> The vast majority of these articles were posted by people who have a well-deserved reputation for a high level of accuracy. The most extreme case is perhaps that of Henry Spencer, who makes errors so rarely that in the space newsgroups it was an event, involving the awarding of a (virtual) "I corrected Henry Spencer" T-shirt, when anyone else found an error of his.

Imagine an Internet where accuracy was valued. Are there forums around now which are like that? What does it take?

> Imagine an Internet where accuracy was valued.

LOL even in the "good old days" Usenet was absolutely chock-full of assholes and cranks, the above archive is just the _good_ parts.

    'even in the "good old days" Usenet was absolutely chock-full of assholes and cranks'

Yes, but that's what kill files were for :-)

EDIT: Formatting.

I'm looking forward to looking at what he thought were my good posts!

It's not that rare for technical topics. People like Henry Spencer participated in technical newsgroups, and yarchive looks to be very technical.

This forum is good for such topics, and so are many subreddits. If you want accuracy about other stuff like politics, good luck :) It's worse now but it was still bad in the 90's.

> This forum is good for such topics, and so are many subreddits.

IME, in my fields of expertise (in IT), this forum is bad - as in, if you don't already have expertise, you shouldn't trust what you read; most of it is wrong or bad advice. Subreddits are worse, a waste of time (of course, maybe I don't know about good subreddits because I hardly ever read them).

Gatekeeping against some minimum level of effort and technical ability to connect, and a culture that celebrates accuracy and the scientific method and honest discourse.

I pointed out a radio astronomy mistake in one of Henry's postings, and no surprise, he wanted to quibble about it. He's not a radio astronomer.

Why do you think that forum is accurate? How do they do it?

It takes a firm tone from the beginning that weeds out and does not tolerate fools.

Once fools have taken over > 50% of the messages and start writing in an authoritative manner, all is lost. See the python-dev mailing list after the woke revolution for example.

I was a big reader of the rec.bicycles forums in the 1990s. That contributor Jobst Brandt was amazing in his reasoning on bicycles and their technology. Reading this brings back great memories.

I did not agree with him all the time, but he profoundly changed my views on endless debates at the time such as the effect of weight on the tradeoff of durability versus speed.

If you've ever used the network interface InfiniPath -> TrueScale -> Omni-Path, the spark that lit the fire was an argument I had on comp.arch with Nick Maclaren, where I asserted that an appropriate endpoint design could minimize MPI overhead. I later sold a big supercomputer with this chip to Nick's organization.

comp.arch was a real gem. I learned a lot from reading Mashey's posts.

Is there a proper usenet archive search online any more?