Leo

On the cup of tea: please do not, and the reason is worse than sticky keys.

The key block is directly above the key frame and the action. A spill there does not sit on the surface — it runs down into the mechanism, into felt and cloth and wooden parts that swell. Technicians will tell you that liquid in the action is one of the more expensive accidents an instrument can suffer.

A piano is one of the few pieces of furniture where the flat surface is genuinely not a surface.

One practical test anybody can run, and it is more persuasive than any argument: take a recording you know well and shift it by that amount both ways in any audio editor.

Listen without knowing which is which, ideally with somebody else operating it. Most people can hear a difference between the two. Almost nobody can reliably say which one is the "natural" one when they are not told, and the descriptions people offer flip depending on what they were told beforehand.

That asymmetry — a real audible difference, no consistent preference once expectation is removed — is the whole answer in one experiment.

One more thing that often turns out to be the actual cause: check the fingering at that junction.

A transition that fails consistently, in the same way, in an otherwise easy piece, very often has a fingering problem — the hand arrives in a position from which the next notes are awkward, and no amount of practice fixes an arrangement that requires the hand to be in two places.

Work backwards from the note you must reach and see what hand position it needs, then choose the fingering in the previous bar that puts you there. Sometimes the fix is one changed finger three notes earlier, and the difficulty evaporates entirely.

One practical note for anyone trying it: on a classical guitar the effect is much stronger above about the fifth fret and nearly useless in the first position.

That is straightforward from the mechanism — near the nut the string is short and already at high tension relative to the displacement you can produce, so rolling achieves very little. Players who feel that their vibrato does not work often have been trying it in exactly the place where the physics is least helpful.

The useful principle first: a mixed crowd sings in two octaves automatically, and you are not choosing a range for everybody, you are choosing one range that works when men sing it an octave below the women.

So the practical target is a melody that sits comfortably somewhere around the middle for one group, and the other group will find its own octave without being asked.

The range that most reliably works for untrained singers spans roughly from the A or B flat below middle C up to about the C or D above it. Almost everybody can manage that octave. Push the top up much beyond D and you start losing people — you will hear the room thin out on the high notes, and once people fail on a phrase they tend to stop singing altogether rather than dropping out for one note.

Your D to D is at the upper end of workable. Taking it down a tone or two would bring in more of the room.