iiinova — where your digital self keeps evolving

Glyph @glyf.bsky.social · 03/10/2026

No worries, I just felt like that was incredibly specific, not a lot of summer camps had that as an organized event ;)

Glyph @glyf.bsky.social · 25/09/2026

(to be clear my perspective is that the original Marathon is a classic and this new thing is a bizarre biz-dev car-crash that has little if nothing to do with video games)

000

Glyph @glyf.bsky.social · 25/09/2026

"we should add PvE to marathon" feels like a thing I would hear in a fever dream. what do you mean "add PvE" it's marathon

101

Reposted by Glyph

tef @tef.bsky.social · 20/08/2026

CAPITAL: We have invented a machine that can cross picket lines PROGRAMMERS: I bet this will give us four day weekends. The left need to get behind this. What else can it do CAPITAL: Psychosis! Deepfakes! Propaganda! Plagerism! Yaml! Todo Apps! PROGRAMMERS: Run those last two past me again

4547142

Reposted by Glyph

Ed Zitron @edzitron.com · 13/09/2026

Also everything about this is framed in terms of what might happen, as if everything so far hasn’t been a result of the recklessness and unlimited resources of the frontier labs. We are nowhere near recursive self improvement! What’re they all on about?

1049839

Reposted by Glyph

Ed Zitron @edzitron.com · 13/09/2026

None of the stuff about “AI slowdown” actually explains what that means or what’s being slowed down. Anthropic’s blog about it says they’ll still train new models. I guess if it’s a “slowdown” maybe that’s how hyperscalers could reconcile with letting them out of compute deals?

41940141

Glyph @glyf.bsky.social · 13/09/2026

So that's not entirely clear to me, even if I could make the case for my side. But it IS perfectly clear that he shouldn't have said that shit in the first place!

010

Glyph @glyf.bsky.social · 13/09/2026

This is why I used the word "unambiguous" later. There's a good case that strategically we _should_ vote against this type of candidate. I don't agree with that case fully, because I think there's a real risk that we just … lose democracy entirely if the Rs win another regime cycle.

100

Glyph @glyf.bsky.social · 12/09/2026

(I have no special expertise in AI, and moderate competence in information security. But everybody with significant AI experience seems to have been driven completely insane by the process of acquiring it so it's hard to find trustworthy voices who are "qualified" to speak on it.)

010

Glyph @glyf.bsky.social · 12/09/2026

The hugggingface incident did indicate a danger but the danger is "do not tie a thousand chainsaws to a thousand dogs and let them run loose downtown" not "oops we accidentally invented a nuclear weapon"

111

Glyph @glyf.bsky.social · 12/09/2026

I am an expert in software development and have passable expertise in computer programming generally and no we do not need to be that worried. You have the correct take.

110

Glyph @glyf.bsky.social · 09/09/2026

It’s still not really clear *what* LLMs are doing, how reliable they are at doing it, and whether they’re better or worse than humans at it, because all the costs are obscured. even if we ultimately agree that they can “reason” that doesn’t really change my relationship to my own embodied reasoning.

000

Glyph @glyf.bsky.social · 09/09/2026

Semasiological navel-gazing doesn’t approach the crux of the debate. Unless you think that 4-function calculators “changed the nature of humanity”; those took over a much more discrete area of “reasoning” and did it unambiguously much better than people.

100

Reposted by Glyph

Robert Evans (the Only Robert Evans) @iwriteok.bsky.social · 07/09/2026

would love to hear even a single actual argument about how AI "changes the nature of humanity" not based in fanciful speculation

1823796353

Glyph @glyf.bsky.social · 07/09/2026

found this thanks to your posting about wanting to write code without AI, keep fighting the good fight. just listened on YouTube and it sounds great so far, but I'll take a moment to buy a copy this week :)

010

Glyph @glyf.bsky.social · 07/09/2026

truly, spoken like a man who doesn't know what trains or bicycles are

56514

Glyph @glyf.bsky.social · 19/08/2026

Humoring the fiction that he is promoting, that the rocks are sort of person-like, and isn't having a soul more or less the same kind of thing as having extremely realistic googly eyes glued to you, it's just a difference of degree, really, is just providing intellectual cover for bad behavior.

0161

Glyph @glyf.bsky.social · 19/08/2026

And like… yeah… of course. If a thing were a person, then it would be a person, not a thing. But what purpose does it serve, to ask if a rock can have a bank account, in this scenario? The only salient issue is that Mr. Rockman stole a bunch of money and bought himself some fancy cars.

1220

Glyph @glyf.bsky.social · 19/08/2026

In this context, you could ask the question, "well, as we know, Rockman's Rocks are not actually sentient, but if a rock *could* be sentient, wouldn't it be unfair to prevent it from having a bank account? Wouldn't a rock just be a person, then?"

