6 ms·

Occam’s Razor suggest he was exfiltrated.

TheBlight 2y ago
Seems like the most likely explanation to me. If he were arrested for being a spy, presumably the DOJ would want to showcase it.
- runarberg 2y ago
  Think like a Bayesian. How many spies have been caught lately? of them, how many did the DOJ showcase? Wikipedia has a list of imprisoned spies in the USA. It is pretty sparse, the last spy to be convicted in the USA was in 2016. https://en.wikipedia.org/wiki/List_of_imprisoned_spies https://en.wikipedia.org/wiki/List_of_imprisoned_spies Searching the DOJ homepage with the search term “spy” (to see which they have showcased) gives us 209 results, including a bunch of news around the 2009 conviction. The most resent news I found was from 2022 (https://www.justice.gov/archives/opa/pr/five-individuals-charged-variously-stalking-harassing-and-spying-us-residents-behalf-prc-0 https://www.justice.gov/archives/opa/pr/five-individuals-cha...) a very vague news which is actually more about harassing private citizen (presumably piggy-banking out of the anti-asian hate following Covid). The most resent news about actual spying against the US government was from 2020 (https://www.justice.gov/archives/opa/pr/former-cia-officer-arrested-and-charged-espionage https://www.justice.gov/archives/opa/pr/former-cia-officer-a...) Alexander Yuk Ching Ma. He was actually found guilty in may 2024 (so he is missing from that Wikipedia list). The DOJ article about the conviction doesn’t use the word spy, so it is missing from my search, but here it is: https://www.justice.gov/archives/opa/pr/former-cia-officer-sentenced-10-years-prison-conspiracy-commit-espionage https://www.justice.gov/archives/opa/pr/former-cia-officer-s.... Given the above, I think it is safe to bet that if he was arrested for being a spy, the DOJ would indeed want to showecase it. However, spies are very rare, and the likelyhood of OP’s case being a spy is very low. It is much more likely that he disappeared for other reasons.
  - Spooky23 2y ago
    A guy was arrested for being a PRC spy, stealing sensitive documents from GE Vernona. 2-3 years ago. I remember it because he was using stenography to embed documents in landscape photos sent via email. They also use various means to charge people. Lying to a federal agent is a serious crime. Almost any misuse of federal funds can a a serious crime.
    - fsiefken 2y ago
      how did they catch that?
      - Spooky23 2y ago
        IIRC, he was sending pictures of the sunset on his jog every day, and the pictures were unusually large. https://www.theregister.com/AMP/2023/01/04/ge_turbine_china_prison/ https://www.theregister.com/AMP/2023/01/04/ge_turbine_china_...
  - zmgsabst 2y ago
    There’s 3 soldiers arrested this year. https://www.youtube.com/watch?v=3LOP645eW-M https://www.youtube.com/watch?v=3LOP645eW-M
- atian 2y ago
  Sounds like entrapment or he was heavily induced (by whatever) into whatever they got him for. Maybe he'll get pardoned like Ulbricht.
runarberg 2y ago
I know this is HN and we love these simplistic universal laws here (Hanlon's razor being another beloved law) but Occam’s razor isn’t really applicable here, or at the very least, it isn’t very helpful, and it certainly isn’t conclusive. That he was exfiltrated by the Chinese government also requires the same number of variables as that he was disappeared by the American government. So Occam’s Razor actually suggests nothing here. Occam’s razor is a rule of thumb which suggests that if several hypotheses explain the same thing, the simplest one should be favored. When we have a missing person with several unknown variables, until we see some evidence, any guess is as good as other. And no hypotheses should be ruled out. On HN we also like Bayesian analysis, so instead of looking for the simplest explenation, we should be asking what we already know. What is the probability that this person was disappeared by the government, given that 3 other students have been disappeared by the government lately? Given that these 3 other students were also people of color? Etc.
- echoangle 2y ago
  > When we have a missing person with several unknown variables, until we see some evidence, any guess is as good as other. Really? So that he was abducted by aliens is also on the table and has the same probability? Obviously there can be some prior information that makes „he was exfiltrated by Chinese intelligence services“ more likely than „he was disappeared by US intelligence services“. I’m not saying there is, but philosophically, you don’t need to know anything more than „A person with relations to china disappeared“ to be able to determine one hypothesis as the most likely one.
