[drostie]

It is extremely inconvenient that time is base-24 and base-60 while everything else in the metric system is base-10; it leads, for example, to a difficult conversion between the "meters per second" of the physics class and the "km/hr" of the highway. One surprisingly frustrating thing is that if you view a day as a million "instants" then an instant is actually a very human number, about the human reaction time or so; and a kiloinstant is a very human timescale too; 86.4 seconds or just shy of a minute and a half. We already know that when someone says "be there in five minutes" they mean 7-8; if they said "be there in five ki" they would be more accurate.

Fun things:

1. I wrote an HTML5 base-10 clock here with some togglable layers. Try out base-10 time: http://drostie.org/time/ . It's actually much easier than reading a normal clock because it's a digital readout, "8, 5, 6" rather than "two past a quarter after 8, that's 8:17." (It might not seem that way at first -- but that's because we spent long hours learning to tell time.)

2. For the exactly opposite view, that the number system should be base-12, see http://www.dozenal.org/ . It's actually a good way to work with numbers, and I've used the "counting on the joints/pads of your proper fingers" trick a number of times; sometimes you only have one hand free and want to count to something that's less than 48 (you can encode 2 extra bits with "hand facing up, hand facing down, hand facing up again, hand facing down again"; I've found it gets confusing after 4 or 5 of these though.)

3. I am very sympathetic to Feynman's "we don't need more units!" claim, but the reason we use various units is because we have different interests -- masses in eV/c^2 for example reflect someone who is interested in the atomic interaction energies (eV) of relativistic particles (c^2); energies in Kelvin reflect someone who is interested in how much they need to cool their experimental apparatus to see certain effects; energies in inverse centimeters reflect people who have spectrometers. Following this, I've tried to think whether the base-10 clock could be used to construct a set of "rational units" which would try to get the "human scale things" right while making all of these other units amount to a power-of-10 difference. I've not condensed these speculations to a final form yet but the speculations are themselves at: https://github.com/drostie/essay-seeds/blob/master/misc/rati... .

[ctdonath]

"we don't need more units!"

Yes, some units are useful for particular interests. Create them as they're needed and useful. Base-10 time would make more sense. Celsius at least coordinates notable values with common materials.

Some units are just stupid. The lead article gives an unsatisfactory explanation of base-12/60 time (admitting near the end the reason for base-60 is unknown); trying to explain clocks to toddlers is proving annoying (I can't explain it if it doesn't make sense, and it doesn't make sense). Fahrenheit is just an arbitrary marking on a scale and seeing how reality happened to line up.

http://xkcd.com/927/

[dragontamer]

Fahrenheit was supposed to be 0 degrees for freezing, and 100 for body temperature.

It turns out that Fahrenheit's reference thermometer was a bit... off... however.

[gumby]

Interestingly, Fahrenheit has 180 (212-32) degrees between freezing and boiling. Zero is set to a brine solution -- a reproducible metric about as cold as you could make. So it's not an insane system.

[MarkPNeyer]

0 was the coldest temperature he could reliably reproduce - which is salt water just above the freezing point.

[pyre]

How did it manage to be off by 32 (!) for freezing, but fairly close for body temperature? I would expect a constant (or at least same ballpark) error.

I imagine it had something to do with the medium used for the thermometer?

[stephenhuey]

He was trying to find a salt mixture that went through a phase change at 0 degrees:

http://antoine.frostburg.edu/chem/senese/101/solutions/faq/z...

That wasn't so reliable, and neither was using 100 for human body temperature, so when he discovered that water boiled at 212 he began calibrating with that.

[Crito]

> We already know that when someone says "be there in five minutes" they mean 7-8; if they said "be there in five ki" they would be more accurate.

But then they would actually show up in 7-8 ki, meaning 10-12 minutes!

On the other hand, maybe this is just the cure for high-intensity fast-paced society that we need. ;)

[qbrass]

What about scaling instants up to a year? A kiloday would be 2.74 years. Which doesn't seem to mean anything in human terms.

[dragontamer]

All of the english system measurements were built on "human" terms.

A furlong is the distance a horse can plow in one day. This also gives a relationship between horsepower and furlongs. A chain is the distance you keep crops apart, and one chain by one furlong is an acre.

If you had 20 acres of land, you knew a horse could plow the land in 20 days.

The aging English system of Feet, Yards, Furlongs, Miles, Rods, Chains, Acres, Horsepower... they don't mean anything to non-farmers. So, we can replace these with a different system: the Metric system.

However, everyone uses the Point / Pica (1/72th an inch, and 1/6th an inch respectively) system for font measurement. As we are computer literate people today, font-size measured in "points" is quite important.

12-Point font (aka, font one pica in size) is a standard font size. No one cares that 12-points is really 4.233 mm. Hell, saying 4.233mm font means nothing to me.

But the day, night, moon, and sun cycles (approximated by day, month and year) is obvious to everyone. As such, an arbitrary 10-base system like Metric Time ignores the "reality" for most people... that the day, moon, and sun are related.

[Someone]

Nitpick: horsepower is not an old English unit; James Watt invented it in the late 18th century (http://en.wikipedia.org/wiki/Horsepower)

[Crito]

I doubt most people understand or think of "Point" as a measure of length. It is just some completely arbitrary scalar that exists only due to poor UI design and so that teachers can mark off points for it not being set to "12". In many other cases, such as modern web browsers, it is being replaced with the concept of "zoom" which just uses natural numbers with "100" being arbitrarily defined as some loose concept of "default".

They would certainly never dream of using it for anything other than choosing how large they wanted their font, or possibly line.

[drostie]

The basic answer is that (1 year) / (1 day) = 365.242199... is a pure number with no units associated to it, so there exists no set of units which can eliminate it.

The question then is, should we synchronize on the year, or on the day? The advantage of using the year is that it's a more stable unit of time; the length of a day is actually incredibly noisy and days are slowly getting longer every century as the Sun's and Moon's tidal forces deform the Earth, turning rotational energy into heat.

Unfortunately, there are a bunch of disadvantages. There are a lot of calendars in effect today, and they don't always settle on the same definition of "year" -- for that matter, there are a lot of astronomical definitions of "year" -- since the Earth's axis doesn't point at the same stars eternally, do you mean orbiting the Sun once relative to the distant stars (ignoring the axis) or do you mean coming back to the same tilt relative to the Sun so we can start the seasons over again (including the axis)? Or do you just mean a full cycle of phases of the moon, as lunar calendars do it?

The best way that I can see is to settle approximately on a day with an atomic clock; and push the question of actually trying to keep dawn at the same time each day (correcting for the slowdown of the Earth) to the time zones, which already sometimes try that in the sense of Daylight Saving Time.