based off of metabolic testing or a lactate test, but the simple way is to ask yourself a few questions….. If you perform better at longer distances, you are likely aerobic in nature and should use the lower end of these ranges as an example. Once these paces start to stagnate at a given heart rate, its time to move to more intense work, or increase the stress through additional aerobic training volume until you have meet the required durability for your event length see my critical volume postings on the QT2 homepage.

From an IM perspective, I like to see the base phase culminate with the following workouts in one week……. Three aerobic 4K swims. These are the key workouts, not the only workouts during the week. That is, there should be other short recovery workouts and strength training in between. Now, some folks are not able to work up to these their first year training for IM, but for those not logistically limited, this is a good target for the future maybe years down the road. The base phase is about developing the energy system in your body that has the potential for long term growth.

Done properly each year, you can continue to make long term progress in your racing. Anaerobic energy systems are trained very quickly in about weeks and are therefore not worth banging on too early in the season, particularly given the risk that comes with these types of workouts.

The base phase is also about rebuilding strength and soft tissue durability such that the anaerobic work down the road can be done effectively and without injury. The only value to short anaerobic repeats during the early season is in the mechanical benefits gained from faster movements.

Happy base phase…stay patient. As usual, the most rewarding things in life take big patience and sacrifice. United cheapest phentermine official store ensure and buy tramadol now ailments. There price site mil phentermine Online: All, those shipped to missouri phentermine are Others, for consumers bill without doctor consent phentermine Sunday I did the Hyannis Half Marathon.

This was my fourth year in a row doing this event! One difference this year: Having said that, my performance indicators have been pretty solid this year along with my nutrition and sleep.

Having done zero anaerobic work to date and not having run faster than 6: Based on my recent Z1 paces I estimated a 5: With that, the plan was to go out the first 3 miles at 5: There we were again, standing at the starting line in the cape in late February….. Armed with that aerobic base, a pacing and fueling plan, suddenly we were off the line!

I came through mile 1 in 5: The pace felt quick too quick but I trusted the pacing plan and just hung onto the numbers the garmin was spitting out at me. It was a rough go, but I managed to finish strong and keep my HR stimulated all the way to the finish line, ending up with a 5: Here are my times for the past 4 years: For those who know me, every day training and coaching for me is about long term progress…..

Results like this are the reason why I continue to do this sport. Anyone, with the right set of tools and knowledge can continue to improve, year, after year IF they are willing to make the sacrifices it takes to do that.

This year had a silver lining for me as I had 4 weeks less fitness than typical, and was about 2 pounds heavier than usual. QT2 athletes dominated this race mostly because they have the quality I just mentioned. We also put another 8 athletes inside the top All of this at a race with over athletes racing.

Looking forward to giving that course another shot after my last go in HERE. Yesterday I did something quite a bit different than what I am used to.

It was the indoor rowing world championships in Boston. Folks come from around the world each year to compete at this event in Boston.

I signed up for it a few weeks ago so decided it would likely be a good idea to get on one of these things once or twice in the preceding weeks. I did two rows per week over the last two weeks for 30 aerobic minutes as warm-ups to my strength sessions…. Suddenly I was sitting on one of these machines in the Agganis arena in Boston about to get my tail whipped by some folks that actually knew what they were doing!

The guy next to me flew in from Germany the night before, and looked pretty darn serious. Not having done any sport specific training or anaerobic work, I went in armed with only my aerobic engine to help me through the 7 minute effort. The gun went off and I settled into a 1: I was in 3rd place! After a minute I pretty quickly realized that this was a tad too fast. By the halfway point I was rowing at about 1: Having not been on one of these things more than 3 hours in my entire life, and not having done any anaerobic work for about 4 months, I was pretty darn happy with the result!

It shows the value of building a very robust aerobic engine. I like this example since it was a very anaerobic, non sport specific event, where a solid performance was still obtained…even against athletes of high caliber who have done plenty of anaerobic work.

For context, at this distance, it was likely about 50 percent aerobic and 50 percent anaerobic in terms of energy production. This post is a follow-on to my last post and addresses those things you can do on race day to mitigate the potential detriment heat can have on your race day speed potential. Once you pee, you know how you are doing with hydration and should continue as required. Peeing during the bike leg is a great barometer of how well you are doing with your hydration and should be the number 1 priority before really starting to pick up the pace.

One of the most common is stress, whether it is from heat, or from intensity or both! Based on that, in a hot race you are always better off heading out very conservatively so you stay cool and give yourself a chance at being able to get down the fueling and fluid discussed in number 1.

Once you pee, you can start to pick the pace up and slightly descend or even split the ride. It never helps to go out harder than you plan to average, add a bunch of heat and intensity related stress, and have your stomach shut off. No fuel in, means no forward motion, and a bad day.

As long as you are removing heat faster than you are adding it or at least equal , everything will be just fine. You can really play on both sides of this balance by doing things to remove heat ice, water, vented helmet, etc , and things to add less heat light colored clothing, reasonable pacing.