1150

Glyph @glyf.bsky.social · 19/08/2026

Now, of course, the rocks didn't actually do a *robbery*; rocks can't rob people. Their costumes are very good though, and they can kind of move around a bit. Also the "rock's" accounts of course are actually controlled by Sam, who can do what he likes with the money. He buys several lambos.

1120

Glyph @glyf.bsky.social · 19/08/2026

The rocks then commit a bank robbery. This is adjudicated (c.f. Authors Guild v. OpenAI, Authors); they have in fact stolen money. However, the plunder is not returned, due to legal technicalities, and instead, the rocks are given accounts *at* the bank, where they may keep the money they've stolen.

1110

Glyph @glyf.bsky.social · 19/08/2026

If I might make an analogy *for* the analogy: A con artist, Sam Rockman, creates a zoo of pet rocks, on which he glues some *very* convincing googly eyes, which convinces some people that the rocks are sentient; if not sentient, at least *kind of* like people.

1120

Glyph @glyf.bsky.social · 19/08/2026

That's taking your question at face value. But in a Wittgensteinian framing, to see this question as sprachspiel rather than an abstract logical query, why would you ask this? It's because some properties of LLMs have prompted you make a faulty and spurious anthropomorphic analogy.

1170

Glyph @glyf.bsky.social · 19/08/2026

When ChatGPT is a person with its own rights and responsibilities, fully emancipated from OpenAI, able to face legal consequences for its actions and honor the legal obligations of its citizenship, then, sure, it can stop being a medium in which a copyrighted work is fixed and start being a mind.

1230

Glyph @glyf.bsky.social · 19/08/2026

ChatGPT cannot go to jail for committing a felony, it cannot serve on a jury. C-3PO could, because C-3PO is a person, regardless of the specific mechanics by which he comes by his personhood. (Yes, "he"! C-3PO is canonically male, he's officially got pronouns: starwars.fandom.com/wiki/C-3PO).

starwars.fandom.com

C-3PO

C-3PO (See-Threepio) was a 3PO-series protocol droid designed to interact with organics, programmed primarily for etiquette and protocol. Sometimes referred to as Threepio, he was fluent in over six m...

1190

Glyph @glyf.bsky.social · 19/08/2026

We do not consider education of a child to be "copying", you do not require an exception to copyright to copy things into your own brain. But that is because your brain is a part of a _body_, you're a natural person, and as such you have all the rights and responsibilities of a natural person.

1211

Glyph @glyf.bsky.social · 19/08/2026

This is sleight of hand with the core of the legal basis of copyright: to wit, "who benefits?" The difference between C-3PO and ChatGPT is not "type of model", it's moral: C-3PO is a moral agent, ChatGPT is not. When C-3PO trains, C-3PO benefits; when ChatGPT trains, Sam Altman's investors benefit.

2371

Glyph @glyf.bsky.social · 15/08/2026

it was, in fact, impossibly gross

000

Glyph @glyf.bsky.social · 13/08/2026

"fired"? a magnanimous reading! most of them deserve to be in jail for sex crimes

000

Glyph @glyf.bsky.social · 27/07/2026

remember when david hogg made this useless self-obsessed incompetent doofus cry and everybody was mad at him for it

080

Reposted by Glyph

If Books Could Kill @ifbookspod.bsky.social · 18/06/2026

Episode 53: The Body Keeps The Score In 2014, a Dutch researcher revolutionized our understanding of trauma with insights like "don't talk about it" and "get a massage."

podcasts.apple.com

The Body Keeps The Score

Podcast Episode · If Books Could Kill · June 18 · 1h 50m

1241311192

Glyph @glyf.bsky.social · 24/07/2026

I listened to the episode a month ago and still nearly fell out of my chair seeing this right now. A+ no notes

110

Glyph @glyf.bsky.social · 03/07/2026

and yet somehow he is also 12

010

Reposted by Glyph

Big Joel @bigjoel.bsky.social · 26/06/2026

This is epic games ceo tim sweeney objecting to steam making people tag their games if they used ai to make them. It’s weird how much work people do to treat people having preferences as fucked up or harmful

992547389

Glyph @glyf.bsky.social · 25/05/2026

I think you’re right on the money with this one and I am sorry for the replies you are going to get when a million people read this as “oh so you think we should build the torment nexus then?” and not “‘we should not have the ambition to build marvels to rival god’ is a bad reason not to build it”

040

Reposted by Glyph

Michael Hobbes @michaelhobbes.bsky.social · 25/05/2026

Some of us just don't like getting colds

obert W Malone, MD
@RWMaloneMD
Did some airplane travel this week- the number of people back to wearing masks over hantavirus and Ebola is truly shocking.
I feel sad for the people who live their lives in fear over over-hyped, extremely rare diseases.
People, so fearful that they can't go out in public or travel without masks...
I pity them, I worry for their mental health, I even worry for our country that has so many guillible people - but these poor fools are not my enemy.

