It will be easier to boost the Earth’s spin back to a 86,400-second day than to fix all the code.

It's been suggested that perhaps one way to find evidence of extrasolar life, would be to find exoplanets whose years are integer multiples of their rotational periods, i.e. they've done what you describe just to eliminate leap years and make their calendars easier.

That is fascinating! Do you have a reference?

I think he might be sarcastic. There is no known method I know for measuring the rotational period of extrasolar planets, without having enough observational power to defiantly notice the astronomical engineering required to alter an inhabited planet’s rotation rate.

Actually I think if we follow the current trend in technology advances in astronomy, we will definitely be able to measure planetary rotation well before we have the angular resolution to spot anything of consequence on an exoplanet. We currently have the technology to measure stellar rotation bia Doppler shift. https://en.m.wikipedia.org/wiki/Stellar_rotation

This technique will definitely be applied to exoplanets as the angular resolution and spectroscopic sensitivity are improved. In fact, improvements to both if these numbers are being relentlessly pursued by a veritable army of scientists around the world.

Spectral broadening only needs one pixel, and enough spectral sensitivity that the broadening isn’t entirely in the margin of error. We are pushing For ever more accurate spectral sensitivity as we try to use Doppler measurement to hunt for exoplanets.

Meanwhile the ongoing efforts to obtain greater angular resolution, has basically been one of the core pursuits of the astronomical community for centuries. We are building massive telescopes and advanced compensation systems in the form of adaptive optics to get the very best out of these telescopes.

So yeah I think we’re going to have rotation rate measures before we see much more than a pale blue dot.

"You see that big city past that mountain range? Use it as a marker to measure the planet's rotation."

If we measure this city's position then we'll know if there's life on the planet.

You get the joke.

In the Three Body Problem series by Cixin Liu, the orbits of planets and moons in the solar system are altered to make colonization and terraforming easier.

Wrong. In the book they build the bunker stations behind Jupiter but NOT in orbit of jupiter or its moons and even that station keeping takes a tremendous energy.

But how do we know which base number system they're using!?

Mathematically that's irrelevant to finding planets with an integer ratio of days per year. Integers are still integers no matter what base you express them in.

What if the base isn't an integer?

Base e for life.

Or a prime based number system.

Every base is base 10.

And speak perfect English according to the Stargate universe.

I don't think that was true in the movie.

Indeed.

That's actually a cool idea..

I would just change the duration of a second. It would be probably easier to change a few physics constants here and there than to fix all the broken code.

If only you were manipulating second only. You would also be manipulating metre, candela, and ampere, assuming old definition of SI, or metre, kilogram and kelvin assuming new definition of SI.

Can you explain how the duration of a second affects the metre?

Not OP, but "The metre is defined as the length of the path travelled by light in a vacuum in 1/299 792 458 second" - https://en.m.wikipedia.org/wiki/Metre

The duration of a second is defined using Caesium-133 [1], not using anything else like daytime or rotational speed. It's independent of such factors and will always be the same in a given frame of reference. No need to update those values.

edit: I think it will lead to the same or more problems because everyone would have to change all time-related constants. Easier to add a second to the day constant (because the day constant is a subset of the seconds constant, but the seconds constant affects many more aspects than just the day constant).

[1]: https://en.wikipedia.org/wiki/Isotopes_of_caesium#Caesium-13...

But in that case, doesn't the "second" constant change (or the coefficient) and a metre remains the same length?

A meterstick would remain the same length, but the SI meter's length would change.

Unless these physics constants are, well, constants put in the same code. How many ballistical systems treat ‘standard g’ as a variable? (hint: it depends on seconds squared.)

Someone should write a short story about this.

Code that assumes that is wrong today already. In fact there is no time.Day constant in the Go standard library

Probably not. https://what-if.xkcd.com/26/

I don’t care if this goes against HN guideline and if I am downvoted, but that was the geekiest joke of the millennia :D

Geeks make it go around yo.

In Vernor Vinge's A Fire Upon the Deep and A Deepness in the Sky novels, the characters in the far future track time using megaseconds (~11.6 Earth days) to sidestep the problem of days with different lengths. They still count from the 1970-01-01 Unix epoch, but the characters mistakenly think that date is when humans landed on the moon.

There's also such a thing as a Programmer-Archeologist, can you imagine dealing with 14,000 years of nested dependencies? One key plot point in one of the novels, a totalitarian group is running a omnipresent surveillance network but has no idea about the software running underneath it. They set it up using the GUI and are using it but never looked under the hood. The group attempting to get out from under the grasps of the totalitarians has a programmar-archaeologist who digs into the underpinnings of the software and hijacks it from a lower level of its operating system.

I forgot about the Programmer-Archeologist! Another point from (I think) these novels that stuck with me was Vinge's description of the "software midden heap". I'd like to think we can clean up some middle layers from our computing stack in the future, but with some old software only available in emulators (possibly running in WebAssembly in a browser), I am not so sure.

In fact, I realized I made an nearly identical comment here on Hacker News back in 2011. :)

https://news.ycombinator.com/item?id=2702231

The Matrix expanded-universe also has some surprisingly deep and complex lore regarding such interactions between the humans and machines and everything between and around them. Those movies and their spinoffs are worth revisiting now and then.

Yes but at least that far in the future we'll finally be rid of JavaScript.

That's the good way to write code, surely? Then you can just change it once to DAY_SECONDS=86401.

And they can also pray that their development environment still works on Linux 9.21

Maybe GNU Hurd will be on version 1 by then.

Maybe humans will no longer be responsible for maintaining software

Fine but now you have to find every packaged binary using the old compiled version and update that too.

Quantum compilers will have that done by Tuesday afternoon.

Maybe.

I'm pretty sure some nifty JavaScript code will come to save the day!

Machine learning will analyze your thoughs and output wasm, then run it a os that is sandboxed in a webrowser.

Pretty sure the Y10K problem would be a bigger issue.

A few centuries earlier that would have been worrying about adding extra teeth to the gear wheel driving the mechanical clock.

That would be quite terrible, if in fact 50K years from now there was such a thing as computer code.

They're missing a bunch of possibilities. Maybe a second is lengthened? Maybe we decouple timekeeping from the rotation of the Earth? Maybe we spin the Earth back up?

Lengthening the second is unlikely: for every physical phenomenon with a duration (from the rotational period of other planets to the frequency of atomic transitions), you would forever after have to wonder, is that value quoted in"new seconds" or "old seconds"?

Redefining units leads to ultimate backwards compatibility problems.

Why don’t we just 6 minutes at the end of each year? Seems to be more practical to me

You want December 31st to be 6 minutes longer than December 30th?

I mean it's not like some days are a whole HOUR shorter / longer than others in the countries that follow daylight saving time.

Yeah... the way you ask, makes it seem like a pretty dumb idea :-/ is it?

Could you just add a leap day every 86,400 days?

At the midpoint of that period your clock day would be 12 hours off from your rotational day.

Leap hour every 1440 days would be a nice happy medium.