The picture: an arrow from a chosen corner of the room to the ball. Look at figure s01 — the arrow r and its three shadow-lengths x,y,z on the walls.
Why the topic needs it: everything downstream is "where is the particle, and how fast is that arrow changing?" We cannot talk about energy until we can point at position.
The subscript in ra just labels which particle: r1 is particle 1, r2 particle 2, and so on up to rN for N particles. The bold typeface always means "this is an arrow (a vector), not a single number".
Why a dot and not d/dt? Pure shorthand — in mechanics almost every derivative is with respect to time, so Newton's dot saves ink. r¨ (two dots) would be acceleration, but the parent topic never needs it.
The picture: stand the particle at one instant, then a heartbeat later; the tiny arrow connecting the two positions, divided by that tiny time, is r˙. It points along the direction of travel and its length is the speed.
The special case we use most is an arrow dotted with itself:
v⋅v=v12+v22+v32=∣v∣2=v2,
the length-squared, i.e. speed-squared. That is why kinetic energy is written T=21mr˙⋅r˙ — it is 21mv2 dressed in vector clothes.
The picture: figure s02 shows two arrows and the two things the dot product measures — how much one lies along the other, and (when equal) the plain squared length.
The picture: every ball carries a little energy bag proportional to its mass times its speed-squared; T is all the bags emptied into one pile. Nothing here is mysterious — it is Newtonian energy. The whole art of the parent note is rewriting the r˙a using better coordinates.
Why n and not 3N? A single free particle needs 3 numbers; N of them need 3N. But constraints (a bead trapped on a wire, a pendulum of fixed length) forbid most of those. The number n that actually survives is often much smaller. See Generalized coordinates and constraints.
The picture: figure s03 — a bead on a bent wire. Its full room-position needs x,y, but along the wire one number q (arc length) says everything. That one number is a generalized coordinate.
The indexi in qi just labels which generalized coordinate we mean; q˙i (with the dot) is that coordinate's rate of change — the "speed along that natural direction".
The extra slot t (time) appears only when a constraint is itself being moved on a schedule we impose from outside — a wire someone is spinning, a support being raised. Such a constraint is called rheonomic (moving). If nothing external moves the constraint, t is absent and the constraint is scleronomic (rigid, time-frozen).
The picture: two dials (q and, if present, a clock t) feeding into a machine that spits out the room-arrow ra.
The straight-d derivative d/dt changes everything at once; the curly-∂ derivative changes one thing, holding the rest fixed. We need the curly one because our map has several inputs.
Two flavours matter:
∂qi∂ra — motion caused by you turning coordinate i.
∂t∂ra — motion caused by the constraint moving on its own (only nonzero for rheonomic systems).
Why the topic needs both: the velocity of the particle is the sum of "motion I command through the qi" plus "motion the moving constraint forces on me". That split is literally the parent note's Step 1.
Read it as a sentence: total room-velocity= (for each dial: "which way it pushes me" × "how fast I'm turning it") summed up, + ("which way the moving constraint drags me").
When we square the chain-rule velocity, products q˙iq˙j appear for every pair(i,j). Writing ∑i,j means "sum over all ordered pairs i and j". Each such term carries a coefficient
Mij=∑ama∂qi∂ra⋅∂qj∂ra,
a table of numbers (a matrix) indexed by two labels.
Mij is symmetric (Mij=Mji, because the dot product doesn't care about order) and positive-definite (real motion always costs positive energy).
Each arrow means "you must understand the tail before the head makes sense." The two streams — the energy stream (left) and the coordinate stream (right) — meet at the topic itself.