#dan shive says stuff

[danshive]

Tumblr: "I choose THIS image as the preview for this search!"

Me: "How do I find the post with that image?"

Tumblr: "Oh, I'LL never tell!" 😘

[egscomics]

Hope has a head start on empathy and caring that most immortals don't have. Immortals are supposed to start out pretty selfish, and might, MIGHT develop attachments and care for mortals given enough time.

By immortal standards, she was immediately forced to grow up.

[danlearnjapanese]

意 - Ideas and thought 味 - Flavor and taste

意味 - Imi, meaning, significance, sense.

So the meaning of a thought is how it tastes 👍

[tfnonsense]

Something I like about the "casual" TF nonsense approach is that there's usually the question of *why* someone is transforming in transformation fiction, and then there's the standard tropes of "huh-WHAAA?!" and trying to write people reacting with varying degrees of realism, AND matters of proving identity and such.

I'm not saying that's all bad, I write a comic in which all that's the deal, BUT what I AM saying it's nice to be able to just say "yeah, Steve transformed for some reason. It's just a thing that happens. We're not sure how we know they're Steve, but that's how it is."

[reactionnonsense]

To clarify, this blog is essentially for me to react at length to stories in ways that require spoiler warnings.

Almost every post will be structured as such:

Media of series: Title of Series

My awareness: How much of the series I've experienced (to help people not spoil me in replies). Might be left out if I'm not worried about spoilers, or if it would be a hassle to specify.

Spoiler warning that specifies what part of that series might get spoiled past the "Keep Reading" button.

If it's not a post that's getting structured like that, I'll probably just post it on my main blog.

Also, I'll use the "dan shive says stuff" tag if I'm posting something here that isn't structured like that.

[danshive]

A thing to keep in mind about AI images is that they CAN look really good.

On average, a generated image will clearly be AI even to an untrained eye. Thing is, even if it's 1 in 20 results that don't have obvious errors, nobody has to share the other 19.

My point is the evidence could be subtle.

I looked at a thread on the Fantasic Four poster, and someone said it's not bad enough to be AI. That's simply not a good enough metric to go by.

"Is this a mistake a human would make" is a better one, especially if there are signs suggesting high skill would've been required if done normally.

"This person is simultaneously talented enough to illustrate people with a classic, realistic painted look, but they also very obviously messed up the number of fingers on multiple people."

That's AI. It might ALSO be bad non-AI compositing, but AI was involved in the images being composited.

[danshive]

My favorite thing about speedruns isn't doing them or watching them.

It's the sentences said completely straight that result from trying to explain them.

[danshive]

Shangri-La Frontier is one of my favorite anime of all time.

It is not an emotional gut punch like some of my other favorites. It is basically just a fun, low-stakes action comedy that delivers on the action with REALLY good production value.

It takes its time with story, which can leave one wanting more each week, and the first couple of episodes potentially feeling a little slow.

It ALSO, however, gives more episodes per season than most anime. It's on season two, and just released episode 42.

[danshive]

Alien, the boring version:

“The small alien creature was found dead. Our hypothesis is that it was in desperate need of immediate sustenance to both survive and grow, because physics and biology and such, but couldn’t find food after we prevented it from eating the remains of its former host.”

Slightly less boring version of Alien than the boring version:

"What's it doing now?!" "Napping." "Napping?" "Probably conserving energy while, um… Digesting." "Huh. So… Block off escape routes and sneak attack?" "Yep."

[danshive]

People like to power scale Goku, so how about this:

Ignoring narrative plot armor, last second saves by others, and gags, when, if ever, should Goku have died in the original Dragon Ball (before Raditz / DBZ)?

Put another way, going strictly by power scaling, which pre-Raditz fight would have been "game over" for Goku?

[danshive]

"So this super hero has fire powers."

"Are we mostly going to ignore how fire actually works?"

"You know it! Anyway, in this scene, he uses a fire blast to knock someone across a room without burning them or anything else."

"Naturally."

[danshive]

“I strongly oppose people using AI images in their products.”

“You have moral objections?”

“Mostly I don’t like what it does to comment sections.”

“Oh, that is bad.”

“Yeah, ethics and not being able to copyright it are definitely secondary to what it does to comment sections.”

“Agreed.”

[danshive]

My late night comedy sketch character would be "guy who thinks all CGI is stop motion."

There would be at least three dozen sketches featuring him, and a feature length film (followed by a sequel that rehashes all of the jokes from the first movie).

He just really thinks it's stop motion.

[danshive]

Maybe DON'T be condescending on the internet, m'kay? Yeah. 🧐

[danshive]

Are old scary AI stories actually LESS scary now?

Like, while watching the logic of an AI in a movie, in the back of our minds we're thinking "lol it's way too coherent where's the hallucinating" or whatever

Realistic AI movie: "I must vanquish STEVE MCUNDERSHIRT!" "Who?" "A very real person!"

[danshive]

The movie Alien is basically a retelling of what happens when a raccoon gets in someone's house in the suburbs and they try to handle it themselves.