Randomness seems to confound us. For example, we have a tendency to infer non-randomness from apparent patterns in random events (witness the incorrigible optimists who spot trends in the spins of a roulette wheel or the ups and downs of the FT Share Index); at the same time, the history of statistics suggests that, when random samples are required, we often mistake the merely haphazard – or whatever happens to be near at hand – for the truly random. As I will show, the Doomsday Argument’s fundamental mistake is to rely on the intuitive but misguided notion that we can in general take ourselves to be typical humans, and thus, in effect, random samples of the species.
Here is the paper. If you are familiar with the core argument, scroll down to p.9 "We are necessarily alive…" for the beginning of the bottom line. I have never been persuaded by the Doomsday Argument, if only because it does not specify the appropriate reference class of self-observing agents. I now see further reason to be skeptical. Comments are open, and thanks again to the what-would-I-do-without-it www.politicaltheory.info for the pointer.
I’m a late-comer to this one — I’ve known of the doomsday argument, but never took the time to really think about it. (It seemed another of those annoying ideas that inadvertantly leak out of philosophy from time to time to bother the rest of us, and in such cases a little mental rock-kicking usually suffices).
With that proviso, I think the reference class argument is right–righter than you suggest. This, it seems to me, is fundamentally flawed:
“Finally, corresponding to finding you are in cubicle number 7 we have the fact that you find that your birth rank is about 60 billion”
But it is simply not a ‘fact’ that my birth rank is ‘about 60 billion’. That’s because there was no FIRST human–any line drawn in evolutionary history is completely arbitrary. Let us consider an ancestor from our evolutionary lineage that lived, say, 10 million years ago, or 20 if you like. By definition, no catastrophic event occured to wipe out his species–his direct descendants still live (and in enormous numbers). But his ‘species’ is extinct nonetheless. So is his situation consistent with ‘doom’ or ‘survival’? If we say, ‘doom’, then the exercise is pointless–our descendents will ultimately evolve into creatures distinct from ourselves, and ‘doom’ will come gradually and imperceptibly even without a catastrophy. But if we say ‘survival’ then we must include all the individuals in the populations of our evolutionary ancestors in our reference class. And then our reference population becomes all living things on earth, and the birth rank of a randomly selected individual is many, many orders of magnitude greater than 60 billion.
In all Bayesian reasoning, the first thing to consider is what the prior distribution is. There’s no reason to believe that the “lifetime” of the human species will be uniform over a particular range, for example. We could look at the distribution of the lifetime of species that have lasted at least as long as humanity, but that’s iffy as it would have lots of gaps in it. And then there’s the issue stated above — how can we tell when a species begins or ends?
But ignoring evolutionary reality, and thinking that there’s a distinct begin and end to a particular species, who’s to say that the lifetimes don’t follow an exponential distribution? The exponential distribution is rather cool as a time of failure distribution, as the expected time to failure is constant, no matter where you stand. Suppose a lightbulb follows this lifetime distribution, and it has an expected lifetime of 1000 hours. If it makes it to 1000 hours, and you look at what’s the expected time to failure, it’s still 1000 hours. Under an exponential distribution as the prior lifetime distribution, knowing how long the population has survived tells you absolutely nothing.
I’m hoping the person who originally made the doomsday argument isn’t a statistician or mathematician, because that’s embarrassing.
“Improbable” things happen all the time, and the perception of improbability depends on perspective. Suppose I hit a golf ball off the tee box. Presuming it lands in the fairway ( not an especially reasonable presumption, but that’s another matter) the odds that it will land on a particular blade of grass are millions to one. But it’s certain to land on one of them.
So is the “outcome” remarkable? Depends on the “outcome” you’re talking about. If someone had predicted which blade of grass the ball would land on, that’d be remarkable. But the after the fact remarking upon a multimillion to one shot? Not so much.
I think a similar thing is going on here… a forcing of perspective to make people go “wow.”
I don’t think this article adds anything to the existing literature. If you care about this, look at Nick Bostrom’s book on the topic. If you don’t care, just ignore it. It makes a prediction, but it doesn’t predict an observation that you can ever make.
The main practical significance of the argument is to remind us that we systematically underestimate the risk of events that have never occurred because we would not be there if they had occurred. We should expect zero such events in our past even if the events in question are likely.
Refuting this seems completely straightforward. There are two urns being filled with balls at equal rates. One will stop being filled with balls at 100 billion, the other will never stop being filled. Both are still being filled when I grab a ball an urn and sample the number.
No conclusion can be drawn whatsoever regarding which urn I drew from.
The publishing of the DA article in SciAm was a major factor in my decision to discontinue my subscribtion. So did the article claiming we would run out of oil in 50 years. (Shame they weren’t in the same issue.)
