5 ms·
I think the way the headline is structured make it seem like this is merely a political issue. However the government staff's inability to understand the epoch
by muddi900 2y ago
I think the way the headline is structured make it seem like this is merely a political issue. However the government staff's inability to understand the epoch in ISO standard is definitely a technical issue.
- JPLeRouzic 2y agoIf government computers were still using older than ~45 years COBOL versions with May 20 1875 as a baseline, wouldn't many more people be 150 yo than a few dozen?
- chrisandchris 2y agoFrom Musk statement: > we've got people in there that are about 150 years old From that I don't get whether it is 1 person or N people, whereas N is more probable IMHO. However, how big N is, is unknown.
- muddi900 2y agoThere are thousands, if not millions, of Americans with FNU, or First Name Unknown. And many with LNU, Last Name Unknown. Having well documented existence since birth is fairly recent. Not everyone has clear birth certificates. For example, my father and his younger brother have the same birth year. So it checks that really old people have unknown birthdate.
- rented_mule 2y agoYeah, it's a pretty obvious tell that people haven't worked in any depth with large amounts of complex data when they are surprised that there are huge anomalies. Many, many assumptions do not hold at scale. My mom's birth county's records building burned down in the 1940s. She has no birth certificate. When government agencies hear this, they ask if there are baptism records. The records from the church where she was baptized burned in the 1950s. It's been a massive pain for her, and government data systems are exposed to cases like this. Data at sufficient scale is incredibly messy because many seemingly impossible things have a reasonable probability of being present.
- dkjaudyeqooe 2y ago'People' is plural. If he only found one he would say 'a person' or 'someone'.
- AlecSchueler 2y agoThat only works when people aren't employing hyperbolic rhetoric.
- someothherguyy 2y agoIt says when they don't know the birthdate (a empty value in the field ABS(TIMESTAMPDIFF(YEAR, NOW(), birthdate))).
- omarjobs 2y agoNo, it would be 149years now, right? It will only turn 150 on May 20th this year
- dkjaudyeqooe 2y agoDo you understand the issue? What value would you have entered into the birth date field in this situation? In the absence of a null value and proper null processing, zero seems to be the right value. It's clearly exceptional and intuitive to anyone who encounters it in the data, and it would allow the rest of the system to work correctly. Social security eligibility doesn't have a maximum age but it does have a minimum. If a person claims Social Security without proof of age then their eligibility is dealt with administratively. So the headline is right, this is 100% a political issue, since any fair minded person with relevant experience wouldn't jump to claiming fraud in this situation but rather they'd assume a data issue.
- wakawaka28 2y agoMost people don't expect modern government records to be so chaotic. If so many incomplete and even erroneous entries are allowed (they are, as the social security numbers are in fact reused or else entered wrong) then there will be some significant fraud to be uncovered. That fraud might be harder to find than simply looking for people who are too old. But it's there, no doubt, and more fraud happens every single day. If you look into this stuff you'll find dead people getting payments, and children paying social security taxes as their data is stolen.
- muddi900 2y agoIt reveals the lack of technical expertise at play for the most powerful bureaucrats in the world.