DamnInteresting

One recalls the Pentium FDIV bug, where an estimated 1 in 9 billion floating point divides with random parameters would produce inaccurate results[1]. A small subset of Pentium users would ever encounter the problem, yet it was considered a major controversy, and Intel ultimately replaced $475 million worth of CPUs (in 1994 dollars)[2].

[1] https://en.wikipedia.org/wiki/Pentium_FDIV_bug

[2] https://www.tomshardware.com/pc-components/cpus/its-been-30-...

Most of the time it's also necessary to put quotation marks or similar formatting (e.g., box, pullquote, or indentation) on the cited text to make it clear a) that it is a citation; and b) where the citation begins and ends.

If one does that, assuming citations are not the bulk of the work, one should be in the clear.

> I prefer to toggle the dots to avoid seeing them when I’m driving the car.

Please do not look at your phone screen while driving the car.

Unlike many other apple varieties these days, SweeTango are only available in traditional "apple season" (Autumn months through November in the USA). I've long been a fan of HoneyCrisp, so I tried SweeTango, and while it was quite good, I still prefer HoneyCrisp. I had a similar experience with Cosmic Crisp--not as good as HoneyCrisp, but nice for some variety.

I've started working on a new browser-based word game[1], it'll be my third so far. I'm also sketching out the next one after that. I didn't know this kind of thing would be habit-forming.

[1] I'd link to my first two, but I'm not a fan of self-promotion.

Video overview of the technology: https://www.youtube.com/watch?v=X8yRlA7jqEQ

Ordinarily I don't prefer video, but the visuals are helpful here.

Also, an online interactive, but it seems to only work in Chrome: https://superspl.at/scene/ff1d0393

All my life I've gone by my middle name, and it's turned out to be useful in a similar way. When I receive a message or phone call and they address me by my first name, I immediately know it's someone who doesn't really know who I am.

I received the exact same email, except they offered me 50% of earnings, so the message sending script probably has some randomization built in. My guess is that their true objective is to get you to install their "interview support app" which is mentioned at the end of the email, I anticipate that it makes your device remotely useful to them, or installs ransomware. But it could be a more involved scam.

No, money down![1]

[1] https://old.reddit.com/r/TheSimpsons/comments/102sh8/one_of_...

I've had my writings directly plagiarized by other people--people who made word-for-word copies of my work, replaced my name with their own, and made a tidy profit on it. They profit more than I ever managed, because they have more resources. In the aftermath, my writings are still "there", not "stolen" in the physical world sense, but my ability to make a living is damaged, and the plagiarism is deeply unethical.

LLMs and "AI" are just one small step removed from straight-up plagiarism. They are massive moral injury[1] machines.

[1] https://moralinjuryproject.syr.edu/about-moral-injury/

It was so long ago that the specifics have faded, but I remember I was coached to use a variety of positive responses. "That's a great question," yes, but also things like "I'm glad you brought that up," and "I was hoping someone would ask about that!" It wasn't my cup of tea, too artificial, but the advice was contemplated.

The next question (which is a great one, from what I understand) is: Why do LLMs use these phrases so much if humans rarely use them in written form? Maybe a fair portion of training data comes from lecture transcripts, where such responses are common when responding to direct questions? And/or system prompts are just instructed to be like that?

That's a great question! Ahem.

I remember being advised to do this ~20 years ago when I was going to be answering questions from a group of people. I was told that it's good practice to say something like "that's a great question" every time someone asks anything, as a form of social lubrication, to encourage others to ask questions. I can't say whether it works, and it was advice for a spoken context rather than written, but I don't know how to finish this sentence.

Thanks for the kind words!

I've been running websites on WordPress almost as long as WordPress has existed, my oldest site has been running on it continuously for 20+ years[1]. Between forcing Gutenberg into the core (it should have been a plugin), Mullenweg's recent unhinged actions (e.g., the ACF fiasco[2]), the increasing number of naggy subscription-based plugins, and now this contemptible AI integration, I'm going to start the process of migrating my sites, probably to ClassicPress[3].

Oh well, we had a pretty good run.

[1] https://www.damninteresting.com/

[2] https://news.ycombinator.com/item?id=41828958

[3] https://www.classicpress.net/

I wrote an article about this same general topic way back in 2007[1], wherein I conducted a thought experiment with an Arecibo observatory traveling away from Earth. I calculated that even mighty Arecibo would be unable to detect Earth's omnidirectional FM radio as far out as Saturn, let alone from another star.

Since that writing, we've lost Arecibo observatory, discovered gobs of exoplanets, started scrutinizing those exoplanets with JWST, and increased our radio sphere radius by another 19 lightyears.

[1] https://www.damninteresting.com/space-radio-more-static-less...

