Guest post: User story on ratings and edits
This is intended to be an example of the activities that the encyclosphere standards and protocols should support. It follows on from user story 1, which introduces “sciepedia.org”, a hypothetical online science encyclopedia.
Kevin is a regular user of sciepedia.org, and has contributed to a number of articles on the subject of high-energy physics.
He sees a tweet on Twitter about a new paper on kaons and reads the paper. He thinks there is information that is worth adding to the kaon page on sciepedia
He logs into the sciepedia.org website with his credentials and reads the latest “official” page — this is the page chosen by the site which has the best ratings from users recognised as members of the sciepedia community. As an editor, the site also shows him an information box that says there are two known edits to this version of the page that do not have ratings.
He pulls up the first suggested edit, showing the differences. It is advertising copy for a pharmaceutical site. He pulls down a rating menu and flags it as spam. This will hide it except for a page where users can review ratings. It will also penalise other edits from the same author ID.
The next edit adds a line of detail about the error bounds of the charge. Kevin doesn’t know about that. He checks that the metadata on the edit makes sense, and he selects the rating menu and rates the revision as well-formed. He’s not vouching for accuracy, just that it is not obvious rubbish or misdescribed. The revision will not be preferred over the current one, but it will still be offered to other editors for review.
He then goes to the editing screen for the page and adds a brief paragraph and a link describing the paper. The new revision he creates keeps the same metadata as the previous revision, but adds an entry referencing the previous revision, and indicating that he is the author of the most recent edit. He saves his revision, which results in it being added to sciepedia’s feed of new changes.
All of Kevin’s ratings, and his revision to the kaon article, are signed by the site with Kevin’s sciepedia username and the sciepedia current signing key. This is OK, but Kevin prefers to have his contributions associated with a key he controls. He therefore next runs a signing app on his desktop. It pulls from sciepedia.org (and snowboardingworldboard.com, which he also is a member of, and which uses the same encyclosphere rating and signing protocols) the feed of his contributions. He checks they are correct, then clicks “select all” and “sign contributions”. It prompts him for a password and then signs the ratings with his own personal key. It then sends those contributions to a sciepedia API endpoint as additional signatures.
His revision to the kaon article doesn’t immediately replace the current revision, but if one other sciepedia member in good standing rates it as verified then it will. All the policies concerning what ratings and signatures make one revision preferred to another are up to sciepedia, and would be different for a different online encyclopedia.