6 ms·
Conspiracy and an off-by-one error
- eevilspock 12y agoSuperb investigative journalism. And love the idea of using GitHub, where anyone else can suggest improvements or additions, and it's all transparent.
- TazeTSchnitzel 12y agoWell, it's a Gist, so it's tracked by git and can be forked, but I don't think you can make pull requests.
- agf 12y agoOff by one errors affecting geopolitics. That's a scary thought. Any other examples out there? There are things like false missile launch positives during the Cuban missile crisis...
- eudox 12y agohttp://en.wikipedia.org/wiki/Stanislav_Petrov http://en.wikipedia.org/wiki/Stanislav_Petrov Sunlight reflected on clouds, misidentified as the launch of a nuclear missile.
- Demiurge 12y agoThis doesn't really affect geopolitics any more than any other conspirology. No fingers changed direction because of this.
- agf 12y agoI agree, but change the context just a little and the same error could have convinced people the evidence was faked. The potential is there.
- cpeterso 12y agoIt's not quite geopolitics, but I read that Facebook showed that some of the Winklevoss lawsuit's emails were (likely) fake because the email headers specified the wrong daylight savings time zone for the year that were supposedly written. Of course I can't find a reference online now..
- ChuckMcM 12y agoWell there was the time when the country was mislabelled and the army invaded accidentally. (http://searchengineland.com/nicaragua-raids-costa-rica-blames-google-maps-54885 http://searchengineland.com/nicaragua-raids-costa-rica-blame...)
- mikeash 12y agoPeople know the Cuban Missile Crisis came close to a nuclear war, but many don't know how close. A human off-by-one error there could have started it all. There was a moment on a Soviet submarine where the captain decided to use a nuclear torpedo. He needed two other men to agree to it, and one refused and resisted their attempts to persuade him. If he had been a little weaker willed, or was in a bad mood that day....
- Houshalter 12y agoOther close calls prevented by a single person: https://en.wikipedia.org/wiki/1983_Soviet_nuclear_false_alarm_incident https://en.wikipedia.org/wiki/1983_Soviet_nuclear_false_alar... https://en.wikipedia.org/wiki/Norwegian_rocket_incident https://en.wikipedia.org/wiki/Norwegian_rocket_incident
- jcr 12y agoEcoded date from the downloaded youtube file: Encoded date : UTC 2014-07-17 10:44:19 Exact upload date using the API: <yt:uploaded>2014-07-18T10:43:10.000Z</yt:uploaded> > "you will find that the encoded date timestamp is 24 hours behind the upload date" Excellent analysis, but the last statement is reversed. It's probably just a typo/oversight (like that's never happened to me before (sigh)). The encode timestamp is 24 hours before the upload timestamp, which is impossible since it must be uploaded before it's encoded.
- fixermark 12y ago"Behind" means "before," depending on whom you are speaking to. Relative-time prepositions in English are pretty badly messed up.
- harshreality 12y agoWhat's messed up about "before" and "after"? I don't think native speakers will use "in front of" or "behind" to describe time coordinates. Those words are used to describe relative positioning of events, and in such use I think they have the opposite interpretation from what you suggested. "Event A occurred behind event B", for example, or "A is behind schedule" (although you could then argue that scheduling has its own semantics for "behind"). However, even "I'm running behind" means running late, not early. Even on a timeline, it's not clear whether the point of perspective for determining "in front of" or "behind" is a point in the past or a point in the future. Simple: use before and after, or earlier and later. btw, the overall semantics of "a means b, depending on whom you are speaking to" made me cringe even though I know you meant "A usually means B, but it depends..." (which I don't think is even true—see above—but that's a factual not semantic complaint). Taken literally, you could just as easily have written "a means (not b)..."
- fixermark 12y ago"Before" and "After" are fine. "Behind" is not incorrect in the original text, but it can be terribly, frustratingly ambiguous. As you've noted, "I'm running behind" means "I'm running late," i.e. I will arrive after the correct time. But "My clock is running behind" means "My clock is running slow," meaning it shows a time before the correct time. Similarly, when projects are behind schedule (i.e. likely to complete after the correct time), we sometimes solve this by pushing the schedule back... Meaning by pushing the correct time into the future, as opposed to most contexts where "pushing something back" would mean "pushing it further behind." So we push schedules back to move deadlines forward. Personally, I usually use "into the future" and "into the past" to be precise. Treating time with spatial-positional-relative language is fraught with inaccuracies. (Fun fact: did you know "biweekly" means either "twice a week" or "once every other week?" I can't think of any other measurement unit that has a factor-of-four error built into its very definition).
- dj-wonk 12y agoThere is considerable overlap between good investigative skills and debugging software. I'm glad to see both being used together!