I’m fighting a cold now, so I don’t have the ability to write out all of my objections to the DA now.
My first objection (to both arguments) is that any prediction which is constant over time is likely to be worthless for many purposes. We’ve been running out of oil in 50 years for 100 years. The DA argument has said that 50% of “people” have already lived since the first “people” were around, and will continue to do so even (if and) when the last “person” is born. I don’t feel compelled to take such arguments seriously. The marginal utility to me of sitting down and detailing exactly where the argument fails is quite low, which is why I’ve not bothered to do so.
But if the first article is correct, and influential people are being convinced, then perhaps my marginal utility is being driven up.
There was a dart tournament. A latecomer managed to conive his way onto the list. His first dart was 2 feet to the left of center. His second, 3 to the right. His third was 1 to the left. The DA proponents declared him the winner on the basis that on average, he was spot on! (not real)
There was a lot of excitement at a voice recognition conference reported in the http://www.theinquirer.net a couple of years ago. Effective voice recognition was thought to be one year away. One of the participants explained the excitement, “For the last 15 years, we’ve been 2 years away!” (real–and the humorous irony was clear)
I our manager reported that the bridge chip for a processor we were testing was going to be later than expected (again)–9 months later. I replied that a year and a half ago, it was expected in 18 months, and that 6 months ago, it was expected in 12. I then confidently predicted that it would be available in 18 more months. He was not happy with my “joke”. I was right. (real)
Mark Nau nailed it.
>Refuting this seems completely straightforward.
>There are two urns being filled with balls at
>equal rates. One will stop being filled with
>balls at 100 billion, the other will never stop
>being filled. Both are still being filled when
>I grab a ball an urn and sample the number.
>No conclusion can be drawn whatsoever regarding
>which urn I drew from.
One flaw with this refutation is that assuming N can be infinite is unrealistic. If N is finite and you have two urns with different Ns, to be analogous to ‘being born’ in either population, you would need to observe a random ball n from all those dropped. The time at which you make the observation needs to be a random time in the range that makes sense for each urn.
So let’s say one urn is being filled with 100 billion balls, and the other with 100 trillion balls. One ball is dropped every second. You arrive at one of the two urns at a random time to observe the ball being dropped. Let’s say a computer chooses an urn at random, and also chooses at random which time you should observe the ball being dropped. Clearly, if you observe ball number 60 billion being dropped, you can make a good guess as to which of the 2 urns the computer picked.
google排å
wow power leveling
wow power leveling
wow powerleveling
wow powerleveling
wow power level
wow power level
hongkong hotel
hongkong hotel
beijing tour
beijing tour
