Mastodon Feed: Posts

Boosted by glyph ("Glyph"):
normplum@fosstodon.org ("Norm") wrote:

Today I learnt (realised) that #Python's `json.loads()` and `json.dumps()` functions aren't plural versions of `json.load()` and `json.dump()`. The `s` stands for `string` (since they read/write from strings instead of files).

#TIL #programming

Boosted by glyph ("Glyph"):
alilly@solarpunk.moe ("Athena L.M.") wrote:

Tired: Push notifications via HTTP
Wired: Push notifications via XMPP
Hired: Push notifications via PostgreSQL NOTIFY/LISTEN

Boosted by dysfun@treehouse.systems ("gaytabase"):
ricci@discuss.systems ("Rob Ricci") wrote:

@atsuzaki @dev getting there

pzmyers@freethought.online ("pzmyers 🕷") wrote:

"scientifically accurate" is not a term to throw around lightly, especially not to promote a movie with sloppy science.

https://freethoughtblogs.com/pharyngula/2026/03/24/im-still-cranky-about-project-hail-mary/

https://www.youtube.com/watch?v=jIgzqt7zLg0

Boosted by glyph ("Glyph"):
nixCraft ("nixCraft 🐧") wrote:

lmao 🤡

zkat@toot.cat wrote:

me before: PR going back to Spain is a hard-line right-wing position for sure

me now: wait. Becoming a Spanish "nationality" would basically mean PR is heavily autonomous, while still having "federal" representation, protecting its culture/language stuff, and people wouldn't need to be afraid of us "ending up like our neighbors", which seems to be a big hangup? And we'd get EU membership as a nice side?

I feel like I need to think about this more, but it's looking really fucking nice compared to becoming a US state or the current status, and it would ease fears about having to fare for ourselves?

Other boricuas only: What are y'all's thoughts?

#PuertoRico

Boosted by glyph ("Glyph"):
thomasfuchs@hachyderm.io ("Human Brain Enthusiast") wrote:

I’ve a great idea for an AI startup, using leftover Juiceros to squeeze more performance out of GPUs.

Pls send $10 billion

Boosted by glyph ("Glyph"):
SnoopJ@hachyderm.io wrote:

@xgranade me, sowing references: haha, fuck yea!!! Yes!!

me, reaping use-after-free: Well this fucking sucks. What the fuck.

Boosted by adam@social.lol ("Adam"):
neatnik@social.lol ("Neatnik") wrote:

It’s Tuesday, so it’s time for yet another Mastodon object storage migration! Seriously, hopefully this one will be the last. More info here: https://status.neatnik.net

TL;DR: You may notice some weirdness with *remote* images over the next several hours, but it’ll be temporary. If you notice any super weird issues, feel free to ping me.

dysfun@treehouse.systems ("gaytabase") wrote:

enable javascript to continue

no lol?

Boosted by GuillaumeL@hachyderm.io ("BigSaur G"):
jasongorman@mastodon.cloud ("Jason Gorman") wrote:

Karpathy posted something not so long ago about how much he was letting Claude agents write his code, and the boosters were sharing it far and wide as "proof" the agentic age had arrived.

Fact is, not even Karpathy believes that.

*You* are the agent.

[

Boosted by NfNitLoop ("Cody Casterline 🏳️‍🌈"):
EmilyEnough@hachyderm.io ("Emily 🏳️‍🌈🏳️‍⚧️") wrote:

My biggest problem with the concept of LLMs, even if they weren’t a giant plagiarism laundering machine and disaster for the environment, is that they introduce so much unpredictability into computing. I became a professional computer toucher because they do exactly what you tell them to. Not always what you wanted, but exactly what you asked for.

LLMs turn that upside down. They turn a very autistic do-what-you-say, say-what-you-mean commmunication style with the machine into a neurotypical conversation talking around the issue, but never directly addressing the substance of problem.

In any conversation I have with a person, I’m modeling their understanding of the topic at hand, trying to tailor my communication style to their needs. The same applies to programming languages and frameworks. If you work with a language the way its author intended it goes a lot easier.

