For the curious:

  $ python -c 'print(list(map(__import__("unicodedata").name, "ന്ന")))'
  ['MALAYALAM LETTER NA', 'MALAYALAM SIGN VIRAMA', 'MALAYALAM LETTER NA']

(The "pypyp" package, by Python core dev and mypy maintainer Shantanu Jain, makes this easier:)

  $ pyp 'map(unicodedata.name, "ന്ന")'
  MALAYALAM LETTER NA
  MALAYALAM SIGN VIRAMA
  MALAYALAM LETTER NA

Yes, but from OP's perspective this is a distinction without a difference.

Clearly, but OP would be well advised to apply Hanlon's razor. The victimhood narrative does not improve understanding, which is needed if better outcomes are to be hoped for in the future.

> You apparently did not read the article.

Please don't say things like this in comments (see https://news.ycombinator.com/newsguidelines.html).

I don't think "LLM" and "hallucinated" are accurate; different kinds of AI create images, and I get the impression that they generally don't ascribe semantics to words in the same way that LLMs do, and thus when they draw letter shapes they typically aren't actually modelling the fact that the letters are supposed to spell a particular word that has a particular meaning.

That's pretty hard to reconcile with OP's claim:

> In 2010, I wrote A successful Git branching model and created a diagram to go with it. I designed that diagram in Apple Keynote, at the time obsessing over the colors, the curves, and the layout until it clearly communicated how branches relate to each other over time. I also published the source file so others could build on it.

If you mean that the Microsoft publisher shouldn't be faulted for assuming it would be okay to reproduce the diagram... then said publisher should have actually reproduced the diagram instead of morging it.

I'm glad I actually checked TFA before asking here if "morging" referred to some actual technical concept I hadn't previously heard of.

If we are here, lets at least coin it for something relevant!

This is the first I've heard of it but I was very impressed. Multiple outbound links captured my attention.

For those who want to have the stuff built, yes, it absolutely is.

I'm not sure what distinction you're trying to make, but it seems like you might be distinguishing between keeping out substandard work versus keeping out the submitters.

In which case, I kinda disagree. Substandard work is typically submitted by people who don't "get it" and thus either don't understand the standard for work or don't care about meeting it. Either way, any future submission is highly likely to fail the standard again and waste evaluation time.

Of course, there's typically a long tail of people who submit one work to a collection and don't even bother to stick around long enough to see how the community reacts to that work. But those people, almost definitionally, aren't going to complain about being "gatekept" when the work is rejected.

To be fair, one probably needs at least one idea in order to build stuff even with AI. A prompt like "write a cool computer program and tell me what it does" seems unlikely to produce something that even the author of that prompt would deem worthy of showing to others.

I think Show HN is far more overloaded with "I one-shotted an automation I find useful and then asked an LLM to explain why this is actually revolutionary".