Hacker News Reader

Comment by nmjenkins | Hacker News Reader

What I don't get is, why did they dismiss the plans out of hand 200 years ago when the guy that wrote them was literally famous for producing a revolutionary clock? I mean, with that kind of track record, you'd have thought someone would give it a go.

View on HN

In addition to the other replies to your post which make valid points, you have to remember that Harrison was a self-educated carpenter and clockmaker. The Longitude Committee was composed of the great and the good, and preferred the lunar distance method for getting the longitude at sea.

Edit: child post points out the Longitude Board's examination of Harrison's watches for finding the longitude at sea was much earlier than the publication of Harrison's book. Wikipedia page has the time-line.

You've got your timeline mixed up - by the time the book was published, the committee had fully accepted the clocks and showered Harrison with awards.

The watches? I recollect that Harrison was in dispute with the committee for decades, resolved only by the direct intervention of the king? He never received the full longitude prize that I am aware of. Would welcome correction.

However, clocks were expensive to manufacture and few ships had them. The lunar distance method was a mainstay of navigation until the latter 19th century.

Because the ideas were buried in a book that was mainly about dissing his rivals, and no-one saw past the scandal.

Its worth noting that this recreation uses materials and fabrication techniques not available at the time, too. So there is still no definitive proof that it would have worked.

So to really prove his claim someone would have to create the clock according to the design, using only the technology available 250 years ago? That would be impressive. Perhaps now that the design has been validated someone will attempt it.

Someone did. Martin Burgess. It is mentioned in OP's link. And here: http://www.clockmakers.org/wp-content/uploads/2014/05/Decodi...

Neither link is clear about the methods used to manufacture the clock.

Hijacking your comment to recommend those fascinated by time (and those that aren't!) to look into this year's FOSDEM main track dedicated to time. I especially recommend the talk "Precise Time", which goes in depth about how to track time, what tracks time, to what level of accuracy, how to measure inaccuracies etc. It's not a very technical talk, you can watch it with any layman:

https://video.fosdem.org/2015/main_track-time/

Thanks for posting the link. Is anyone having trouble viewing the videos? I can watch them from my browser plugin, but from VLC or mplayer the video is buggy. In all cases the audio amplitude is very low.

200 years ago, how would you validate his claim? You would have to have a more accurate time reference, and I'm not sure exactly what that would be.

Harrison measured the accuracy of his clocks by observing eclipses of stars by chimneys. We know that for a given star such eclipses happen every 86164.09056 seconds.

For those who might be wondering why it's not every 24 hours, or 86400 seconds, look up "Sidereal Day"[0][1][2].

[0] https://www.google.co.uk/search?q=sidereal+day

[1] http://simple.wikipedia.org/wiki/Sidereal_day

[2] http://en.wikipedia.org/wiki/Sidereal_time

At that level of accuracy, however, you will find that the rotation of the Earth itself doesn't keep time that well.

The Shortt-Synchronome clock [0] was the most accurate pendulum clock, known to have an error of less than 1 second per year. It was used in 1926 to detect tiny seasonal changes in the Earth's rotation rate. An evaluation of the clock in 1984 revealed it was even more accurate than thought --- it was in fact accurate to within 1 second per 12 years, the discrepancy being due to "the slight changes in gravity due to tidal distortions in the solid Earth caused by the gravity of the Sun and Moon."

Harrison's claim of 1 second error in 100 days is clearly in the same order of magnitude as the Shortt-Synchronome's known error rate of 1 second per year in 1926, and thus would have been also affected by the tiny seasonal changes in the Earth's rotation rate.

[0] https://en.wikipedia.org/wiki/Shortt-Synchronome_clock

This is why we talk about mean time and still have to account for discrepancies as leap seconds. That's not an issue with the clocks, just the earth.

However, the situations are actually pretty different. If you look into the study of the Shortt-Synchronome clock, you'll see it's actually talking about the gravitational effects of the sun and moon on pendulums. The Harrison clocks had no pendulums.

