PMA Publisher: Published from the Principia Machina source

2026-10-15T19:30:41Z

Published from the Principia Machina source

← Older revision		Revision as of 19:30, 15 October 2026
Line 1:		Line 1:
	{{DISPLAYTITLE:What happens at the edges}}		{{DISPLAYTITLE:What happens at the edges}}
			{{Noether's theorems map}}

	''Mathematical counterpart: [[Noether's theorems/First theorem/The boundary term\|The boundary term]].''		''Mathematical counterpart: [[Noether's theorems/First theorem/The boundary term\|The boundary term]].''
Line 107:		Line 108:
	''No mathematics is used on this page. The mathematical treatment is on the		''No mathematics is used on this page. The mathematical treatment is on the
	counterpart page named above.''		counterpart page named above.''

BFP Rauch: 1 revision imported

2026-09-10T13:29:55Z

1 revision imported

← Older revision		Revision as of 13:29, 10 September 2026
(No difference)

NT>NoetherVolumeBuild: Noether theorems volume, initial import (B-06)

2026-09-10T00:00:00Z

Noether theorems volume, initial import (B-06)

New page

{{DISPLAYTITLE:What happens at the edges}}

''Mathematical counterpart: [[Noether's theorems/First theorem/The boundary term|The boundary term]].''

[[File:NT-F-14.png|thumb|right|360px|alt=A solid rectangle and a shifted dashed copy of it, with the strip that enters shaded on one side and the strip that leaves shaded on the other. The overlapping interior is marked unchanged. A caption says the whole cost of moving the region is a border term, and that this term becomes the conserved current.|The origin of the boundary term]]

This is the hinge of the whole argument, and it turns on being careful about a
question that sounds trivial: changed compared with what?

== Two thermometers ==

A river is warming through the afternoon, and it is warmer downstream than up.

Drop a floating thermometer in the water. It drifts with the current and records
a rising temperature. Some of that rise is the afternoon warming everything. Some
of it is the thermometer arriving somewhere that was already warmer.

Now fix a second thermometer to the bridge. It also records a rise, but only the
afternoon's. It never goes anywhere, so nothing it records is due to arriving
somewhere new.

'''Two thermometers, two different numbers, both correct.''' Which one you want
depends on the question, and confusing them is a mistake that shows up as a wrong
sign rather than as obvious nonsense.

== Noether keeps three, not two ==

Most treatments of her theorems use two symbols where she used three. The middle
one is the one that disappears.

'''The drifting thermometer.''' The total change: the new situation at the new
place, against the old situation at the old place. Everything that changed, from
every cause.

'''The bridge thermometer.''' The change at a fixed place: the new situation
compared with the old at the same spot. This is the total change with the effect
of moving subtracted out.

'''The nudge.''' The ordinary change you make when you test a shape by pushing
it, as on
[[Noether's theorems/Explainers/Nudging a path and seeing what changes|the nudging
page]]. Here nothing is being transformed at all — you are just poking the system
to see if it pushes back.

The first two differ by exactly the amount the drift accounts for. Noether writes
that difference down as its own equation, and it is one of the most quoted lines
in the paper. Compress the three into two and you lose the equation that
subtracts the drift, which is precisely where sign errors come from.

== Moving the frame ==

There is a second piece of bookkeeping on this page, and it produces the thing
everything else is built on.

Photograph a scene. Now slide the frame slightly and photograph again. The two
pictures differ, and the difference is entirely at the border: a strip has come in
on one side, a strip has gone out on the other. The middle is untouched.

Transforming the variables moves the frame in just this way, and the cost of
moving it is exactly a border term.

'''That border term is where conserved quantities come from.''' It is not an
error term to be assumed away. It is the answer.

== Losing the integral sign ==

The last step on this page is short and it changes everything after it.

You end up with a statement that some total, added up over a region, is zero.
That would be a mild statement if it were about one region. But it holds for
'''every''' region you choose — big, small, anywhere.

A quantity whose total is zero over every possible region has to be zero
everywhere. There is nowhere for it to hide.

So the statement stops being about totals and becomes a statement about each
point separately. From here to the end of both theorems there is no adding up at
all, only algebra.

That is why the results hold point by point, and why they hold whether or not
anything is obeying the laws of physics.

== Reading the wrong thermometer ==

Here is the failure this page exists to prevent.

Suppose you are working out the conserved quantity for a shift in time, and you
reach for the drifting thermometer when the argument wanted the bridge one.

The two differ by the drift term. So your answer differs from the right answer by
that same term. And the drift term is not small, and it is not obviously out of
place — it has the right units, it has a sensible size, it varies the way you would
expect a piece of the energy to vary.

What you get is not an error message. It is a wrong conserved quantity that looks
entirely reasonable, and which happens to be conserved in the special case where
nothing is moving.

'''The most common symptom is a sign.''' The two comparisons differ by the drift,
and the drift enters with a minus, so a treatment that muddles them tends to
produce an answer that is right except for being backwards. A minus sign is easy
to blame on a convention and hard to trace to its cause.

That is why Noether kept three symbols where two would have looked tidier, and why
this volume does the same.

''No mathematics is used on this page. The mathematical treatment is on the
counterpart page named above.''

Noether's theorems/Explainers/What happens at the edges - Revision history

PMA Publisher: Published from the Principia Machina source

BFP Rauch: 1 revision imported

NT>NoetherVolumeBuild: Noether theorems volume, initial import (B-06)