exampleListCuration.md

Curated Lists example: decentralized curation of a list of nostr clients

The point of this example is to illustrate the genuinely decentralized nature of the dcosl protocol, which is capable of curating lists without putting any single entity in charge of the curation.

In this example, a list of nostr clients has been submitted by an anonymous user and curated by your grapevine. (Anyone can submit a new list, so you can imagine this to be any list you want.) In addition to submission of the list itself, there are three categories of nostr notes that are relevant to this list:

• list item submission. Anyone can contribute an item to any list.
• curator endorsement. Anyone can endorse any other user (thumbs up or down) as a curator of this list.
• list item endorsement. Anyone can endorse any submitted list item (thumbs up or down) as an item that belongs (or doesn't) to this list.

From the above data, each submitted list item is partitioned into the ACCEPTED, the REJECTED or the PENDING (not enough info yet) category. But if Alice and Bob don't "trust" the same people to curate this list, they may not see the same results!

In the panels below, the "seed user" selector determines whose perspective is being depicted (whose grapevine is being used for the curation). Results are shown below from the perspectives of two different users:

• wds4, a regular user
• joker4, a putative bad actor promoting a scam client created with nefarious intent

Note that, in accordance with DIP-01, the results of curation depend upon the choice of seed user.

LEFT PANEL: From the perspective of wds4, "scammy fake client" is rejected from the list, with several legit clients accepted onto the list and several others awaiting curation by the grapevine. RIGHT PANEL: From the perspective of the joker, "scammy fake client" is the only one accepted onto the list, with one legit client having been rejected.

Why do they see different results? How are the lists sorted? How does wds4 know that joker4 is a bad actor? Short answer: wds4 endorses someone who endorses someone who rejects joker4 as curator for this list. Perhaps multiple such paths exist from wds4 to joker4. So it is not too hard to understand, intuitively, that wds4 the regular user and joker4 the bad actor would see different results. Users do not have to dig into the calculational details if they don't want to. Like an automobile, you don't have to know the details of how it works to use it. The take home message for this page is simply that different users do not necessarily see the same results. (Although it is postulated by the doctrine of loose consensus that oftentimes they will see the same results, in the absence of some controversy or irreconcilable reason for disagreement, like that one is a bad actor and the other is not.) Ultimately, you are at the apex of the decision making tree; no one is ultimately in charge of this process other than YOU. In this way, and for this reason, there can be no single point of failure. It is truly decentralized.

But for users who are even a little bit curious about how it works under the hood, or who doubt the conclusion of the previous paragraph, they will be able to dig in all they want. See this page for an overview of how list items are accepted and rejected.