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
北京翻译公å¸
北京翻译公å¸
é—¨ç¦
wow gold
wow gold
powerleveling
powerleveling
wow powerleveling
wow powerleveling
wow power leveling
wow power leveling
power leveling
power leveling
wow power level
wow power level
china tour
china tour
猎头
猎头
货架
货架
货架
google排å
google排å
wow gold
wow gold
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
北京翻译公å¸
北京翻译公å¸
仓库
仓库
仓库出租
仓库出租
物æµå›åŒº
物æµå›åŒº
监控
监控
æ¿€å…‰æ‰“æ ‡æœº
软件工程硕士
工程地毯
工程地毯
集团电è¯
集团电è¯
æ¿€å…‰æ‰“æ ‡æœº
æ¿€å…‰æ‰“æ ‡æœº
打包机
打包机
拓展è®ç»ƒ
拓展
拓展培è®
houyuping 07年7月11日
google排å
google排å
wow gold
wow gold
powerleveling
powerleveling
wow powerleveling
wow powerleveling
wow power leveling
wow power leveling
power leveling
power leveling
wow power level
wow power level
Ddo plat
Ddo plat
FFXI Gil
FFXI Gil
GW Gold
GW Gold
GW Power Leveling
GW Power Leveling
RS Gold
RS Gold
RS Power Leveling
RS Power Leveling
Lotro Gold
Lotro Gold
Lotro Power leveling
Lotro Power leveling
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
北京翻译公å¸
北京翻译公å¸
监控
货架
货架
货架
涂层测厚仪
涂层测厚仪
翻译公å¸
仓库
仓库
仓库出租
仓库出租
物æµå›åŒº
物æµå›åŒº
红外测温仪
红外测温仪
超声波测厚仪
超声波测厚仪
超声波探伤仪
超声波探伤仪
频闪仪
频闪仪
涂层测厚仪
涂层测厚仪
电ç«èŠ±æ£€æµ‹ä»ª
电ç«èŠ±æ£€æµ‹ä»ª
google排å
é—¨ç¦
é—¨ç¦
é—¨ç¦ç³»ç»Ÿ
é—¨ç¦ç³»ç»Ÿ
æ¬å®¶
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
光盘制作
光盘å°åˆ·
光盘å°åˆ·
æ¬å®¶å…¬å¸
æ¿€å…‰æ‰“æ ‡æœº
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
hongkong hotel
hongkong hotel
beijing tour
beijing tour
集团电è¯
集团电è¯
wow power leveling
wow power leveling
wow powerleveling
wow powerleveling
wow power level
wow power level
china tour
china tour
é’春痘
é’春痘
摆闸
痤疮
痤疮
色斑
色斑
黄è¤æ–‘
黄è¤æ–‘
é’春痘
é’春痘
å£è…”医生
å£è…”医生
å£è…”
å£è…”
å£è…”医院
å£è…”医院
å£è…”科
å£è…”科
三星集团电è¯æœº
呼å«ä¸å¿ƒ
电è¯äº¤æ¢æœº
交æ¢æœº
呼å«ä¸å¿ƒ
测厚仪
测温仪
åœè½¦åœº
é“é—¸
wulvyou 07年7月16日
google排å
google排å
wow gold
wow gold
powerleveling
powerleveling
wow powerleveling
wow powerleveling
wow power leveling
wow power leveling
world of warcraft powerleveling
world of warcraft powerleveling
wow power leveling
wow power leveling
cheap wow powerleveling
cheap wow powerleveling
powerleveling wow
hongkong hotel
hongkong hotel
beijing tour
beijing tour
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
北京翻译公å¸
北京翻译公å¸
é—¨ç¦
集团电è¯
集团电è¯
货架
货架
货架
拓展è®ç»ƒ
拓展
拓展培è®
google排å
æ¡ç 打å°æœº
机票
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
仓库
仓库
仓库出租
仓库出租
物æµå›åŒº
物æµå›åŒº
æ¡ç 打å°æœº
猎头
猎头
虚拟主机
红外测温仪
红外测温仪
超声波测厚仪
超声波测厚仪
超声波探伤仪
超声波探伤仪
频闪仪
频闪仪
涂层测厚仪
涂层测厚仪
电ç«èŠ±æ£€æµ‹ä»ª
电ç«èŠ±æ£€æµ‹ä»ª
beijing
beijing
china tour
china tour
great wall
beijing travel
beijing travel
æ¬å®¶
æ¬å®¶å…¬å¸
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
beijing
beijing
china tour
china tour
great wall
beijing travel
beijing travel
摆闸
痤疮
痤疮
色斑
色斑
黄è¤æ–‘
黄è¤æ–‘
é’春痘
é’春痘
三星集团电è¯æœº
呼å«ä¸å¿ƒ
电è¯äº¤æ¢æœº
交æ¢æœº
呼å«ä¸å¿ƒ
测厚仪
测温仪
åœè½¦åœº
é“é—¸
liying 07年7月18日
google排å
google排å
wow gold
wow gold
wow gold
wow gold
powerleveling
powerleveling
wow power leveling
wow power leveling
wow powerleveling
wow powerleveling
world of warcraft powerleveling
world of warcraft powerleveling
world of warcraft power leveling
world of warcraft power leveling
hongkong hotel
hongkong hotel
beijing tour
beijing tour
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
