September 25, 2026

Open Thread 197

It's time once again for our regular open thread. Talk about whatever you want, so long as it isn't culture war.

Overhauls are Understanding Hull Symbols, Auxiliaries Part 3, ICNW Part 5, the Arleigh Burke class, Standard Part 1, Evan's review of Turner Joy, and for 2025, Exotic Hulls Parts two, three and four, as well as Fighting the Last War.

Comments

  1. September 26, 2026quanticle said...

    This is a question inspired by my recent acquisition of a mechanical watch as a (early) birthday present. One difference between mechanical watches and quartz watches is that a mechanical watch is far less accurate than quartz. Whereas even cheap quartz watches are accurate to within a second a month or so, a mechanical chronometer will be off by several seconds even after a day. So given that the measurement of longitude required accurate timing of star sightings, how did mariners keep their clocks synchronized prior to the advent of electronic timepieces?

  2. September 26, 2026personofinterest said...

    The short answer is marine chronometers are not nearly as constrained as wrist watches and are much better.

    For example, the escapement used in a wrist watch is not orientation independent; it runs at a different rate depending on which way gravity goes. The watch is factory set with a rate that ought to correspond to average orientation experienced during casual wear, give or take. Nowadays balance wheels that help correct this deficiency (tourbillon) are reserved for extremely expensive watches mostly for marketing reasons. As an interesting consequence of this, there is a market for artificial wrists that move your watch collection around while not wearing them in order to simulate casual wear and reduce the runout. Similarly, wrist watches basically never contain temperature compensation. They are factor set assuming they are warmed by body temperature in moderate ambient conditions. When you take them off and they are no longer warmed by body temperature they run at very a different rate, whereas changes in ambient temperature make them run at modestly different rates. Said artificial wrist may include a heater.

    By way of illustration, during the first sea trial of John Harrison's first sea watch (H4), the runout was about 5 seconds over 81 days. Corresponding to a longitude error of about 1 nautical mile. They only got better from there.

  3. September 26, 2026personofinterest said...

    The short answer is marine chronometers are not nearly as constrained as wrist watches and are much better.

    For example, the escapement used in a wrist watch is not orientation independent; it runs at a different rate depending on which way gravity goes. The watch is factory set with a rate that ought to correspond to average orientation experienced during casual wear, give or take. Nowadays balance wheels that help correct this deficiency (tourbillon) are reserved for extremely expensive watches mostly for marketing reasons. As an interesting consequence of this, there is a market for artificial wrists that move your watch collection around while not wearing them in order to simulate casual wear and reduce the runout. Similarly, wrist watches basically never contain temperature compensation. They are factor set assuming they are warmed by body temperature in moderate ambient conditions. When you take them off and they are no longer warmed by body temperature they run at very a different rate, whereas changes in ambient temperature make them run at modestly different rates. Said artificial wrist may include a heater.

    By way of illustration, during the first sea trial of John Harrison's first sea watch (H4), the runout was about 5 seconds over 81 days. Corresponding to a longitude error of about 1 nautical mile. They only got better from there.

  4. September 26, 2026Hugh Fisher said...

    A mechanical chronometer for marine navigation is much more accurate than a mechanical watch, and the early ones came into use because they didn't need synchronisation. They're also precision bits of machinery. John Harrison's H-3 took about twenty years of research and development, and weights 27 kg. There's a good book "Longitude" by Sobel, Dava about the problem and how it was tackled.

  5. September 28, 2026quanticle said...

    I didn't realize that even early marine chronometers were that accurate.

    Also, while I was looking into it, I found out that the US was transmitting time signals by radio as early as 1904, which allowed ships at sea to synchronize their chronometers at sea.

  6. September 28, 2026Anonymous said...

    personofinterest:

    Similarly, wrist watches basically never contain temperature compensation.

    A quartz watch could do that at the expense of power usage (for real accuracy atomic clock chips small enough to fit in a wristwatch are available, but even ordinary quartz watches are accurate enough that you could usually get away with only setting them when the cell is replaced) but could a mechanical watch do that and still fit on a wrist?

  7. September 29, 2026Fionn said...

    quanticle, consider reading Dava Sobel's Longitude, in case you haven't already read it. It's pretty wild looking back that some guy way out in the countryside managed to build a series of chronometers when the elite watchmakers were struggling to get anywhere near the required performance standard. The book is dramatised, but OTOH the story is of a total outsider doing the impossible so some drama is warranted.

  8. October 09, 2026Eltargrim said...

    Test comment for testing

  9. October 10, 2026Ian Argent said...

    @quanticle

    Also, while I was looking into it, I found out that the US was transmitting time signals by radio as early as 1904, which allowed ships at sea to synchronize their chronometers at sea

    I wouldn't have known that before today, but that seems like one of the very first things I'd do with radios and one of the simplest, so I would expect it to show up very early in the timeline

  10. October 21, 2026Said Achmiz said...

    Hi bean and blog readers,

    You may have heard about the huge PHP vulnerabilities that were announced today. As the software that this blog runs on is PHP-based, I just wanted to let you know that the appropriate server updates to fix those vulnerabilities have been applied, and no security problems are expected. (That is why the site was down from last night to just now—to prevent any attacks before the updates were in.)

    As always, please let me know about any problems you encounter. Otherwise, nothing needs to be done about this, as the problem has been taken care of.

  11. October 21, 2026bean said...

    Thank you again for taking care of all the technical stuff, and particularly for getting the fixes done in a timely manner.

  12. October 22, 2026QB said...

    As of yesterday, more oil was coming out of the middle east than before the war started. Only trouble is: it costs $20 to get it out, versus $3 per barrel before the war.

    ... today is yet to be recorded.

  13. October 23, 2026Hugh Fisher said...

    @QB, from the Tanker Trackers data via Sal Mercogliano nope, the average oil transits per week / month are still down. The USN appears to be escorting tankers through Hormuz in convoys rather than individually sailing as was done pre-war. So pick the day a convoy goes through, lots of oil coming out. Pick a different day, none.

  14. October 24, 2026ike said...

    @Said

    That probably explains why there is now a backslash in the number of comments at the end of a post.

    by bean 13\ comment(s)

  15. October 30, 2026QB said...

    @Hugh USN is using small smuggling boats, not tankers. Very not surprised if tanker traffic is down. Also, there was the small matter of Iran starting shite directly after the "we're getting more oil out" comment.

Comments from SlateStarCodex:

Leave a comment

All comments are reviewed before being displayed.
Name (required):

E-mail (required, will not be published):

Website:

You can use Markdown in comments!


Enter value: Captcha