Hacker News Reader

Comment by monocasa | Hacker News Reader

Luckily clock skew already happens and is handled more or less by the current systems. The difference between the system moving at relativistic speeds versus a crystal being slightly out of spec add up to the same result. Incorrect number of ticks for a given proper time interval.

That's part of the genius of Leslie Lamport, embracing the fact that your distributed clocks aren't consistent, so start looking at effects through distributed systems from a light cone perspective just like you do with relativity.

View on HN

Clock skew is in error. Relativistic time dilation is not. Both clocks will be right and disagree. You can't escape answering the question: do I care about the time elapsed at home or the time elapsed here?

At the end of the day there's always skew; crystals in systems aren't made to perfect tolerances. In fact I'm not even sure how you would define perfect tolerances for clocks in a meaningful way.

So at the end of the day it doesn't matter if it's in 'error' or a fundamental to the physics even under perfect conditions. Neither are really under your control and both have the same answer.

> You can't escape answering the question: do I care about the time elapsed at home or the time elapsed here?

We already have that concept, in the difference between CLOCK_WALL and CLOCK_MONOTONIC. You use monotonic for "local temporal reference frame" and wall for "proxy of globally visible temporal reference frame". Wall shifts around anyway as NTP or what have you figures out how skewed off of "actual" time you are.

If you need time elapsed then you implicitly need a frame of reference in the same way that a clock time needs a time zone to be meaningful in a global basis. In practice if you wanted this it would mean having a timer colocated within your frame of reference as there is no scalar representation of a frame of reference that would allow you to do timestamp calculations. Timestamp math only works as a hack because we mostly stay relativistically close to Greenwich.

Astronomers use mathematical models of the solar system to translate between barycentric time, geocentric time, and terrestrial time, which are all different relativistic frames of reference. Only TT has actual clocks: TAI ticks at the same rate as TT, tho for historical reasons there is a fixed offset. That isn’t counting the clocks derided from observing the orbits of the moons and planets…

> You can't escape answering the question

Yes you can. When error swamps the difference between two things, you don't have to choose.

Do you set your analog alarm clock to wake you at 6:00:00 or 6:00:05?

Is that not choosing?

The answer I was implying is: You can't distinguish between them, you put the arm in the vague vicinity of both. So no it's not choosing. You could choose which to aim at if you really wanted to, but the end result would be basically the same. You can't choose which one it's set to, because it's so far below the noise floor.

Today it is. My point is if a ship goes on an 80 year journey at an appreciable speed then its wall clock will disagree by an appreciable amount. If we have hundreds or thousands of such ships and they communicate with each other, someone will need to choose what clock they care about and it won't be the Earth's clock.

Well I meant the real you/me. Not a hypothetical spaceman.

Anyone on Earth can just say their computer clocks are set to "Earth", ignore relativity, and use their normal clock skew correction algorithms.

It's not "You can't escape answering the question". It's "Interplanetary spacemen can't escape answering the question." And those spacemen might not even use seconds.

We already define time relative to 1970-01-01, relative to the mean solar noon in Greenwich.

We should also define the length of a second to mean "at sea level in Greenwich".

This just means the "length of a second" wanders around randomly and is thus useless. So now "a second" isn't any particular amount of time, and you've lost all ability to really measure time except in a vague inexact sense.

The quest to eliminate irreproducibility from metrication is why SI did all that work to finally eliminate the kilogram prototype. The second was defined reproducibly decades ago because trying to define it based on our planet's inconstant spin was such a failure.

The fact that the Earth's spin is not constant is perhaps frustrating, but like other facts just insisting it isn't true doesn't help you.

Right, but general relativity matters for how astronomers define their timescales, and how they are measured and maintained in practice: there are different scales for barycentric time (at the centre of gravity of the solar system), geocentric time (centre of the earth) and terrestrial time (on the surface of the rotating geoid). TAI is supposed to tick at the same rate as TT, so it depends on a model of an equal-gravitation version of sea level (the geoid) and the relative height of the time labs. In particular, NIST is a mile high in Colorado so their clocks tick a lot faster than (say) NPL by the Thames in west London.

Sure, the quest for Universal Time is despite the name necessarily a local quest so you can't meaningfully export your results to creatures orbiting a distant quasar as you can with the SI units.

But the size of the discrepancy between Colorado and London is far smaller than the discrepancy Leap Seconds are concerned with right?

Yeah, gravitational time dilation is about 1e-16 seconds per metre of elevation near the surface http://leapsecond.com/great2016a/index.htm so Boulder is about 1e-13 fast. The rotation of the earth is stable to about 1e-9 http://leapsecond.com/ten/ which is also about the same as the rate difference between UT1 and UTC (0.2 ms / 86400).

During the discussions leading to the inception of the SI second the bigwig standards folks nixed the consideration of relativity, basically paraphrased as "If you physicists do not stop arguing about relativity we will never redefine time in a way that removes it from the astronomers."

> This just means the "length of a second" wanders around randomly and is thus useless

Why would it? My redefinition just means the length of a second is defined as a given number of cycles of caesium in a gravitational field equivalent to that of Greenwich.

If you're in a different gravitational field, you'll just have to convert the values easily.

These days, the length of a second is defined by the difference in frequency between two ground-state energy levels of a caesium-133 atom, which works out to 9192631770 cycles per second. No need to check tide tables.

That definition is still relative. It still requires as additional factor the place where the measurement is taken, due to Einstein's theories.

That's why I proposed changing the second definition to 9192631770 cycles in a gravitational field equivalent to that of Greenwich.

Sea level when the tide is in, or out?

“sea level” is short for “mean sea level”

Not trying to troll, but... What happens if/when global warming makes sea level rise? Do we have measurements that are still based on "sea level"?

https://tidesandcurrents.noaa.gov/sltrends/faq.html

> Tidal datums must be updated at least every 20-25 years due to global sea level rise. Some stations are more frequently updated due to high relative sea level trends.

I mean, I feel like that's fairly obvious, and the playful pedantry of my comment in a discussion on the very definition of basic unit of measurement was overlooked.

If we are going so far as to rigidly include "sea level" in the definition of "a second", it would probably help if we say that that we mean mean sea level.

Know what I mean?

Reply on news.ycombinator.com