[physarum_salad]

"Wikipedia, in my mind, has two main purposes: A quick visit to find out the basics about some city or person or plant or whatever, or a deep-dive to find out what we really know about genetic linkages to autism or Bach’s relationship with Frederick the Great or whatever."

Completely agree with the first purpose but would never use wikipedia for the second purpose. Its only good at basics and cannot handle complex information well.

[ajross]

I think that's actually wrong, or hangs on a semantic argument about "complexity". Wikipedia is an overview source. It's not going to give you "all" the information, but it's absolutely going to tell you what information there is. And in particular where there's significant argument or controversy, or multiple hypotheses, Wikipedia is going to be arguably the best source[1] for reflecting the state of discourse.

Like, if there's a subject about which you aren't personally an expert, and you have the choice between reading a single review paper you found on Google or the Wikipedia page, which are you going to choose?

[1] In fact, talk pages are often ground zero!

[physarum_salad]

The best source is the one that provides the widest breadth of information on a topic.

This is a good use of wikipedia: "Like, if there's a subject about which you aren't personally an expert, and you have the choice between reading a single review paper you found on Google or the Wikipedia page, which are you going to choose?"

But that is like skim reading or basic introductions rather than in-depth understanding.

[ajross]

> that is like skim reading or basic introductions

No? How do you learn stuff you don't know? Are you really telling me you enroll in a graduate course or buy a textbook for everyone one?

Like, can you give an example of a "deep dive" research project of yours that does not begin with an encyclopedia-style treatment? And then, maybe, check the Wikipedia page to see if it's actually worse than whatever you picked?

Again, true domain experts are going to read domain journals and consult their peers in the domain for access to deep information.[1] But until you get there, you need somewhere you can go that you know is a good starting point. And arguments that that place is somehow not https://wikipedia.org/ seem... well, strained beyond credibility.

[1] Though even then domains are really broad these days and people tend to use Wikipedia even for their day jobs. Lord knows I do.

[generationP]

Yeah, encyclopedias are meant to be indexes to knowledge, not repositories thereof. The WP feature-creeped its way to the latter, but it is not reliably good at it, and I'm not sure if there is an easy way to tell how good a given page is without knowing the subject in the first place.

[skeeter2020]

what I think it IS good at is parlaying the first purpose into a broad, meandering journey of the basics. I would never use it for deep study of genetics & autism or Bach and Fredrick the Great, but I love following some shallow thread that travels across all of them.

[dragonwriter]

Its often good for the latter when, as a tertiary source should be, it is used not just for its narrative content but for its references to secondary sources, which are themselves used for both their content and their references.

[spankibalt]

> Its only good at basics and cannot handle complex information well.

Poppycock! Because of MediaWiki's multimedia capabilities it can handle complex information just fine, obviously much better than printed predecessors. What you mean is a Wiki's focus, which can take the form of a generalized or universal encylopedia (e. g. Wikipedia), or a specialized one, or a free-form one (Wikipedia, in practice, again). Wikipedias even negotiate integration of different information streams, e. g. up-to-date news-like information, both in the lemmata (often a huge problem, i. e. "newstickeritis"), in its own news wiki (Wikinews), or the English Wikipedia's newspaper, The Signpost.

And to take care of another utterly bizarre comment: Encylopedias are always, per defintion, also repositories of knowledge.

[physarum_salad]

Don't understand the implications of this:

"And to take care of another utterly bizarre comment: Encylopedias are always, per defintion, also repositories of knowledge."

We should just accept wholescale editing and knowledge production when we personally agree with it? Otherwise its verboten? You are aware of "edit-a-thons"? Are these biased?

Why not just have an AI print all of the currently available information on a topic with minimised (not zero) biases?

If we lived in a utopia where wikipedia could randomly allocate tasks to a diverse group of expert level civilians and then aggregate these takes/edits into a full description of a topic I would agree with wikipedia maximalists. This does not happen and a bunch of bad or naive actors have reduced the quality.