This is not just exposing the existing Thanks feature to anonymous senders. It is a proposal for a new feature that lets you thank for an article, without identifying a particular revision. Then, some group of editors is informed.

Did someone work on this project during Wikimedia-Hackathon-2015? If so, please update the task with the results. If not, please remove the label.

I suggest declining as too similar to the [[mw:Article feedback]] fiasco. The correct solution is to make users use action=history and expand thanking permissions to unregistered users.

The rationale of https://www.mediawiki.org/wiki/Author_Thanks points more to a generally thankful "Like/Star this page" than to our Thanks framework, which is designed around interpersonal thanks.

Seeing that a page has been liked/starred by N users might be useful. I would keep it only for registered users, to reduce the risk of abuse (imagine school Wikipedia workshops or assignments where students compete to get the highest number), showing the values for everybody (which might be a way to encouraging readers to register, and from there...)

Liking / starring has become a popular action in the Internet, and there is a chance that this would work in Wikimedia projects too. The Article Feedback problem was the extra work it brought to editors, but this feature would not affect them negatively in any way. Editors active in page X will feel thanked if they see that 4.000 people like the page.

risk of abuse [...] this feature would not affect them negatively in any way

Contradictory statements.

In T89577#1426332, @Nemo_bis wrote:

risk of abuse [...] this feature would not affect them negatively in any way

Contradictory statements.

I don't think so. Even if a page gets 1999 non-legit thanks, how does that affect negatively the editors? Requiring user registration reduces the risk of fake numbers, that's all.

In T89577#1422345, @Nemo_bis wrote:

I suggest declining as too similar to the [[mw:Article feedback]] fiasco. The correct solution is to make users use action=history and expand thanking permissions to unregistered users.

This is in no way similar to article feedback. Article feedback created a backlog of work for editors to do (and the backlog wasn't as clean and easily actionable as it could have been).

This does not create a backlog of work.

IMHO it's very simple. "Feedback" is meaningful if it integrates with the wiki process. "Thanking" some abstract entity for the existence of an article, without understanding the history of the page, its authors, the stratification of contributions, is the opposite of understanding and integrating with the wiki process. Therefore, it can only generate noise and harm.

P.s.: I'm changing summary because readers don't exist and there are no features restricted to unregistered users.

Readers do exist, and your edit does not serve to clarify the purpose of the bug.

It does, because reader thanking of editors is already possible, by visiting the history. (I've further clarified by quoting the relevant snippet of the mediawiki.org page for this proposal.)

In T89577#1465721, @Nemo_bis wrote:

"Feedback" is meaningful if it integrates with the wiki process. "Thanking" some abstract entity for the existence of an article, without understanding the history of the page, its authors, the stratification of contributions, is the opposite of understanding and integrating with the wiki process. Therefore, it can only generate noise and harm.

I disagree. "Thanking" might be more meaningful when you are aware of all the details behind a result. Still, if you like the result and it is useful to you, saying Thank You is still meaningful for a lot of people. It's like saying Please, You Are Welcome, or smiling: it makes most people feel better in both ends of an exchange.

Let's take https://en.wikipedia.org/wiki/The_Harlem_Tulip as an example (literally a "Random article"). Looking at the topic and the history, I'm sure those contributors would be likely happier if they saw that NN users thanked them for that page. And probably the NN users than thanked them felt good performing that action.

Those NN thanks would not create any scientifically meaningful information [1], but would help establishing a richer social bonding between readers, editors, and the content they are both interested about.

Anyway, it looks like the opinions of a sample of actual readers and editors would be useful here. I wonder if a small survey or some quick study could be made to bring this discussion further with some data.

[1] Although with enough time and scale those Thanks might entertain researchers.

So a reader comes along and reads a page and things 'dang this is good stuff', and wants to make sure people know that they liked it. Maybe this alerts any of the pages editors that someone liked it, but the editors would be able to disassociate from such things, or disable the alert at all. And there would be no 'this editor has been thanked for these pages' data, just 'this page has been thanked this many times. Maybe with expiry due to significant edits since? Such that a user could thank again later?

This idea is completely separated from the mediawiki Thanks extension as said in T89577#1269819, hence removing the tag and adding MediaWiki-extension-requests.