Why is SQL so successful?

A while back someone on lobste.rs asked:

Why is SQL so successful despite its shortcomings?

To which I answered the following.

Human and machine efficiency:

1. Human efficiency: If I learn e.g. MongoDB, FaunaDB, or Datomic’s proprietary languages, and they go out of business, I have literally wasted my time. If I learn SQL, I have an industry-standard skill I can re-use, probably for decades. Multiply that across thousands of people working with “data stuff” across the entire world.
2. Machine efficiency: We have decades of research papers and industrial experience in optimizing SQL query performance. Some one-off DB with its own query language may or may not be able to take advantage of all that has been learned over the past 40+ years about SQL query optimization. The fact that someone has invented their own query language does not mean they don’t have this knowledge, but it’s a signal that they may not. (Which feeds back into point 1: I need to think your DB is worth spending my time learning, so you need to signal to me in various ways that you know what you are doing. SQL is one of those signals, since it is data’s lingua franca).

Tubes, by Andrew Blum (2012)

This is a good book. I’d recommend it to anyone interested in the internet’s physical infrastructure. Since it was published over 10 years ago now, details have probably changed, but the basic information is probably still broadly right. Or at least, of interest

If you’re going to read this, you should probably also read the essay Mother Earth, Mother Board by Neal Stephenson (from 1996 or so). It’s easy to find online.

Hachette v Internet Archive re: “Controlled Digital Lending”

If you’re a proponent of Controlled Digital Lending by libraries, this court ruling is worth reading:

Memorandum & Opinion – #188 in Hachette Book Group, Inc. v. Internet Archive (S.D.N.Y., 1:20-cv-04160) – CourtListener.com courtlistener.com

I think the Internet Archive didn’t help the cause when they went beyond a strict 1:1 correspondence. Perhaps they were trying to provoke the suit tho

According to other laws cited in the opinion it doesn’t sound like even a 1:1 ratio would have mattered. But it does appear that going way beyond that ratio is what prompted the suit back in 2020. Again, perhaps intended.

I’d like to see CDL become the law of the land but the opinion cites a lot of fair use case law that makes it seem like CDL via lawsuits isn’t gonna get it done. I think Congress would have to pass legislation, which seems unlikely

See also https://controlleddigitallending.org/ which states

“Through CDL, libraries use technical controls to ensure a consistent “owned-to-loaned” ratio, meaning the library circulates the exact number of copies of a specific title it owns, regardless of format, putting controls in place to prevent users from redistributing or copying the digitized version”

The internet Archive did not appear to have such controls in place with partner libraries, nor did they maintain the 1:1 ratio for a while in 2020, so it’s hard to see how what they were doing was really “controlled digital lending” according to the definition

Why approximately everyone hates math

Thesis: math “sucks” because the algebra nerds won and rewrote everything using algebra and made everyone use it. My probability book doesn’t mention tree diagrams until p. 60 and then says “oh btw this is what the old guys who started the field used in 1700s” like wtffffff

It’s not hard to see why P(A|B) != P(B|A) if you use pictures but if you start right in plugging and chugging algebra formulas they’re easy to mix up (one example)

Big Algebra got their beloved “Generality ™” and now math is hated by people who could actually enjoy it

Why I think everything after R5RS was a mistake

Today in “things nobody cares about”, I’m kinda mad at the R6RS authors for creating their own DEFINE-RECORD-TYPE that is apparently different from SRFI-9? because… something something PL nerd bullshit and now I can’t re-use my code

Feeling more and more like everything after R5RS was a mistake. Like sorry Scheme lost but I don’t think PL nerds breaking my code to write new semantics or whatever is gonna make fetch happen for the modal ALGOL programmer in industry anyway

On the recent public layoff emails

The public layoff email is one of the most important communications an exec can make. That most so far are poorly written rambling jargonfests that are somehow also self aggrandizing (signs they clearly didn’t get a professional edit) is telling IMO

Like you can’t formulate a coherent 3 paragraph email amongst your entire team with all your resources but tell me again how you are good at strategy and execution

How school attendance worked in the 90s

In 1995/96 I skipped school one day a week. You could miss up to 30 days of school a year in NY and still pass. In 1997 they changed it to 18 days

Why old car analog dashboard controls are better

This is me adjusting the volume with a finger by feel. There are detents. Push to turn on/off. I can do this safely while driving.

Car designers, please study the classics. iPad on the dashboard ain’t it. This 2008 Subaru design defeats all touchscreens

Why I don’t like the new big pickup trucks

My critique of the new big pickups is primarily aesthetic. They look like a 4 year old boy’s idea of a “big man truck”. They represent a new low point in the development of faux masculine objects. Fundamentally insecure. “2 generations of boys with no dads at home” vibe

My parents divorced when I was young so I’m sympathetic but only to a point

On issue tracker consistency

To paraphrase Alan Perlis, it’s better to have 1000 people at your company using 1 issue tracker than 10 groups of 100 people with each group using a different issue tracker