aboutsummaryrefslogtreecommitdiffstats
path: root/tools/feed-wheel-reverse-test.js
Commit message (Collapse)AuthorAgeFilesLines
* The wheel coasts wherever the pointer is while a feed or a thread is up.Adam Soltys10 days1-7/+12
| | | | | | | | | The coast only took a notch whose target sat inside the feed or the thread, so a pointer resting on the rail, the header or the wallet column beside the feed on a wide screen got the browser's plain scroll (Adam, 2026-10-16: "when my cursor is hovered over the right side bar or any part of the page that's not the feed, wheeling scrolls without coasting. can we make it coast no matter where my pointer is?"). Now the question is whether a feed or a thread is the page on screen; the pointer's spot only matters when it is over something that takes the wheel itself — a text box, the chips strip, the lightbox, the pager, a box that scrolls on its own — and on a page with no feed on it (the wallet) the wheel stays the browser's. feed-wheel-reverse-test: the fixture's page carries a feed unless told otherwise; a notch with the pointer over the wallet column, or over nothing in particular, coasts; one on a page without a feed does not. Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_01J1Kon8bxJoHRpJdSzshCkp
* A turn back against a flying feed brakes first: that notch eases the coast ↵Adam Soltys10 days1-33/+31
| | | | | | | | | | | to a stop, the next is the gentle start back, the ones after wind up. The gentle half push straight away from a flight still overshot (Adam, 2026-10-16: "the first click of the wheel should just stop the coasting/flying maybe with a quick easing rather than a sudden stop and then the next notch should start going slowly in reverse and the next ones should start speeding up"). Now, when the page is flying (over 1.2 px/ms — a notch from rest peaks near 0.9), the first notch back only brakes: the coast's decay is shortened to 40 ms from that moment (the frames a busy thread missed don't travel first), so a flight at 8 px/ms stops within ~300 px, eased. The notch after it, up to 2.5 s later, is the gentle start — half a notch's travel at half the speed — and the ones after wind up as usual. Against a stroll or a coast that only just died there is nothing worth braking: the turn back is the gentle start itself, as before. feed-wheel-reverse-test: the brake moves nothing by itself and eases out within a short way, never the other way; the next notch creeps back half a notch; the ones after wind up; a notch back a second and a half later is still gentle, three seconds later a full push; a stroll turned back starts gently at once. Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_01J1Kon8bxJoHRpJdSzshCkp
* The wheel coasts over a video and on a thread, and a turn back against a ↵Adam Soltys10 days1-19/+38
| | | | | | | | | | | | | coast is a gentler half push that winds up again from the next notch. A notch with the pointer over a clip fell to the browser's single-notch scroll: video and audio were on the wheel handler's hands-off list from the first commit, though the wheel does nothing to a player itself. The thread page was outside the handler's scope altogether, so a thread scrolled plain while the feed coasted. Both coast now, like the words around them (Adam, 2026-10-16). A reversal was a full notch the other way followed by two notches without wind-up, and felt like an overshoot. It is now half a notch's travel at half the speed, so the reader who flew past a post settles back onto it; the next notch the same way is a plain one, winding up from whatever speed the half push left, so a spin on after a reversal gets going at once. (The five-notch quarter-size hunt stays gone.) feed-wheel-reverse-test: the half push, the wind-up from the second notch, a notch over a video / audio / a thread coasts, one in a text box or the pager does not; the fixture's target can sit inside a given element and on a thread page. Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_01J1Kon8bxJoHRpJdSzshCkp
* Feed wheel: after a reversal the next two notches are plain, then a spin ↵Adam Soltys11 days1-1/+10
| | | | | | | winds up again. The 700 ms spell without wind-up after a turn back (63fc25d) also covered a spin begun inside it, so correcting the direction and spinning on felt slow for most of a second (Adam, 2026-10-16: "it slows down for a while when I don't want it to"; "overall the scrolling still isn't as fast"). The calm is now a count: the reversal's next two notches don't wind the speed up — a couple of correcting notches still travel a couple of notches — and from the third the wind-up is back, so a spin after a reversal goes as far as one from rest. tools/feed-wheel-reverse-test.js (a reversal plus four quick notches goes well past five plain ones, and about as far as five from rest). Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_01J1Kon8bxJoHRpJdSzshCkp
* Feed wheel: a turn back is a plain notch again, with no wind-up for a ↵Adam Soltys12 days1-18/+14
| | | | | | | moment. The "hunt" on a reversal (3ff588d: a quarter-size notch settling fast, then four growing ones) read as a stutter on every change of direction and a slow patch after it (Adam, 2026-10-16: "it stutters and sometimes when I'm scrolling it slows down for a while"). Now a reversal still catches the coast at once, and the notch travels its full distance at the usual pace; for 700 ms after it notches don't wind the speed up, so a couple of correcting notches travel a couple of notches rather than launching a flight the other way — which was the complaint the hunt had answered. tools/feed-wheel-reverse-test.js (full travel on a reversal, a reversal plus a quick notch vs two wound-up ones from rest, a reversal against a coast that only just died); feed-momentum-test passes. Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_01J1Kon8bxJoHRpJdSzshCkp
* Feed wheel: a turn back against the coast hunts, in short steps that grow. ↵Adam Soltys2026-10-121-3/+22
| | | | | | Flying down the feed and turning the wheel back for the post just passed used to send the page flying the other way (a full notch's coast, 300px) the moment it caught — the reader overshoots again, and hunts by overshooting. Now a reversal against a coast (or one that died within 400 ms) catches the page and takes that notch as a quarter of its travel, over a quick coast (110 ms, not 325); each notch on in the same direction grows the step — 0.35, 0.5, 0.7, 0.85 of a notch — and the sixth is a full push with the usual wind-up, so a few flicks have the wheel flying again. Another turn back while hunting starts the small steps over, and two seconds without a notch ends the hunt: a notch from rest is the push it has always been. Keys, a touch or a grab end it too. tools/feed-wheel-reverse-test.js (short step on reversal, the ramp, the restart, the pause, the just-died coast, the full push from rest); feed-momentum-test still passes. Co-Authored-By: Claude Fable 5.1 <[email protected]>
* Restart wheel coast timing and respond immediately on reversalAdam Soltys2026-10-091-0/+50