Has anybody produced Dunning-Kruger curves for people who think they know what Dunning and Kruger did?

The lesson Andrew learned was

Do not trust documentation blindly; it could be wrong.

which is ok. Another, more important lesson to learn is

Program defensively, make sure every case is handled. (The Elements of Programming Style, Kernighan and Plauger, McGraw-Hill, 1978).

A third, more obscure lesson available for learning is

Sometimes talking to the duck isn't enough, and you should talk to something that will talk back.

Andrew Tannenbaum, Lessons Learned from 30 Years of MINIX, Comm. ACM, March, 2016:

I mentioned [...] MINIX running on the simulator but not on the hardware to my student, [...] who said [...] the 8088 generated interrupt 15 when it got hot. [...] there was nothing in the 8088 documentation about that [...] I inserted code to catch interrupt 15. Within an hour I saw this message on the screen: "Hi. I am interrupt 15. [...]."

What lesson did Andrew learn? What lessons should he have learned?

Is the "dictatorship of the proletariat" an oxymoron?

Here's a thought: sentences like "Catalogs contain articles, so we introduce a class CITEM." should never appear in any formalization of anything. Avoid pointlessly multiplying concepts by using either "Catalogs contain items, so we introduce a class CITEM." or "Catalogs contain articles, so we introduce a class ARTICLE.". (Decoupling suggests the former, avoiding jargon suggests the latter.)

Mr. Perfect was the idol of our manager, Nelson, and he almost never came in without a shockingly gorgeous girl, rarely the same one twice. They would sit down at the bar, Mr. Perfect and the girl, and the predictable theatrics would start right up, so the moment he appeared I'd resign myself to a night of watching a wallet flirt with a price tag.

Deborah Eisenberg, Cross Off and Move On
Your Duck is My Duck

Hear true stories of American health care:


Start listening at 29:12. For maximum irony load, listen from the segment start at 28:54.

My answer is the difference is the same as the difference between long-term trends and short-term fluctuations. Permafrost melts due to year-after-year temperature increases and means possibly significant change. Freakish weather is a happenstance due to particular circumstances and doesn't mean much past its own existence.

There is an answer in the thread [1], but it isn't helpful: it doesn't address the question, it wanders all over the place, and it's hard to understand (to be fair, I suppose, the first sentence does sort of point towards what I think is the right answer).

[1] slatestarcodex.com/2019/06/26/

A blog item discusses melting permafrost and existential dispair. [1] Someone in the comments is just asking questions:

"Also, how is the report of melting permafrost different from a conservative saying that a particularly cold winter disproves the climate change?" [2]

What is your answer?

[1] slatestarcodex.com/2019/06/26/
[2] slatestarcodex.com/2019/06/26/

r. clayton boosted

NGC 1566: The Spanish Dancer Spiral Galaxy

Image Credit: NASA, ESA, Hubble; Processing & Copyright: Leo Shatz

apod.nasa.gov/apod/ap190702.ht #APoD

Sentences that make you go hmm:

Brian Vanderhall had been a friend since college, probably my best friend through graduate school and all the drama surrounding our careers at the New Jersey Super Collider.

David Walton, Superposition

nah: People good at making flutes signalled reproductive fitness and successfully spread their genes.

yeah: People good at sex signalled reproductive fitness and successfully spread their genes. Everybody else had time to make flutes or whatever.

A UDS advantage should be better performance (but, as always, benchmark). A UDS disadvantage is intramachine communication only (but maybe not a problem for IPC).

Mastodon @ SDF

"I appreciate SDF but it's a general-purpose server and the name doesn't make it obvious that it's about art." - Eugen Rochko