[fapjacks]

I don't think "hopelessly wrong" is correct, but we live in the era of clickbait titles, so of course it's not correct. Also, how "long term" is he talking? Because sure, in a billion years, provided Voyager isn't swallowed up by a star, the map is not going to be useful. I actually designed (and got) a tattoo based on this design, except that I updated the data on my version with data from newer astronomical catalogues, in particular a 2002 Australian survey that updated most of the p-dot values for these pulsars. Decoding the original period/p-dot values and locating the pulsars from the catalog data wasn't too hard, and some time after that effort but before I actually got the tattoo, I discovered a work online in which someone else had also decoded the p-dot values and found which pulsars they used. I compared his list to my own and was happy to discover we'd found the same. I also re-encoded the p-dot values which originally had Voyager's "launch date" with the value of my birthdate. I dropped a couple of the pulsars because the 1969 data was very far off from the 2002 data, crossing some imaginary error threshold in my mind enough to be dropped from the map. I personally believe the p-dot values are sufficiently precise as to be uniquely or nearly uniquely identifier data for the pulsars, I mean in the fantasy universe in which this actually amounts to something. It's estimated there are 200,000 pulsars in our galaxy. Surely ET has a computer he can use? Again, in this imaginary universe where ET finds Voyager (or me) and really sets out to crack the code, and we're not using some stupid scenario in which it's found in a billion years...

[sliken]

It's not that ET doesn't have a computer, it's that without omniscience they won't know which of those 200,000 pulsars were pointing at one of 100 billion stars way in the past.

Unless they already orbit all 100 billion stars, but even then it's 100,000 years to collect the data.

Basically with 200k pulsars sprinkled among 100B stars sprinkled over 200,000 light years it's very easy to get lost. A list of frequencies gets less useful over time. You saw a pretty big difference in the data between 1969 and 2002. Imagine the differences in 100k years.

Even a single unknown gravitational interaction could throw the probe way off in speed/direction.

[fapjacks]

Well, I should be clear that the difference in data between 1969 and 2002 was due to measurement error in the original surveys, not due to some kind of actual movement. I have already conceded that over hundreds of millions or billions of years, the map becomes a jumble, but for the next few million years (or from a few million years ago), the map is definitely not "hopelessly incorrect" as the guy states. And actually the whole point of choosing 14 pulsars was so that you had plenty of elbow room for errors arising from e.g. one of the pulsars being flung far and wide by a rogue black hole.

With a computer and observations of "a fair percentage" of the 14 original pulsars, it would still be "fairly easy" to determine which pulsars we're talking about, because from the period alone, you can reduce the total number of candidates to a handful. And so from 200,000 pulsars, from the original 14 period measurements (or 11 in the case of my tattoo), you should be able to reduce the total pulsar pool down to perhaps a hundred or so candidates, just from looking at reasonable possible pulsar periods. Then you can permute through all combinations using a computer to determine which of the pulsars all shared the particular recorded period values at the same time. These values are so precise, and so varied in nature, that you could easily take all 200,000 pulsars in the galaxy, and sort out which 14 you were talking about, if you also could guess that the 14 values were all measured at the same time, that the precise moment chosen to record the period value had to coincide with the precise moment chosen for the other 13 pulsars. Only this select group of 14 pulsars will have had these precise period values at one time, even if many other pulsars in the galaxy have had a specific period value at some point in their history, and this is the genius of Carl Sagan's idea, since you then also can encode a specific time in the data (which happened to have been Voyager's launch date on the golden record, and my birthdate in my tattoo). I still concede that along enormous time scales, it would not be possible to do this, but Carl Sagan's original assertion is definitely not unreasonable across the span of a few million years or so, and a reasonably large enough survey of the galaxy.

[qubex]

Am I wrong in summarising that ”hopelessly wrong map” could also be read as ”a map and a calendar (provided the eventual receivers have good enough records of pulsars going back eons”?

[fapjacks]

I'll explain it from the start, even though you probably know some of this already. The period of a pulsar is the period of its rotation, so how fast it rotates. The "p-dot" of a pulsar is the speed at which the pulsar period decays over time, how quickly the spin rate slows (since everything is always slowing down). And so this is why you wouldn't need a calendar, because by observation, you can get both the period of the pulsar, and the pulsar's p-dot (just by measuring the period a handful of times and doing the math). So ET wouldn't need a calendar, they'd just need to have cataloged the same pulsars in at least a few different sky surveys. Or at least once, because if I gave you a period value of 0.34567890 (hypothetically ignoring units), this is enough information to reduce to the total number of candidate pulsars to one. Unless you want to talk about enormous time scales (e.g. millions of years), but then you'd still be able to reduce the total number of candidates to a handful since pulsars have wildly different periods (and p-dots).

So this is actually how the map "encodes" the launch date of the Voyager probe (and how I was able to change this data and encode my own birthdate in my tattoo): We measure the period of a pulsar at a specific point in time. The pulsar above with period 0.34567890 is a specific period that will change over time due to the natural slowing of its spin (it's p-dot). So all the periods "encoded" into the original map were December 1969. I identified the pulsars, updated the data (distance, direction) with data from the new sky surveys, then advanced the pulsar period value by the p-dot value between December 1969 and the precise date of my birth, then re-encoded those new period values into binary, placed them on the "new" map with updated direction and distance data, and got the tattoo.

I discovered all of this stuff in the research phase of my tattoo. I had some discussions with a handful of professors in the Astronomy department at UH -- where I was going to school at the time -- and for a time it became this "thing" that was well-known in the department. I'd go visit someone for the first time casually during their office hours and they'd mention they'd heard of my idea. Probably one of the best compliments I ever got was from an old engineer that had worked on the SSME (Space Shuttle Main Engine, one of the most complex machines ever made) who said that my tattoo was the nerdiest thing he'd ever seen. Felt good man.

[qubex]

That's how I understand it... the monotonic rate of decay of pulsars' rotation rates allows one to backtrack to when the map was made. The question, therefore, is whether any phenomena can introduce discontinuities in those rates of decay in a manner so unpredictable as to jeopardise the time-keeping properties of the map.

[fapjacks]

Sure, it's a great question. And I don't know for sure, but I would imagine a powerful enough gravitational force, or certainly a collision with a massive body (e.g. another star) would completely change this value. But for that to have happened to enough of the 14 original pulsars so as to make the map legitimately completely useless... I think it would be an amazing, spectacular and infinitesimally-unlikely event.