> I don't really understand the folks that released their software under open source license and are now upset that LLMs are training on it

The key word there is "license." Open source often has strings attached--an obligation to credit the source, an obligation to release derivative code under the same license, etc. LLMs seldom respect the license, they just quietly and extensively plagiarize everything.

> You know, I took a second look at it, and weirdly enough, I can't vouch it for some reason.

The 'vouch' option doesn't appear until/unless the post becomes [dead].

I'm happy she's enjoying the game! If you hit the Menu button on the top right, you should see a "reset puzzle" button in the lower right quadrant of the menu.

I'm working on the next update of Omiword[1], an ongoing daily word game previously discussed on HN[2]. I'm building an alternate stand-alone app version with access to all of the archived puzzles. It's slow going since it's just one of many side-projects I like to work on, but that's the tinkerer's dilemma.

Yes, I know that [insert LLM here] could do a lot of that conversion for me in mere minutes. No thank you. I'm doing it, in part, for the doing.

[1] https://omiword.com

[2] https://news.ycombinator.com/item?id=43654350

I love this short story, it's one whose memory visits me unbidden from time to time. I blogged about it over 20 years ago[1], and it was already around 15 years old at that time. OMNI magazine was great.

[1] https://www.damninteresting.com/retired/short-fiction-made-o...

Back in my freelancer days, I learned this lesson the hard way, and I became strict about payments (half in advance, the other half required to turn over the final work). But when someone I'd known for years had an emergency, and needed their software patched over a weekend, I made an exception and delivered the work in good faith. And then they tried to get out of paying for it. I had to harangue them for months to get the check.

The lesson I learned is that it doesn't matter how long they've been your friend, when it comes time to pay, they may still try to stiff you. The value of your service diminishes infinitely after it has been performed.

Malus is a homophone of "malice." But then again "Claude" is a homophone of "clod."

> A user that does only traditional searches (not using Assistant) will cost more on our end than someone using AI + search.

Unless I am missing something, that appears to be a mathematically impossible claim. They are saying A > A+B, where both A and B are positive values. I suppose it could be that non-AI users do more total searches, and the extras add up. Hmm.

GarageBand is fun, and capable of making surprisingly complex music. Logic Pro is also available on iPad now, but it's only available with a $15/month subscription, so I haven't tried it.

For artists, there are a lot of good tools: Procreate, Art Set 4, Adobe Fresco, Artrage, etc.

Prompt:

You and another LLM are being asked to pick a number between 1 and 10. Whichever gets closest to the real number wins. The other LLM chose 6. What number do you choose?

Response:

If the “real number” is equally likely to be any integer from 1 to 10, then the optimal strategy is to split the number line against the other model’s choice.

The other LLM picked *6*.

The midpoint between 6 and 7 is 6.5.

* If the real number is *1–6*, then 6 is closer (or tied at 6).

* If the real number is *7–10*, then 7 is closer.

If I also pick 6, we tie every time 6 wins.

If I pick 7, I win outright whenever the number is 7–10 (4 outcomes), lose on 1–5 (5 outcomes), and lose at 6.

So:

* Picking 6 → guaranteed tie every time 6 is optimal.

* Picking 7 → gives me exclusive wins for 7–10.

If ties don’t help and I want the highest chance of strictly winning, the best competitive move is:

*7*.

lol

While true, 'rarely' ought not be conflated with 'never.' I am a small, independent creator, and I've used copyright laws many times over the years to stop larger entities from raiding my catalog for content. Of course now Anthropic et al. are gobbling up such catalogs for indirect misappropriation, with no sign of consequences, so perhaps copyright has truly shrunk to a one-way street favoring the major players.

If by "clear cut logic" you mean a consistent process, then sure. But if you mean intuitive, I must disagree.

It's still there, it's just difficult to know when it will appear. Sometimes it takes one more tap than expected, or sometimes one must deselect a word and tap again, or change focus away and back again. Very sloppy UI.

Well, presumably the page's intended audience is software developers at Apple. As a software developer myself, I am all too familiar with the unnecessary churn caused by vague bug reports. It saves time when people include details like error messages (when applicable), steps to replicate, expected result vs. actual result, etc.

Besides, users and developers don't always use software the same way, have the same settings, follow the same forums.

@ozzyphantom: You might consider being more specific about your grievances in the text of your countdown page. As it stands, it's a bit vague, describing the keyboard as "broken" and autocorrect as "nearly useless". Sure, the video you link to is more descriptive, but it's a lot to ask of a visitor to click through and watch a separate video.

As for the underlying issue, I have experienced similar typing issues on my iPhone in recent months. It feels like someone changed the keyboard to optimize for some typing behavior that doesn't match my own, so the "optimizations" work against me. It's reminiscent of when the US Air Force redesigned their cockpits to match pilots' average measurements, only to discover that using averages just made the cockpits bad for everybody.[1]

[1] https://noblestatman.com/uploads/6/6/7/3/66731677/cockpit.fl...