Hacker News: mikequinlan

New comment by mikequinlan in "The Atari 800 – By Paul Lefebvre"

mikequinlan — Tue, 12 May 2026 20:07:56 +0000

It was owned by the University across the street (Colorado State University). The mother of one of my friends was a Systems Programmer. We had a blast learning all about the machines.

New comment by mikequinlan in "The Atari 800 – By Paul Lefebvre"

mikequinlan — Sun, 10 May 2026 17:42:54 +0000

The Atari 800 was my first personal computer (i.e. one I had at home). I bought one when they first came out and learned 6502 assembler language and had a blast.

I had experience with assembler language on larger computers (DECSystem 20 and CDC 6400 in high school/, IBM System 360/44 in college then IBM System 370/145 at work) but having my own personal computer was incredible. Of course I moved on to the IBM PC when it first came out (and learned 8086 assembler language), and many PC compatible computers after that, until I retired and got my first iMAC (Intel x32 and x64), then now to my M1 Macbook Pro.

New comment by mikequinlan in "Ask HN: What Is the "Lore" of HN?"

mikequinlan — Mon, 27 Apr 2026 22:27:25 +0000

https://github.com/minimaxir/hacker-news-undocumented/blob/m...

New comment by mikequinlan in "XOR'ing a register with itself is the idiom for zeroing it out. Why not sub?"

mikequinlan — Wed, 22 Apr 2026 07:50:31 +0000

As TFA says, on x86 `sub eax, eax` encodes to the same number of bytes and executes in the same number of cycles.

New comment by mikequinlan in "Ask ChatGPT to pick a number from 1-10000, it generally selects from 7200-7500"

mikequinlan — Sat, 21 Mar 2026 07:14:26 +0000

The prompt doesn't say to pick a random number. I asked to pick a number from 1-1000 and it chose 7,381. Then I asked why it picked that number and it said

Nothing mystical, I’m afraid. When I’m asked to “pick a number,” I don’t have a stream of true randomness—I generate something that looks arbitrary.

In this case, I leaned toward:

• something comfortably away from the edges (not near 1 or 10,000),

• not a round or patterned number (so, not 7,000 or 7,777),

• and with a bit of internal irregularity (7-3-8-2 has no obvious rhythm).

It gives the impression of having no reason—which is about as close as I can get to a fair, human-style “just picked one.”

New comment by mikequinlan in "ChatGPT's porn rollout raises concerns over safety and ethics"

mikequinlan — Sun, 25 Jan 2026 23:22:11 +0000

https://archive.ph/QLyKT

New comment by mikequinlan in "The Genius Whose Simple Invention Saved Us from Shame at the Gas Station"

mikequinlan — Sat, 03 Jan 2026 02:46:30 +0000

The arrow that shows you which side of the car the gas cap is on.

New comment by mikequinlan in "Why do AI images look right but fail in production?"

mikequinlan — Tue, 23 Dec 2025 19:46:36 +0000

Do you have a better link

New comment by mikequinlan in "Why do AI images look right but fail in production?"

mikequinlan — Tue, 23 Dec 2025 18:17:33 +0000

I don't see anything related to your title at that site.

Maybe you meant this

https://archive.is/pIM27

But it is from 2018.

New comment by mikequinlan in "Apple to focus on 'quality and underlying performance' with iOS 27 next year"

mikequinlan — Sun, 23 Nov 2025 18:24:11 +0000

I sure hope that is true.

New comment by mikequinlan in "Ask HN: Talk to your users. What is your opinion on an Online Farmers Market?"

mikequinlan — Wed, 22 Oct 2025 00:04:37 +0000

How is this different from the local Reko markets?

New comment by mikequinlan in "Solution to CIA’s Kryptos sculpture is found in Smithsonian vault"

mikequinlan — Thu, 16 Oct 2025 11:57:21 +0000

>Jim Sanborn planned to auction off the solution to Kryptos, the puzzle he sculpted for the intelligence agency’s headquarters. Two fans of the work then discovered the solution.

Gift link https://www.nytimes.com/2025/10/16/science/kryptos-cia-solut...

New comment by mikequinlan in "Ask HN: Is Fortran the first high-level language?"

mikequinlan — Sun, 28 Sep 2025 07:53:43 +0000

Plankalkül - https://en.wikipedia.org/wiki/Plankalk%C3%BCl was designed but as far as I know never implemented (until 2000).

ENIAC Short Code - https://en.wikipedia.org/wiki/Short_Code_(computer_language) was implemented but very little used.

New comment by mikequinlan in "Any thoughts about not accepting cards as payment to avoid chargebacks?"

mikequinlan — Sun, 28 Sep 2025 03:05:36 +0000

As a customer, if I can't do a chargeback in the case where you don't provide the advertised services, I simply won't do business with you and will go to one of your competitors.

New comment by mikequinlan in "Ask HN: Should I upgrade to tahoe (macOS 26) or not?"

mikequinlan — Sat, 20 Sep 2025 08:24:14 +0000

Certainly wait until 26.0.1 or whatever the first bugfix update is named.

New comment by mikequinlan in "Beyond Meat headed to Chapter 11 bankruptcy"

mikequinlan — Sun, 17 Aug 2025 21:45:40 +0000

Beyond Meat Denies Filing For Bankruptcy In New Statement

https://plantbasednews.org/news/economics/beyond-meat-denies...

New comment by mikequinlan in "Why English doesn't use accents"

mikequinlan — Sun, 06 Jul 2025 22:17:14 +0000

Clearly the early scribes were looking forward to the 7-bit ASCII code and needed to reduce the number of characters that were represented.

New comment by mikequinlan in "Ask HN: What product should the new Commodore release?"

mikequinlan — Fri, 04 Jul 2025 23:35:43 +0000

The original Commodore was known for a super cheap computer that made it practical for people to have one at home, as well as innovative graphics for the time.

I don't know how that might translate into modern times when everyone has a home computer (a telephone at least). Maybe a computer targeted to pre-teens?

New comment by mikequinlan in "Tesla's Cybertruck flop is historic. The brand collapse is even worse"

mikequinlan — Fri, 04 Jul 2025 22:29:11 +0000

The hate is because Elon is literally a Nazi. And no, Nazi's are not part of my "tribe".

New comment by mikequinlan in "[dead]"

mikequinlan — Tue, 01 Jul 2025 23:45:08 +0000

Updated July 1: Microsoft has since updated its blog post, correcting the number from 1 billion to 1.4 billion and confirming Windows hasn't lost millions of users.