  - runarberg 2y ago
    I answered the exfiltration likelihood in a sibling thread. The summary is that spies are very rare and it is not a likely explanation. Keeping the Bayesian hat on we can pretty much rule out aliens. We have never witnessed an alien abduction, so the chances of this being an alien abduction is actually infinitely smaller than exfiltration. We can set the same prior probability to all the possibilities, that is fine. But given the evidence the posterior for alien abduction is very close to 0, if not just simply 0. Given the recent pattern of behavior from the US government, we have seen quite a few people disappeared. The posterior for US government involvement is IMO much higher than exfiltration. That said, no government involvement is higher still (and is actually the hypothesis favored by Occam’s Razor).
trod1234 2y ago
There are no published facts aside from this guys missing, so there can be no adequate accounting of the facts. You've misused Occam's Razor.
- soulofmischief 2y ago
  Yeah, good example of the maximum entropy principle, [0] in which "the selected distribution is the one that makes the least claim to being informed beyond the stated prior data, that is to say the one that admits the most ignorance beyond the stated prior data", a.k.a. the most entropy. Occam's Razor must still be constrained to known facts, and you cannot construct a "simplest answer" out of data you simply do not have. [0] https://en.wikipedia.org/wiki/Principle_of_maximum_entropy https://en.wikipedia.org/wiki/Principle_of_maximum_entropy
- CodeWriter23 2y ago
  There's plenty of other public information. The FBI was searching his house for undisclosed reasons, so most definitely they have some kind of investigation open on him. And his degrees were issued by universities in Nanjing and Shanghai, China, strongly suggesting he is a Chinese National, perhaps naturalized to the United States. That was known at the time of my post. We now also know now at the time of his escape, Indiana University had terminated his Professorship. And has continued to disavow he and his wife. I did not misuse Occam's razor, my knowledge of public facts simply exceeded yours.
  - trod1234 1y ago
    You have a vivid imagination. The facts do not necessarily support what you claimed. Occam's razor requires an adequate accounting of the facts, not speculation. All you have mentioned is speculation, and the facts you have mentioned don't support this narrative, neither when you first mentioned it, nor now. You misused it, and trying to falsely justify that you didn't is deluding yourself.
snickerbockers 2y ago
Wouldn't "murder victim" be the Occam's razor explanation? We know his wife is still living in the house. We also know that statistically speaking your spouse is the person most likely to murder you. I'm not going to say anything more than that because the obvious next statement would be one that is based entirely on speculation, but it takes a lot less speculation than all these ridiculous James Bond fanfics I'm seeing.
- sweetjuly 2y ago
  > We also know that statistically speaking your spouse is the person most likely to murder you. This is slightly misleading. While it is true that the person with the highest probability of murdering you is your partner (at least, if you're a woman), it is more likely that the murder is a stranger to the victim than to be the spouse of the victim. For example, in the FBI's 2011 dataset [1], of the 7076 murders by someone with a known relationship to the victim, 1295 were by a husband, wife, boyfriend, or girlfriend. This is compared to, of course, the 2700 murders committed by acquaintances and 1481 by strangers. From this, we can see ~18% of murders are by their partners and ~60% are by someone quite a bit further flung. The reason why both statistics are simultaneously true is that most people have only a few partners (focusing the entire risk on a small group) where as there are many thousands of people in the latter categories (creating a very diffuse risk). So, if you were to bet one which person murdered someone, you would do well to guess one of their partners. If you had to guess if the victim knew their murderer, you should bet that they did not. [1] https://ucr.fbi.gov/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/tables/expanded-homicide-data-table-10 https://ucr.fbi.gov/crime-in-the-u.s/2011/crime-in-the-u.s.-...

