Reading order is not a ranking

Why position reads as meaning

Readers are trained by everything else they use. The first result is the best match. The top item is the recommendation. The first name in a list is the one that matters. That training is strong enough that a flat list is read as a ranked one unless something actively says otherwise.

A list of addresses is one of the few places where a reader would be better off with the training switched off.

Three things a top entry might be

Arbitrary

The arrangement in which the strings arrived, preserved because reordering them would be an editorial act. This is what the arrangement on this site is.

Sorted by something irrelevant

Alphabetical, or by length, or by whatever order a file happened to be in. Looks meaningful, means nothing.

Sorted by something the publisher measured

Response time from one machine at one moment, or a popularity count. Real information about something, but not about which address you should use.

None of the three helps you

The first two obviously carry nothing. The third is more interesting, because it is real information that still does not answer your question. A measurement taken from somebody else's machine tells you about the path between that machine and the service, and your path is different.

The study on speed as a tiebreak takes that apart in full. The short version is that a route which was quick from somewhere else is not a route that will be quick from here.

The arrangement on this site

The three strings printed at the top of this page sit in the arrangement they were supplied in. Nothing here sorted them, scored them or moved them. There is no first, no principal and no fallback, and no page on this site attaches a number or a label to any of them.

That is a deliberate rule and it is stated on every page so that a reader is never left inferring one from the layout.

Why the rule is worth keeping even when it is inconvenient

It would be more helpful, in a shallow way, to say try this one first. It would also be a claim, and the claim would be unsupported by anything, because this reference never opens the addresses it prints.

A page that ranks without information trains the reader to trust ranking on pages that also have no information. The habit travels, and the next page it is applied to may have ordered its list for a reason of its own.

What to read instead of position

• The count. A page offering more strings than the market publishes is telling you something about itself, which is the argument of a list longer than the set.
• Whether the page says where its strings came from. Provenance is checkable in a way that quality is not.
• Whether the page claims a check it could not have performed. That is the tell described in what a list proves.

The one honest use of position

Position is useful for one thing: finding the same entry again. If you are comparing two sources line by line, a stable arrangement makes the comparison quicker. That is a property of the page, not of the addresses, and it is the only work the arrangement should be doing.

A note on why this site states its arrangement

Most pages leave the meaning of their ordering unstated, which leaves each reader to infer one. Stating it costs a sentence and removes the inference entirely.

It also creates an obligation. Having said the arrangement is the one the strings arrived in, this site cannot quietly reorder them later to put a favourite first, because the claim is on the page. Small commitments of that kind are most of what separates a reference from a page that happens to contain some facts.

What this study is not. It is not a criticism of any page that sorts its list, and a sorted list is only misleading when a reader reads the sort as advice.