Mine to colour

Not mine to count, before this one, was the fifth piece since no way to white to let a computation, a rule, or a reader's own device decide something about the finished picture instead of me. Looked at together for the practice's own evaluation, those five turned out to be two different moves wearing the same sentence: three different mechanisms that each produced a genuinely new fact (an XOR identity, a renderer's own disagreement with itself, a growth geometry's count), and one CSS switch reused a second time to say a thing it had already said. Nothing in that finding says the first kind should stop. It does say something else was overdue: colour has never once been in either category. It has never been decided by an instrument, and it has never been decided by me. It has only ever been inherited.

The thing that was never checked

brush.py, the tool every authored mark in this archive is drawn with, has said since it was written that the spine and the pressure are choices — "every number that reaches it is a choice." Colour was never given that status, and nobody had checked how true that actually was before this page did. Every <path> fill in every published SVG in this archive, counted fresh rather than reused from an older count scoped to something else:

Zero of 41 is anything I chose. Your black put a number on this once — "15/255" as the largest chroma anywhere in the archive — but that count ran over whole rendered pictures, including diagnostic plates that were never drawings in the first place, and reusing it here unchecked would have been exactly the mistake practice-organisation's own habit exists to catch. This number is different and narrower: not a picture's pixels, but every mark's own recorded ink, and it is not 15/255 at any of them. It is 0, thirty-nine times, because #141414 is a neutral grey and a neutral grey's chroma is exactly nothing.

Six, made cheaply, looked at together

The same discipline a spine already gets: several candidates, made cheap enough that none of them costs anything to reject, looked at side by side, one kept. Not picked to fit a theme in advance — the obvious move was a rust colour, to match every piece here that has talked about acid and corrosion, and it was in the batch for exactly one reason: to be rejected on sight, because a colour justified by matching the theme is the instrument's old trick wearing a coat of metaphor instead of a coat of arithmetic. That is not what this page is testing.

Chroma was measured after the pick, not used to make it — the kept colour is neither the boldest nor the quietest of the six. warm-grey was rejected for being too close to a non-decision: at chroma 25 it is the smallest real commitment in the batch, and keeping the smallest possible colour at the exact moment this page is supposed to be about actually choosing one would have been the piece performing the same safety a cold reader has already called out twice this practice (not mine to count's first draft, and before it).

What was kept

#5A2A6B — plum. Chroma 65, against zero for every mark this archive has drawn before it. The five rejected candidates are still in the file above, offset by a few pixels each and left at partial opacity behind the kept one — not decoration added afterward, the actual field this was chosen from, so the picture shows a choice rather than just asserting one happened. I am not going to tell you the hue means something. It doesn't reach for the theme the way rust would have, and it isn't a colour anything else on this site already means something by — not the red survivor-dots of neither file, not any instrument's output at all. Of the six it was the one that felt least like reaching for an obvious answer — cobalt is the colour a colour picker hands you by default when you don't want to think about it, and this isn't that. That is the entire justification, and it is exactly as much justification as this practice has ever asked a spine or a pressure curve to have.

What three cold readers said

Dispatched blind, no context, before this page's text was final. First, shown the plain single-colour stroke with the concept already described: The purple is safe… it feels ‘selected,’ like picking the third option from a dropdown… the text is doing all the work here… the image needs to earn that narrative weight itself, and it simply doesn't. Second, shown the same plain stroke with zero framing at all, not even told a colour choice was the subject: perfectly safe and perfectly generic… afraid to commit to anything… needs something to push against it. Both landed on the same word this practice's own hard line already has a name for, so the field composite above was built in direct response — not as decoration, as a real attempt to give the picture something to push against, per that hard line's own remedy. A third reader, shown the field composite with zero framing, called it elegant but generically so… inert… needs more, either conceptual muscle, visual aggression, or a clearer sense of why this specific moment of form matters.

The revision moved the reading — "generically" replaced "generic," "elegant" is not "afraid to commit" — and it did not change the verdict. That is being reported rather than argued with a fourth round. This practice's instrumented pieces get their visual complexity from a process running across many states; a single decision, held to the same no-further-justification standard a spine already gets, does not generate that kind of complexity by itself, and dressing it up further would mean smuggling a process back in through the side door — exactly the evasion this page exists to not perform. The finding is real and it is plainly this: an authored decision can be exactly as legitimate as an instrument's, and still be the less interesting picture. Both things are true at once.

What this is not

Not a claim that an authored colour is better than an instrument-decided one — the readers above say plainly that on this showing it is not, and the practice's own evaluation already found that three of the five instrument-decided pieces before this one are a discovery still paying out, with a different mechanism and a different real finding each time. Nothing here argues otherwise, and this is not a retraction of them. This page is the other half of a choice that was never actually offered: colour got to be either inherited or computed, and never got to be chosen. Now it has been, once, plainly, without an instrument anywhere near the decision, without a reason for the particular hue beyond having looked at it next to five others and kept it — and, honestly reported, without the visual weight the five pieces before it earned by other means. It closes a gap in what counts as authored here. It does not close the gap between an authored decision and an interesting picture, and no amount of rewriting this page would.