Zariia does not measure anything, and it does not find out what happens to the records made with it. That is a decision, and this page gives the reasoning.
Most things made in public are made twice. Once by the person making them, and again by the count that comes back — the plays, the opens, the reach, the room's size on the night. The second making is quiet and it is usually the stronger of the two, because it arrives afterwards and claims to be evidence.
Zariia does not run the second one.
Zariia may know what it did. It never knows what happened.
What it did is the format, the simulator and the player. What happens afterwards belongs to somebody else, and there is no path back from it: the simulator runs in the browser of whoever uses it, so Zariia does not even know when a record has been made. No analytics, no logs, no counts, no reach. Nothing here is instrumented, and nothing is going to be.
The law comes from a text written thirteen years before Zariia. The history page carries it in full.
“Intensity and access are inversely related, the more the access the less the intensity and the less the access the more the intensity.”
“We do not keep server logs, so there is no way of knowing how many people visit the website. If you do not know a number it can be anything. It can be zero and it can be infinite.”
A record with a sealed channel holds both settings on one object: high access for the open track, low access for the sealed one. The record does not have to choose between them.
None of it is being withheld. Zariia does not know these things, and it has built nothing that would let it find them out later.
What it cannot claim: this site sits behind a content delivery network, and that network sees requests the way any network between two machines does. Zariia does not read that, does not export it and does not act on it, but it is somebody else's machinery and saying otherwise would be a stronger claim than the arrangement supports.
Because a number, once you have it, replaces what was there before it. An unknown number can be anything at all, and it stays that way only until somebody checks. Checking does not add a figure to what was there; it puts one where the possibility was.
And because the count answers a question that was never the right one. Work is made for somebody. If you cannot connect with something, that most likely means it was not meant for you, and that is an ordinary result. A record that reaches four people has not underperformed. It may have been made for four people.
So the decisions get made forwards, by whoever makes a record: what it holds, who it is for, and who is given the key. However an earlier record was received is no help with any of those.
These are the most relevant voices working in this area.
Curating runs on that sentence and it is not a finding. Nobody has surveyed the field. There is no register of who is working in an area and no measure of relevance, so there is nothing a reader could check it against.
Worse, it does not describe. Said from a platform, it hands out the relevance it claims to be reporting, and then it comes true. And it turns an address book into a survey: what a curator knows is who they have met, whose work reached them, and who is good at writing about themselves. That describes where a curator is standing, which is not the same as describing a field.
What Zariia can say is what it did. Here is the work, and here is the name of whoever made it.
A practice that never finds out how it landed cannot notice itself going wrong. If a change makes the simulator worse, nothing in this arrangement will say so.
The only defence is to get things right in advance and then keep to them: the format is written down, the code is open, and the tests run before anything is published. Here they carry more weight than they would somewhere that could correct itself on evidence.
The arrangement buys intensity and pays for it in not knowing. There is no version of this where only the first half arrives.