- TazeTSchnitzel 12y agoThe comments on reddit are quite interesting too: http://www.reddit.com/r/programming/comments/2b4kpg/conspiracy_and_an_offbyone_error/ http://www.reddit.com/r/programming/comments/2b4kpg/conspira...
- powertower 12y agoExcellent review and analysis! I watched the original video (about the time-stamp) yesterday and someone in the comments section said about the same (that there is a bug in the system). But I wouldn't label the original poster a conspiracy theorist over this. It's perfectly normal for someone to assume YouTube is working as it should, in which case, seeing the early time-stamp should definitely set off some question marks. In combination with the current regime's use of indiscriminate violence, misinformation, and downright lies, it reasonable to question anything they put out.
- ajarmst 12y agoWhen "elaborate multi-agency government conspiracy" and "off-by-one bug in non-critical section of code" are your two competing theories, both Occam's Razor and Hanlon's Razor (two important rules of reasoning) compel the selection of the latter. So it's not really "normal" (if by that you mean reasonable) for someone to assume that YouTube is elaborately bug free if that requires the existence of shadowy conspiracies that are unsupported by other evidence.
- A_COMPUTER 12y agoThere seems to be two ways to apply Occam's Razor. The first, used by people actually tasked with investigating something, is to prioritize exploration of simpler explanations before more complex ones. The second, more commonly used by Internet bystanders, is to treat it like an iron law, immediately cut off all alternate avenues of inquiry and give people degrading labels. Occam's Razor is best as an investigatory rule of thumb and not a thought-terminating cliche.
- makomk 12y agoIn this case, even the most cursory attempt to explore the simple explanation would've shown it was the right one, though.
- antocv 12y agoAn astute consipracy theorytician will just claim that Google has changed their API and changes the timestamps on the fly after their initial blunder was discovered. Just to stay, that evidence does not usually persuade beliebers.
- konstruktor 12y agoGreat technical analysis, but poor choice of words. Why does the author deem it necessary to sensationalise it by using terms like conspiracy theory and conspiracy theorists? Those are rather derogatory and loaded, but introduce no actual information except that a conspiracy is involved: http://rationalwiki.org/wiki/Conspiracy_theory http://rationalwiki.org/wiki/Conspiracy_theory Note that both the currently most accepted hypothesis (rebels armed by the Russians did it) as well as the Russian party line (Ukranians did it to frame the Russians) involve government conspiracies and are, thus, conspiracy theories.
- gizmo686 12y agoIsn't the general hypothesis that whoever did it mistook the plane to be military. If so, this does not seem like a conspiracy, as much as a mistake (or incompetence) followed by a cover up.
- jqm 12y agoExactly. Most likely a military mistake similar to Iran Air Flight 655. These things can happen. When political leaders seize on these types of events to promote their agendas however, some people understandably become more suspicious that a conspiracy is afoot.
- acqq 12y agohttp://en.wikipedia.org/wiki/Iran_Air_Flight_655 http://en.wikipedia.org/wiki/Iran_Air_Flight_655 "shot down by the United States Navy guided missile cruiser USS Vincennes on 3 July 1988," "All 290 on board, including 66 children and 16 crew, died" "The Fogarty report stated, "The data from USS Vincennes tapes, information from USS Sides and reliable intelligence information, corroborate the fact that [Iran Air Flight 655] was on a normal commercial air flight plan profile, in the assigned airway, squawking Mode III 6760, on a continuous ascent in altitude from take-off at Bandar Abbas to shoot-down.""
- acqq 12y ago"whoever did it mistook the plane to be military." It doesn't have much sense in the context, the civilian planes flew there all the time: http://www.theguardian.com/world/2014/jul/19/mh17-changing-course-storms-pilot http://www.theguardian.com/world/2014/jul/19/mh17-changing-c... "the route (...) is one of the "aerial motorways" between northern Europe and south Asia. Malaysia Airlines was one of more than a dozen that flew the route on Thursday. Its flight MH17 was only a few miles from an Air India Boeing 787 and a Singapore Airlines 777 when it was shot down."
- anatoly 12y agoVery nice writeup! To finish the puzzle, consider the following comment in one open-source mp4 library: https://github.com/l-smash/l-smash/blob/master/core/box.h https://github.com/l-smash/l-smash/blob/master/core/box.h "According to ISO/IEC-14496-5-2001, the difference between Unix time and Mac OS time is 2082758400. However this is wrong and 2082844800 is correct." MP4 stores timestamps by Mac epochs. The wrong value of the difference between Unix and Mac epochs is wrong exactly by 24 hours and is hardcoded in the MPEG-4 reference software, from which it made its way into a number of libraries. (I work for Google and have contacted the right people in the attempt to address this bug soon).