But LLMs don’t have an understanding of the conversation. There is no intent. It’s just a mostly-likely-next-word generator on steroids. You’re trying to give directions to a lossily compressed copy of the entire works of human writing. There is no mind to model, and no predictability to the output.

If I wanted to spend my time communicating in a superficial, neurotypical style my autistic ass certainly wouldn’t have gone into computering. LLMs are the final act of the finance bros and capitalists wrestling modern technology away from the technically literate proletariat who built it.

Boosted by NfNitLoop ("Cody Casterline 🏳️‍🌈"):
nina_kali_nina@tech.lgbt ("Nina Kalinina") wrote:

When I first watched War Games (1983) I thought "wow, so weird, not only they had terrible password management, but their test 'AI' system was directly linked to the prod".

Ha-ha. :blobcat_thisisfine:

Boosted by NfNitLoop ("Cody Casterline 🏳️‍🌈"):
johnrogers.bsky.social@bsky.brid.gy ("John Rogers") wrote:

Once again, it’s important to remember that the stock market long ago ceased to be linked to a rational valuation of the companies listed and is just a mood ring for rich people.

RE: https://bsky.app/profile/did:plc:lm4eq2cftwbnrmzkzowigpt7/post/3mhpymbhcvk22

Boosted by NfNitLoop ("Cody Casterline 🏳️‍🌈"):
Migueldeicaza ("Miguel de Icaza ᯅ🍉") wrote:

What are the odds

dysfun@treehouse.systems ("gaytabase") wrote:

i'm a little confused. my system load just jumped by about 4. i am running a single threaded program.

Boosted by slightlyoff@toot.cafe ("Alex Russell"):
mattround@crispsandwi.ch ("Matt Round") wrote:

How is this already over 5 years old?! *crumbles to dust*
https://vole.wtf/kilobytes-gambit/

fromjason ("fromjason.xyz ❤️ 💻 ✍️ 🥐 🇵🇷") wrote:

Someone commented on my post on LinkedIn using the term "enterprise-grade AI" and I had such a visceral reaction like sucking a lemon.

Boosted by zkat@toot.cat:
dougiec3@libretooth.gr ("Wokebloke for Democracy") wrote:

" A report published last week by the Cato Institute, written by director of immigration studies David J. Bier, found that the State Department and Department of Homeland Security were receiving millions of applications from immigrants whom Trump has made ineligible for legal status and pocketing the fees without ever processing the requests."
https://www.commondreams.org/news/immigration-fee-fraud-state-dept

#TrumpAndEpstein #Money #Grift #Immigration #GrandTheft #USTreasury

dysfun@treehouse.systems ("gaytabase") wrote:

a part known as the 'flange'

Boosted by isagalaev ("Ivan Sagalaev :flag_wbw:"):
dalias@hachyderm.io ("Cassandrich") wrote:

@elilla "Signal-shaped noise" is an utterly brilliant characterization of what "gen AI" produces.

Boosted by dysfun@treehouse.systems ("gaytabase"):
hannah@posts.rat.pictures ("historic drystone sheep dyke") wrote:

I really hope pete hegseth got the same windows 11 login screen as me today

slightlyoff@toot.cafe ("Alex Russell") wrote:

This post by @mozilla's VP of Engineering (Andrew Overholt) is welcome news.

For far too long, Mozilla has been nearly silent as Apple suppressed real choice in browsers, and any break with that inglorious pattern should be celebrated:

https://blog.mozilla.org/netpolicy/2026/03/23/competition-innovation-and-the-future-of-the-web/

/via @owa
/cc @jaffathecake

Boosted by jwz:
AmyZenunim@unstable.systems ("★ Amy Star ★") wrote:

in case you need a taste of how fucked the tech industry is right now, I'm being required to use AI at work. if I talk about how it fucks up or overcomplicates basic asks, it's because I "don't know how to use it" which indicates a "lack of growth mindset", and thus poor performance. I've been told this directly to my face, starting immediately.

so not only must I use Claude, I have to cover for Claude's mistakes, and then go the extra mile to pass off my own work as Claude's.

