The clock numbers, and which of them anybody can check

Eight quantities in this subject are measured in time. One of them falls out of a piece of software and can be checked by anyone. The other seven are patterns, and each page says so in its first line rather than dressing a guess as a rule.

Mars market addresses

Published, not monitored
mars24pas2vgwtr4drrsy7tlngevbvxyguejynnkeywibjzenet7knqd.onion
marsautbk3di5cj75eh4dakjjrngddnjwqfdltbq2sy6cf7unzkd2bad.onion
marsautkudspgk6j23cxdtrk36ae4fpis2eoe7izu5y2rsksvmfji2ad.onion

These three addresses are printed as given. This site does not test them, does not know whether any of them answers right now, and publishes no availability figure. An address that opens is not the same thing as an address that is genuine, which is what the forty character fingerprint is for.

The eight time numbers

Time is where this subject gets vague. People say delivery is quick, escrow releases eventually, and a dispute takes a while. None of that is a quantity. The pages below take the eight figures that come up most often and give each one a value, an assumption and a limit.

The register runs by size, from a three day parcel to a twenty four hour pause. Read the chip on the right before you read the number. Counted means the value drops out of a specification. Judged means somebody chose an assumption, and the page prints the assumption where you can argue with it.

How much of this section is guesswork

Most of it. Seven of the eight numbers here describe settings an operator picks and does not publish. This site has no account with the market, no contact with anybody who runs it, and no way to read a configuration file. Where a page gives a figure for a dispute window or an automatic release, that figure is the shape these settings usually take, not a policy anybody confirmed.

A reader who quotes one of these as market policy will eventually be wrong, so the page says it before they do it rather than in a footnote afterwards.

The alternative is worse. A site that prints a confident seven days with no source has removed the one thing that would let you check it. You cannot tell whether the author read it somewhere, guessed it, or copied it from another site that guessed it. A number carrying its assumption can be tested against your own screen. A number carrying nothing can only be believed or ignored.

NumberKindWhat it rests on
10 minutesCountedA documented default in Tor, readable by anyone
3 to 21 daysJudgedHow ordinary post behaves over distance
14 daysJudgedThe usual span of an escrow release window
7 daysJudgedThe usual life of a dispute once it is open
5 triesJudgedWhere login limits tend to sit
20 minutesJudgedWhere idle timeouts tend to sit
4 minutesJudgedA timing of one task, done once
24 hoursJudgedA rule this site made up on purpose

Reading a clock number without getting burned

Three habits make these useful. Treat a judged figure as a range with the middle marked rather than a deadline. Find out what your own order screen shows, because a countdown you can see beats any figure published anywhere, including this one. And act early inside a window instead of late, since every clock in this section punishes whoever waits.

Two pages matter more than the rest. The automatic release runs whether you are watching or not, which makes it the only number here that acts on you. The twenty four hour pause is the only number this site invented, and the only one that costs nothing to follow. The ten minute circuit is worth reading once so that ordinary network behaviour stops looking like a market failure.

Where the real number livesNo page in this section is a measurement of this market. If you need a figure you can rely on, read it off your own order screen. Where the screen and this site disagree, the screen is right and this site is background.

Questions people ask

Why not ask the market what these settings are?

There is nobody here to ask. This site has no relationship with the market and publishes only what it can derive or reason about, which is why seven of the eight numbers carry an assumption line instead of a source.

Which time number causes the most trouble?

The automatic release at fourteen days. It is the only clock in the section that completes on its own, so a reader who does nothing still gets the result it produces.

Is anything here measured?

No. Nothing is timed, sampled or probed. The counted number comes from a published default, as the counted page explains, and the rest are reasoned in the open.