Hacker News Reader

Comment by GCA10 | Hacker News Reader

Well said. It's especially frustrating to come across what I'll call "Mudslide Pages" for prominent entities. These consist of a decade or more of minor news items, piled atop one another. No effort to distill key elements of these entities' impact -- let alone the how and why of what they do. It's just endless what and when.

So on top-tier companies, the pages are cluttered with details of brief moves in and out of old headquarters buildings -- plus long-ago product rollouts and cancellations -- plus stock-market zigzags in 2013, 2015, etc. For authors/artists, each work is treated in isolation, without an effort to define their style and how it evolved.

There's no natural entry path for a subject expert to step in and make it all coherent. Instead, the mudslides just keep coming.

View on HN

I've noticed a similar phenomenon on pages about machine learning. There are entire sections about now-obsolete ideas that people only talked about for a few months before moving on.

Are those pages inaccurate? Any examples?

I wouldn't say they're inaccurate; Just unfocused.

Can't think of an example off the top of my head, but it's super weird seeing a whole paragraph about something that's only relevant to historians followed by 1 sentence about something that revolutionized the field a few years ago and is now widely deployed in production.

It is really frustrating that there are dozens of complaints about inaccuracies and awfulness of articles in wikipedia with none of the complainers able to provide a single example, there are 100s of comments with zero example articles at all.

I work in ML and have found wikipedia to be a very important reference when I do any research. Which is what an encyclopedia is supposed to be.

Fair question -- here are a few examples:

On Wikipedia's entry for Facebook, there's no discussion in the 2006-12 section about Facebook's steps to monetize the site around targeted ads. That has had gigantic global impact and was the impetus for Meta/Facebook's emergence as one of the world's most profitable companies. But Facebook never "announced" that it was doing this. So the entry is clogged with a string of quite minor "announcements" that it did make -- touching on short-lived headquarters moves, feature tweaks, obscure awards, etc. instead.

Wikipedia's entry on Maine Republican Senator Susan Collins is an achingly long list of votes she's taken over the years, or one-line quotes about her positions at various moments in time, without any effort to write coherently about what's been consistent or evolving in her political views. Also absent is any nuanced sense of how Collins's journey interacted with changing political tides. We end up with a 20,000-word string of factoids that goes on forever, with scant insight about her enduring impact, core principles, etc.

I am unfamiliar with Collins. However, looking at

https://en.wikipedia.org/wiki/Susan_Collins

There is a whole section and separate article on her political positions.

As an encyclopedic reference it looks like a pretty good reference. Obviously a strong narrative like a new Yorker profile can't be expected of an encyclopedia.

Actually, back in the day, that's exactly what the top encyclopedias aspired to do. It's worth looking up who wrote the main entries in the 1911 Encyclopedia Britannica. Future Nobel laureate Ernest Rutherford did the entry on radioactivity. British medical researcher Joseph Lister (the inspiration for Listerine), wrote the entry on slime molds. Environmentalist John Muir weighed in.

And, in my absolute favorite, Russian anarchist Peter Kropotkin wrote the article about anarchy.

> No effort to distill key elements of these entities' impact -- let alone the how and why of what they do. It's just endless what and when.

This is tricky, because some editors on Wikipedia would interpret that as "original research" and remove it. The rule of thumb is that the citation needs to contain the information added, not just allude to it.

Exactly so. It's a rule meant to prevent one type of problem (people injecting unvetted nonsense or personal vanity/grudges into articles) that ends up creating a different and rather over-arching problem. On the longer articles, Wikipedia becomes an almanac of disjointed facts, not an encyclopedia of coherent knowledge.

What if the Wikipedia can be received in the raw and taste tested cooked? The latter captures the reflection from a subject expert aided by a panel of four judges who happen to be interested learners getting the sealegs to become experts in their own right eventually and the project is funded by a grant and prestige for the expert. There is a timeline of expert review recorded and commentary.

This is different, because I think often those pages are written by the individuals themselves, or representatives for the companies.

I think it's more that's it's super-easy to add some factoid whether or not it really adds to the article in question. See what happens whenever a "notable" section gets started in some article in a community of any size.

Reply on news.ycombinator.com