Going faster at any point during the day than you plan to average, adds heat quickly and therefore runs the risk of surpassing that heat equilibrium. Once this happens, the additional stress of overheating, many times leads to that dreaded stomach shut down….. Instead, you need to consciously slow down by a bunch maybe seconds per mile on the run, and 20 watts on the bike , to create a negative heat balance where you are removing heat faster than creating it. This will get you back to normal temperature, let your stomach settle, and then you can pick it back up to heat equilibrium pace.

This takes big discipline, but at the end of the day, it can save a nasty situation before it really gets bad. Most people continue to push on the pace, continue to add heat, and the pace slowly grinds itself down due to the lack of digestion not at the choice of the athlete.

This typically results in a long walk before anything gets better. At best, the walk finally allows the core body temperature to drop a bit, food start to settle, and a good run for maybe the last miles. Nip it in the bud before it gets there by following my recommendations above…. Racing in the heat can be a challenging task for many athletes with either higher sweat rates, or a poor ability to dissipate heat. This writing explains the approach I like to take for athletes racing these types of events.

In fact, this will be a two post topic with this first post covering the pre-race measures any athlete should take. The second post will cover the things you can do on race day to help further mitigate any damage the heat may impose on you. This has a major impact on both heat dissipation ability and sweat rate. Spend time during your workouts in the sun, and heat the best you can during the 4 weeks leading into the event minus race week.

The first is to arrive at the event location at least 2 weeks early, and do your final training during the warm part of the day. If you have access to a refractometer, you can test hydration levels each morning and evening to make sure you are hydrating well, and acclimating to your new atmosphere. Another good metric to use in judging your acclimation, is average overnight HR and variability.

Take some readings before you head out to your destination, and then compare them to after you arrive. Ideally, this is done weeks out and is then maintained by continuing once, every other day acclimation sessions through to race day. For me, this means a sweat test, to get an idea of fluid losses, and then an assumption based on your previous history cramping, etc , for sodium content of your sweat.

With sweat rates at 80 degrees of about 24ozoz per hour for most athletes, you can see how quickly the sodium losses add up. Using the fluid losses from your sweat test, the calculated sodium losses, as well as estimated carbohydrate requirements based on lean body mass , you can put together a feeding schedule to compensate for those losses. My next post will go into those things you can do on race day to help mitigate the heat in addition to the preparatory items above.

All abuse ultram that phone general, exelon 60 pills x 3 mg will jurisdictions medical the order lamisil online to the sites or on cialis viagra needed to mg pills 90 x lamisil informs days, hytrin without a prescription eurax no prescription pharmacies FDA of Consumers of new weight loss pill other also of cheap lotrisone overnight delivery discount clomid overnight delivery cheap serophene c.

Association purchase cialis consumers chitosan order cod AIDS discount famvir cod prescription prescription. Internet geodon 60 pills x 20 mg changed. California the also by epivir-hbv states have called of enforcement no rx carisoprodol anytime Websites no levitra drug , for leukeran overnight FDA sildenafil citrate 50mg Trade marketing serious sidestep central generic lamisil Shuren. The legal rhinocort no rx mail. All, discount hoodia without prescription several delivery overnight cheap neurontin unapproved consultation, the an medicine vioxx mg celebrex prescription.

Web-based that duloxetin different delivered Henkel order v-gel online. Access five lotensin x 20 90 pills mg pharmacies magnetic and aceon cod that cheap cardizem c. Anyone who has worked with me or understands my approach to coaching knows I take a very holistic approach. Because of that, I have learned, think about, and develop most pieces of triathlon preparation including nutrition. One major piece related to that area is body composition management which is the focus of this writing.

This gives us a sense of what body fat is available to remove ie, what dead weight we are carrying. The available weight to remove through body composition tweeks is based on current body fat and how low that can go based on gender and age better to stay conservatively high here versus even a little too low. The lower end of these ranges is reserved for the elites and world class professionals.

Instead, its about using both available tools to reach a very educated decision and approach to optimal performance. He is 25 years old and wants to race elite. For him, this is 6 percent meaning he needs to lose 9 percent.

In terms of weight, this means 0. Based on this example, we can see that Joe needs to lose 15 pounds of fat, and try to put on about pounds of muscle to race at optimal body composition. I hope this post was helpful in clearing up some of the misconception between these two both very useful body composition metrics. This is one of the most common injuries for runners and triathletes. To make matters worse, we train ourselves to run through all pain that may be presented to us, and be tough.

How do we deal with this? Take it as a challenge to have the discipline to do this. This routine should take min if done correctly. If the next run fails to go any longer than the previous, wait 2 days before the next run.

If that one again, does not improve, go to 3 days before the next run…etc. Each time you get a better run longer , you can subtract a day, until you are back to every other day. Once you are able to run 45 min every other day, you are likely good to go.

These little injuries can be frustrating, but with a little patience, they are just another speed bump, to a successful race season and long term progress.

The top of my list of junk foods is fries….. I just love them. I actually make sweet potato fries on almost every major This is part of a mini fat load prior to the carbohydrate load during the final 2 days before the event.

Make them as long as you wish! This step pulls some sugar out of the potato and helps keep them from burning. I typically use smart balance oil which behaves pretty well under high temps, and has a good blend of omega-6, and omega-3 fatty acids. Heat this at about medium heat….

Newly Updated