Why the set is three and not fifty

Length looks like generosity and is not

A page offering forty strings feels more useful than a page offering three. More options, more chances, more coverage. The feeling survives about a minute of thought and no longer.

A set is only as good as a reader can hold it. Three strings can be stored, compared and checked. Forty cannot be held by anybody, which means in practice a reader keeps whichever ones happen to be on top, and the length of the list has done nothing except make it feel comprehensive.

Long sets cannot be checked

A short set can be compared against another source line by line in a few seconds. A long set cannot be, by anyone, which is precisely why long sets are easy to publish and hard to argue with.

Anybody can put forty onion strings on a page and call it a mirror index. The cost of adding a bad entry to a long list is nearly zero and the chance of it being noticed is close to it. Length is therefore a place where a bad string is cheap to hide, and a short published set removes that hiding place.

What the operator gets from a small set

Each address is a service that has to be kept running. Three is a number an operator can keep straight. Beyond some small number, entries stop being maintained, and an unmaintained address is worse than no address: it is a string that used to be right, which is the most convincing thing a wrong string can be.

So a small set is not stinginess. It is the size at which the set can still be true.

What the reader gets from a small set

• You can hold the whole thing, which means the benefit described in what redundancy buys is actually collectable.
• You can notice when a list you are shown has more entries than the set does, which is a signal that costs nothing to read.
• You can compare two sources quickly enough that you will actually do it.
• You can tell how much of the set has gone quiet, which is the whole basis of the absence section.

Where the diminishing returns set in

The jump from one address to two is enormous, because it is the jump from no next move to a next move. The jump from two to three is smaller and still meaningful, because it lets a reader distinguish between one quiet address and a general problem.

After three, the additional information is thin. A fourth route mostly repeats what the third already told you, while adding another string to hold and another line to check. Small sets sit where they do for good reasons rather than by accident.

Reading the number as a signal

When you meet a page listing WTN market mirrors, the count is one of the few things you can evaluate before you evaluate anything else. A page showing a handful is making a claim it could plausibly keep. A page showing dozens is making a claim nobody could keep, including itself.

That is not proof of bad intent, and plenty of long lists are simply old ones that nobody pruned. It is a reason to treat the length as a question rather than a feature.

The number as something to remember rather than read

Knowing the size of the set is a small piece of knowledge with an unusual property: it is easy to hold and it is the only part of the subject that lets a reader evaluate a page before reading it.

A reader who knows the set is small will notice a page offering many. A reader who has never counted has nothing to notice with, and will read a long list as a generous one. That is the whole practical value of paying attention to the number.

What this study is not. It is not a rule that any long list is dishonest, and this study does not assess any particular list published anywhere.