Do you consider the original Y2K to have met this bar?

Doesn't look like it to me.

https://en.wikipedia.org/wiki/Year_2000_problem#Documented_errors

Seems everyone thinks the threshold should be raised. Would $1 billion be a more reasonable criterion?

@IsaacKing Probably reasonable, but if you change the threshold for "significant" by a factor of 100x I think that warrants making a new market rather than changing the resolution criteria of an existing one.

@A Agreed. There is obviously a threshold for every % bet. Don’t see why you would change existing market

Alright I changed the description slightly, sticking with the general gist. Shouldn't change in any substantive way from here, just clarifications.

@IsaacKing Pretty big change imo. Most expensive bugs are still shielded from the vast majority of the public. Just costs millions of dollars to fix. “Tens of millions” could have just been a week of delays for a thousand developers

I don’t see why you didn’t just make a second market.

@capybara The description said it was a work in progress, and clearly the previous definition was no good, so I think it's reasonable to change it to something that more closely accords with the title.

Making a second market wouldn't solve the problem with this one.

@IsaacKing What was the problem with the last one? Your definition for significant was valid. Just happened that the likelihood was higher than you wanted. No one thinks tens of millions of dollars is insignificant.

Tens of millions of impact is a relatively small threshold I think, but it might depend exactly how strict you are.

I think we can work on the criteria a bit more? Even with multiple people asking I'm still genuinely not sure if the following invented scenarios or do not count:

- In 2035, Walmart discovers that a device used in many places in every store in the country will break when the Y2K38 bug happens. They replace all the devices before 2038 at a cost of 100 million dollars (~$20k per store) and news articles are written about it. The 'bug' was present in the devices but they were generally removed before its impact was felt.

- One supplier of a subcomponent for John Deere tractors fails to fix a Y2K38 bug in its computer systems before the deadline. The entire line needs to be shut down for 2 weeks to get an emergency fix in place. The cumulative effect of the shutdown for this supplier results in John Deere having 100 million dollars of sales lost (Yearly revenue is in the tens of billions, so a small effect but measurable).

@Eliza Yeah, I agree tens of millions is too small a threshold. I would bet YES purely based on that if the criteria weren't still being workshopped.

The Walmart example doesn't count, that's just responsible preparation.

I think the John Deere example is borderline, not sure.

The John Deere example will count if at least 1 million people wanted to buy a tractor during that time and had to wait.

No, the genocidal AI will handle it.

No, because all critical systems will be running 64-bit OS by that date.

@JoePolchlopek and besides the major embedded operating systems all have workarounds that combine two 32-bit numbers in a way that allows them to represent time as though it was a 64-bit system. I've checked around and my embedded systems and IoT devices all have these workarounds in place, betting on NO as well for that reason.

Just clarifying, should the following count as yes: there are IT issues in other countries leading to say supply chain issues or flight cancellations that indirectly affect US above the impact threshold.

@capybara That'll count, yeah.

Hm, I think "tens of millions of dollars" might be too small a standard for significance. Wikipedia says the U.S. spent $13 billion fixing problems in 2000 and 2001: https://en.wikipedia.org/wiki/Year_2000_problem#Cost

Though I'm not exactly sure what that number means. It sounds higher than I would have thought.

@JosephNoonan If you look into the cited article, it says things like "For IT, the millennium bug represented an unprecedented opportunity to modernize. "I don't think we'll ever again get that opportunity to say, 'Hey, we need a blank check to get everything up to date.' We put a lot of fear into the CEOs back then," says Michael Israel, who, as chief operating officer at IT services provider AMC Computer Corp. oversaw the Y2K remediation work at Continuum Health Partners and its affiliated hospitals in and around New York City."

So I'm pretty sure those numbers include lots of miscellaneous upgrades that weren't actually needed to fix the Y2K bug.

Does it have to cause damage via actual computer bugs occurring, or is the fear of this happening enough? Y2K had a significant impact in the U.S. based on the latter standard but not the former.

@JosephNoonan The former.

Not for med. I'll be dead.