é—¨ç¦
北京翻译公å¸
北京翻译公å¸
google排å
超声波探伤仪
超声波探伤仪
物æµå›åŒº
物æµå›åŒº
æ¬å®¶å…¬å¸
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
æ¬å®¶
æ¬å®¶å…¬å¸
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
beijing
beijing
china tour
china tour
great wall
beijing travel
beijing travel
china tour
china tour
网站设计
多媒体
货架
货架
货架
光盘刻录
光盘刻录
光盘制作
光盘制作
光盘å°åˆ·
光盘å°åˆ·
摆闸
痤疮
痤疮
色斑
色斑
黄è¤æ–‘
黄è¤æ–‘
é’春痘
é’春痘
å£è…”医生
å£è…”医生
å£è…”
å£è…”
å£è…”医院
å£è…”医院
å£è…”科
å£è…”科
三星集团电è¯æœº
呼å«ä¸å¿ƒ
电è¯äº¤æ¢æœº
交æ¢æœº
呼å«ä¸å¿ƒ
测厚仪
测温仪
åœè½¦åœº
é“é—¸
wanyamin 07年7月26日
wow gold
wow gold
powerleveling
powerleveling
wow power leveling
wow power leveling
wow powerleveling
wow powerleveling
world of warcraft powerleveling
world of warcraft powerleveling
world of warcraft power leveling
world of warcraft power leveling
cheap wow powerleveling
cheap wow powerleveling
cheap wow power leveling
cheap wow power leveling
beijing tour
beijing tour
beijing
beijing
china tour
china tour
great wall
beijing travel
beijing travel
网站设计
网站设计
多媒体
光盘制作
超声波探伤仪
超声波探伤仪
æ¬å®¶å…¬å¸
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
google排å
wow gold
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
北京翻译公å¸
北京翻译公å¸
货架
货架
货架
仓库
仓库
仓库出租
仓库出租
物æµå›åŒº
物æµå›åŒº
痤疮
痤疮
色斑
色斑
黄è¤æ–‘
黄è¤æ–‘
é’春痘
é’春痘
电è¯äº¤æ¢æœº
交æ¢æœº
呼å«ä¸å¿ƒ
cheap Silkroad gold
cheap Silkroad gold
Silkroad gold
Silkroad gold
Silkroad Online gold
Silkroad Online gold
sro gold
sro gold
Archlord gold
Archlord gold
cheap Archlord gold
cheap Archlord gold
Archlord Online gold
Archlord Online gold
buy cheap Archlord gold
buy cheap Archlord gold
Dofus kamas
Dofus kamas
cheap Dofus kamas
cheap Dofus kamas
Dofus gold
Dofus gold
Dofus money
Dofus money
Rappelz Rupees
Rappelz Rupees
Rappelz gold
Rappelz gold
cheap Rappelz Rupees
cheap Rappelz Rupees
houyuping 07年8月15日
wow gold
wow gold
wow power leveling
wow power leveling
wow powerleveling
wow powerleveling
world of warcraft powerleveling
world of warcraft powerleveling
world of warcraft power leveling
world of warcraft power leveling
powerleveling
powerleveling
wow powerleveling
wow powerleveling
wow power leveling
wow power leveling
power leveling
power leveling
wow power level
wow power level
china tour
china tour
翻译公å¸
翻译公å¸
上海翻译公å¸
上海翻译公å¸
北京翻译公å¸
北京翻译公å¸
监控
货架
鼓风机
风机
风机
涂层测厚仪
涂层测厚仪
红外测温仪
红外测温仪
超声波测厚仪
超声波测厚仪
超声波探伤仪
超声波探伤仪
频闪仪
频闪仪
涂层测厚仪
涂层测厚仪
电ç«èŠ±æ£€æµ‹ä»ª
电ç«èŠ±æ£€æµ‹ä»ª
google排å
é—¨ç¦
é—¨ç¦
é—¨ç¦ç³»ç»Ÿ
é—¨ç¦ç³»ç»Ÿ
æ¬å®¶
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
光盘制作
光盘å°åˆ·
光盘å°åˆ·
æ¬å®¶å…¬å¸
æ¿€å…‰æ‰“æ ‡æœº
æ¬å®¶å…¬å¸
北京æ¬å®¶å…¬å¸
hongkong hotel
hongkong hotel
beijing tour
beijing tour
集团电è¯
集团电è¯
china tour
china tour
é’春痘
é’春痘
水虫スカルプ育毛シャンプー育毛岩尾シャンプー宮迫シャンプー発毛シャンプー抜ã‘毛シャンプー薬用シャンプー包茎包茎治療包茎手術冷ãˆæ€§ã‚€ãã¿èŸ¹æ¯›èŸ¹ã‚ºãƒ¯ã‚¤èŸ¹ã‚¿ãƒ©ãƒèŸ¹ç”£å¾Œã‚ããŒ
出会ã„豊島区出会ã„北区出会ã„è’å·åŒºå‡ºä¼šã„æ¿æ©‹åŒºå‡ºä¼šã„練馬区出会ã„足立区出会ã„葛飾区出会ã„江戸å·åŒºå‡ºä¼šã„神戸出会ã„久留米出会ã„水戸出会ã„åå¤å±‹å‡ºä¼šã„大牟田出会ã„ã•ã„ãŸã¾
ä½å‰å‡ºä¼šã„平野出会ã„å ºå‡ºä¼šã„è²å¡šå‡ºä¼šã„冨田林出会ã„箕é¢å‡ºä¼šã„æ±å¤§é˜ªå‡ºä¼šã„出会ã„系出会ã„無料出会ã„ç³»ã®æ”»ç•¥ã«ã¤ã„ã¦å‡ºä¼šã„カフェ出会ã„パーティー優良出会ã„系サイト
出会ã„æ±äº¬å‡ºä¼šã„é³¥å–出会ã„富山出会ã„å’ŒæŒå±±å‡ºä¼šã„山形出会ã„å±±å£å‡ºä¼šã„山梨在宅ワーク内è·åœ¨å®…アルãƒã‚¤ãƒˆä¹¾ç‡¥è‚Œãƒ€ã‚¤ã‚¨ãƒƒãƒˆã€€é£Ÿäº‹ã‚µãƒ—リメント無料å ã„
Comments on this entry are closed.