Seems like the most likely explanation to me. If he were arrested for being a spy, presumably the DOJ would want to showcase it.

runarberg 2y ago
Think like a Bayesian. How many spies have been caught lately? of them, how many did the DOJ showcase? Wikipedia has a list of imprisoned spies in the USA. It is pretty sparse, the last spy to be convicted in the USA was in 2016. https://en.wikipedia.org/wiki/List_of_imprisoned_spies https://en.wikipedia.org/wiki/List_of_imprisoned_spies Searching the DOJ homepage with the search term “spy” (to see which they have showcased) gives us 209 results, including a bunch of news around the 2009 conviction. The most resent news I found was from 2022 (https://www.justice.gov/archives/opa/pr/five-individuals-charged-variously-stalking-harassing-and-spying-us-residents-behalf-prc-0 https://www.justice.gov/archives/opa/pr/five-individuals-cha...) a very vague news which is actually more about harassing private citizen (presumably piggy-banking out of the anti-asian hate following Covid). The most resent news about actual spying against the US government was from 2020 (https://www.justice.gov/archives/opa/pr/former-cia-officer-arrested-and-charged-espionage https://www.justice.gov/archives/opa/pr/former-cia-officer-a...) Alexander Yuk Ching Ma. He was actually found guilty in may 2024 (so he is missing from that Wikipedia list). The DOJ article about the conviction doesn’t use the word spy, so it is missing from my search, but here it is: https://www.justice.gov/archives/opa/pr/former-cia-officer-sentenced-10-years-prison-conspiracy-commit-espionage https://www.justice.gov/archives/opa/pr/former-cia-officer-s.... Given the above, I think it is safe to bet that if he was arrested for being a spy, the DOJ would indeed want to showecase it. However, spies are very rare, and the likelyhood of OP’s case being a spy is very low. It is much more likely that he disappeared for other reasons.
- Spooky23 2y ago
  A guy was arrested for being a PRC spy, stealing sensitive documents from GE Vernona. 2-3 years ago. I remember it because he was using stenography to embed documents in landscape photos sent via email. They also use various means to charge people. Lying to a federal agent is a serious crime. Almost any misuse of federal funds can a a serious crime.
  - fsiefken 2y ago
    how did they catch that?
    - Spooky23 2y ago
      IIRC, he was sending pictures of the sunset on his jog every day, and the pictures were unusually large. https://www.theregister.com/AMP/2023/01/04/ge_turbine_china_prison/ https://www.theregister.com/AMP/2023/01/04/ge_turbine_china_...
- zmgsabst 2y ago
  There’s 3 soldiers arrested this year. https://www.youtube.com/watch?v=3LOP645eW-M https://www.youtube.com/watch?v=3LOP645eW-M
atian 2y ago
Sounds like entrapment or he was heavily induced (by whatever) into whatever they got him for. Maybe he'll get pardoned like Ulbricht.

runarberg 2y ago

I know this is HN and we love these simplistic universal laws here (Hanlon's razor being another beloved law) but Occam’s razor isn’t really applicable here, or at the very least, it isn’t very helpful, and it certainly isn’t conclusive. That he was exfiltrated by the Chinese government also requires the same number of variables as that he was disappeared by the American government. So Occam’s Razor actually suggests nothing here. Occam’s razor is a rule of thumb which suggests that if several hypotheses explain the same thing, the simplest one should be favored. When we have a missing person with several unknown variables, until we see some evidence, any guess is as good as other. And no hypotheses should be ruled out. On HN we also like Bayesian analysis, so instead of looking for the simplest explenation, we should be asking what we already know. What is the probability that this person was disappeared by the government, given that 3 other students have been disappeared by the government lately? Given that these 3 other students were also people of color? Etc.