The Shortt-Synchronome is still a pretty amazing clock, though.

  > The Harrison clocks had no pendulums.

The clock referred to in the article does have a pendulum.

Oh, I see, I thought they were asking how you would validate his claim of the "revolutionary clock"(s) he was famous for in the first place.

> At that level of accuracy, however, you will find that the rotation of the Earth itself doesn't keep time that well.

If the main purpose of the clock was to measure longitude by observing the position of the stars, "only as accurate as the Earth's rotation is consistent" is as good as you could possibly need.

Eyeballing the figure at http://en.wikipedia.org/wiki/Fluctuations_in_the_length_of_d... it looks like you could expect stellar time to be off proper time by not more than around 0.1 seconds in 100 days, since the width of the band of fluctuations is around 1 ms. (I'd say 0.01 seconds since in a random walk the error grows with the square root of time, not linearly.) Maybe this fluctuation graph has subtracted out seasonal variation?

As long as you're at a fixed location, how about a sundial?

To the second? I'm not a sundial expert but have seen a few and those that I have seen have been vaguely useful for telling which hour it is. Maybe even which half hour. I assume you could just lengthen the vertical to get a longer and more accurate shadow. But it's still a bit hard to see this being a reference.

Edit: here is a link to some accuracy measurement. I have no idea what the record is, but these ones were getting down to 30 seconds ish. http://www.sage.unsw.edu.au/currentstudents/ug/projects/o%27...

That part is easy: you just compare it to local solar noon. Earth's rotation is considerably more accurate than this clock's target.

For more info about the number of seconds in a solar day, see this awesome Wikipedia page: http://en.wikipedia.org/wiki/Leap_second

For more info about the number of seconds in a solar day, I would rather recommend the table at http://en.wikipedia.org/wiki/Solar_time#Apparent_solar_time . The Leap_second article discusses a much smaller variation (~2 ms) on top of the big cyclic variation (~20 s).

The most obvious way would probably be to build two of them and see how accurate they were against each-other.

Alternately, you could use astronomical angle measurements. You can get quite accurate measure if you don’t move around.

(The reason determining time was difficult on ships is that you need to know precisely where the ship is to figure out the time from the stars, or precisely what time it is to figure out position.)

You could simply build 10 of them and test them against each other. Even though statistical significance hadn't been discovered yet, I'm sure someone could have worked out the math.

There would have been no way to validate it because they couldn't measure the accuracy of a timepiece to within even a few seconds.

Can you support this claim? I'm not skeptical so much as curious. How did people keep more or less consistent time back then?

The only option was to compare the reading of a clock, to an astronomical observation such as the zenith of the sun.

Right. And the zenith can be identified by eye or using a sundial to within an accuracy of a minute or so, but not to a matter of a second or two AFAIK. Maybe using a sundial the height of the Eiffel Tower?

It does seem pretty daunting. A tall sundial would not have enough mechanical stability to be accurate, especially due to wind loading. AFAIK at least one of the people who measured the circumference of the earth, did so by observing a shadow at the bottom of a deep well. Still, a shadow would be pretty washed out over the span of, say, 15 meters, unless some sort of focusing optic were involved.

A tall cliff? A building with a spire on top? An obelisk? Guess now I'm going to have to go and find a tall building with a spire and check how recognizable the shadow of that spire is!

I'd certainly be interested in knowing. One thing that fascinates me is how the ancients made any headway with tools that were so crude compared to what we now have. For instance I always enjoy reading a non-algebra or non-calculus based math proof for something where I would instinctively turn to one of those tools.

One thing it shows is the irrepressible drive of human curiosity.

They didn't keep consistent time as we know it. They scheduled things with fuzzy ranges instead of precise timestamps and used public reference clocks which were inaccurate but useful for two people to know the meeting is whenever the church clock says it's 2 o'clock, regardless of the true astronomical time.

Reply on news.ycombinator.com