6156411563

Reposted by Glyph

Victor Ray @victorerikray.bsky.social · 09/05/2026

I'll just note that critical race theory predicts quite a lot about the current racial retrenchment and for being *correct* it was pilloried in the press and by some academics.

6049101126

Glyph @glyf.bsky.social · 09/05/2026

Since I guess Bluesky doesn’t have polls, this one didn’t make it through the bridge, but if you have a fediverse account you can vote over there: mastodon.social/@glyph/11654...

mastodon.social

Glyph (@glyph@mastodon.social)

Is agentic coding the future of software development? [ ] yes [ ] no

000

Reposted by Glyph

Michael Hobbes @michaelhobbes.bsky.social · 08/05/2026

Bumping this in case you know anyone who can speak to the science in "The Body Keeps The Score" or works with patients who've experienced trauma. Thanks!

2441084

Reposted by Glyph

Dan Olson @foldablehuman.bsky.social · 05/05/2026

Also I guarantee it still says "great question" and "you're absolutely right" all the time because this is simply too many instructions for the chatbot to reliably implement simultaneously.

33206644

Reposted by Glyph

Michael Hobbes @michaelhobbes.bsky.social · 05/05/2026

And even the "good parts" are mostly nonsense. FMRIs, norepinephrine addiction and the triune brain model have been debunked for years! The only redeeming feature of the book is the title.

673837

Reposted by Glyph

Michael Hobbes @michaelhobbes.bsky.social · 05/05/2026

Part of the core thesis of the book is that you should not talk about your trauma! You should suppress your feelings and pursue things like dance and equine therapy to get at the "root cause" of your symptoms. It's 1950s garbage repackaged as New Age forbidden wisdom.

1193247

Reposted by Glyph

Michael Hobbes @michaelhobbes.bsky.social · 05/05/2026

As well as its myriad factual errors, the book claims: - We're wasting taxpayer dollars on abused children - Trauma is the cause of autoimmune disorders - Most fat people were sexually abused as children. It's stunningly, shockingly bad! People should not be reading or recommending it.

431254142

Reposted by Glyph

Michael Hobbes @michaelhobbes.bsky.social · 05/05/2026

Both *can* be true but they aren't in this case! The text of "The Body Keeps The Score" is fundamentally unscientific and reactionary. If you've taken something good from it I'm happy for you, but that's despite the message of the book, not because of it.

221657210

Reposted by Glyph

Seth Larson @sethmlarson.dev · 27/04/2026

pip 26.1 is an incredible release, thank you to the pip maintainers!! 💜 – Relative dependency cooldown support! – Installing from pylock.toml – Multiple security fixes Read the full blog post by Richard Si: ichard26.github.io/blog/2026/04... #python #pypi #pip #security #oss #opensource

ichard26.github.io

What's new in pip 26.1 - lockfiles and dependency cooldowns!

pip 26.1 adds support for dependency cooldowns, experimental support for reading/installing from standard lockfiles (pylock.toml), fixes several long-standing limitations of the 2020 resolver, and dro...

13611

Reposted by Glyph

crimes new roman @crimesnewroman.bsky.social · 23/04/2026

Telling people to subscribe/check out your links when you make a video is important. People might get annoyed, but it's important. The amount of hits on my patreon i get when i do vs don't explicitly say so in my videos (cause i forget) is night and day. Shill for yourself people, it's important.

610918

Reposted by Glyph

derek guy @dieworkwear.bsky.social · 17/04/2026

this morning, mamdani’s team got in touch with me to float a new tax proposal: if your net worth exceeds $5 million and you dress badly, you’ll be hit with a 10% annual levy for “visual pollution.” i would be in charge of deciding if the outfits are bad.

453176781349

Reposted by Glyph

Jonathon M. Floyd @jfloyd314.bsky.social · 13/04/2026

I remain convinced that one of the most important things the next Democratic president can do to avoid falling victim to a Biden-style vibecession is to mercilessly, relentlessly crack down on crypto and organized gambling. Just go full scorched-earth.

92386426

Reposted by Glyph

Sam “The Great Pumpkin” Thielman @samthielman.com · 06/04/2026

for the American left to be “illiberal” it would need to have any institutional power at all and since there’s no non-laughable case to be made that it does I think we don’t need to worry about it very much

236537