echoangle 2y ago
> When we have a missing person with several unknown variables, until we see some evidence, any guess is as good as other. Really? So that he was abducted by aliens is also on the table and has the same probability? Obviously there can be some prior information that makes „he was exfiltrated by Chinese intelligence services“ more likely than „he was disappeared by US intelligence services“. I’m not saying there is, but philosophically, you don’t need to know anything more than „A person with relations to china disappeared“ to be able to determine one hypothesis as the most likely one.
- runarberg 2y ago
  I answered the exfiltration likelihood in a sibling thread. The summary is that spies are very rare and it is not a likely explanation. Keeping the Bayesian hat on we can pretty much rule out aliens. We have never witnessed an alien abduction, so the chances of this being an alien abduction is actually infinitely smaller than exfiltration. We can set the same prior probability to all the possibilities, that is fine. But given the evidence the posterior for alien abduction is very close to 0, if not just simply 0. Given the recent pattern of behavior from the US government, we have seen quite a few people disappeared. The posterior for US government involvement is IMO much higher than exfiltration. That said, no government involvement is higher still (and is actually the hypothesis favored by Occam’s Razor).

trod1234 2y ago

There are no published facts aside from this guys missing, so there can be no adequate accounting of the facts. You've misused Occam's Razor.

soulofmischief 2y ago
Yeah, good example of the maximum entropy principle, [0] in which "the selected distribution is the one that makes the least claim to being informed beyond the stated prior data, that is to say the one that admits the most ignorance beyond the stated prior data", a.k.a. the most entropy. Occam's Razor must still be constrained to known facts, and you cannot construct a "simplest answer" out of data you simply do not have. [0] https://en.wikipedia.org/wiki/Principle_of_maximum_entropy https://en.wikipedia.org/wiki/Principle_of_maximum_entropy
CodeWriter23 2y ago
There's plenty of other public information. The FBI was searching his house for undisclosed reasons, so most definitely they have some kind of investigation open on him. And his degrees were issued by universities in Nanjing and Shanghai, China, strongly suggesting he is a Chinese National, perhaps naturalized to the United States. That was known at the time of my post. We now also know now at the time of his escape, Indiana University had terminated his Professorship. And has continued to disavow he and his wife. I did not misuse Occam's razor, my knowledge of public facts simply exceeded yours.
- trod1234 1y ago
  You have a vivid imagination. The facts do not necessarily support what you claimed. Occam's razor requires an adequate accounting of the facts, not speculation. All you have mentioned is speculation, and the facts you have mentioned don't support this narrative, neither when you first mentioned it, nor now. You misused it, and trying to falsely justify that you didn't is deluding yourself.

snickerbockers 2y ago

Wouldn't "murder victim" be the Occam's razor explanation? We know his wife is still living in the house. We also know that statistically speaking your spouse is the person most likely to murder you. I'm not going to say anything more than that because the obvious next statement would be one that is based entirely on speculation, but it takes a lot less speculation than all these ridiculous James Bond fanfics I'm seeing.

sweetjuly 2y ago
> We also know that statistically speaking your spouse is the person most likely to murder you. This is slightly misleading. While it is true that the person with the highest probability of murdering you is your partner (at least, if you're a woman), it is more likely that the murder is a stranger to the victim than to be the spouse of the victim. For example, in the FBI's 2011 dataset [1], of the 7076 murders by someone with a known relationship to the victim, 1295 were by a husband, wife, boyfriend, or girlfriend. This is compared to, of course, the 2700 murders committed by acquaintances and 1481 by strangers. From this, we can see ~18% of murders are by their partners and ~60% are by someone quite a bit further flung. The reason why both statistics are simultaneously true is that most people have only a few partners (focusing the entire risk on a small group) where as there are many thousands of people in the latter categories (creating a very diffuse risk). So, if you were to bet one which person murdered someone, you would do well to guess one of their partners. If you had to guess if the victim knew their murderer, you should bet that they did not. [1] https://ucr.fbi.gov/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/tables/expanded-homicide-data-table-10 https://ucr.fbi.gov/crime-in-the-u.s/2011/crime-in-the-u.s.-...