db@social.lol ("David Bushell 🪿") wrote:

hey look at all those amazing WebKit features in Safari 26.4

shame Apple doesn't support iOS 18 🫠

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
ben@werd.social ("Ben Werdmuller") wrote:

What ICE is doing in US airports today, it will be doing at US polling places in November. These photos aren’t funny or incongruous; they should be giving us nightmares. This has to end now.

dysfun@treehouse.systems ("gaytabase") wrote:

okay i have heard some bad british accents in my time but this is a whole new level 😂 https://www.youtube.com/watch?v=tcRxa6E1MUw&t=1050s

Boosted by dysfun@treehouse.systems ("gaytabase"):
javi@goblin.band wrote:

Firefox updated their Terms of Use? Let's see!

As you type a search query within Firefox, Firefox offers search suggestions to provide you with faster and more direct access to what you’re looking for. Some of the search suggestions come from your search provider (“Search Suggestions”). Others come from Firefox, and are based on information stored on your local device (including recent search terms, open tabs, and previously visited URLs), or content from Mozilla and Mozilla’s partners, including paid sponsors and internet resources like Wikipedia (“Suggestions from Firefox”).

Here chat. Here. This is where Firefox dies.

"information stored in your local device" and "content from mozilla's parners" and "paid sponsors".

This is a very convoluted way of saying "we use your personal data to segment you into something we can sell to advertisers".

This is EXACTLY what chrome does, this is exactly why a lot of us stopped using Chrome and moved back to Firefox.

In some circumstances Mozilla’s partners will receive de-identified search and interaction data, in order to serve relevant suggestions and measure user engagement with suggested content.

This is making me really mad. THIS IS JUST CORPO-SPEAK TO DESCRIBE HOW THE ENTIRE INTERNET ADVERTISEMENT INDUSTRY WORKS. This is HOW FACEBOOK WORK. This is how GOOGLE WORK. This is how the entire programmatic advertisement industry work. This is what we call "sell your personal data". No, no one sells your address, no one sells your name. BECAUSE IT'S ILLEGAL IN A SIGNIFICANT PART OF THE WORLD.

We also work with advertising providers to deliver relevant sponsored content using programmatic technologies. To support this, we may share limited, non-identifying information — such as device type, IP-derived location information, and category of content viewed — to help determine which ads to display. We don’t share any information that identifies you. You can turn off sponsored content in your New Tab settings at any time.

Oh it's so nice of you Mozilla, to do THE MINIMUM LEGAL REQUIREMENTS when selling our data. You don't share information that identify me? so nice of you! you know how else does that? Meta! Google! Tiktok! Somehow big tech mega corporations are willing to comply with the minimum legal requirements as you do, mozilla!

In some cases, we may share or publish aggregated and anonymized data to facilitate research or as part of the lawful business purposes outlined above (such as sharing aggregated insights with advertising partners).

This is called "advertisement segmentation" and it's what it paid for Zuckenberg fortress in Hawaii!! Going places, Moz, you are operating exactly as how Facebook used to do in 2016!

To provide our services as described above, we may disclose personal data to: Partners, service providers, suppliers and contractors

"We never disclose your personal data!!! well, unless it's one of our partners who pays us for it, of course!"

oh wait! they include a table of what kind of data they share with partners!

Technical dataLocationLanguage preferenceSettings dataUnique identifiersSystem performance dataInteraction dataSearch dataBrowsing data

The SHARE FUCKING EVERYTHING. THEY ARE SELLING EVERYTHING. "Unique identifiers" is the closest to personal identifiable data they can sell. That's what advertisers can use to make a profile of you: They may not know your name, but they will know everything else about you.

This is the same information that google collects and sells from you. THE SAME.

Fucking ghouls. This is where Firefox died, folks.

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
emilymbender@dair-community.social ("Prof. Emily M. Bender(she/her)") wrote:

Looking forward to this response to "AI literacy day" on March 29, with other members of the cast of Ghost in the Machine

https://notaidoc.com/literacy

Virtual event 9am Pacific/noon Eastern/18:00 CET

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
sue@glasgow.social ("Sue Smith") wrote: