Kind: Reposts

Vanilla HTML/CSS/JavaScript is the “universal language” of modern networked computing. Know that, and know it deeply, and you can speak “internet” in nearly any context, in nearly any job.

That’s the power of international standards and open protocols. They’ll never go out of style.

—@vanilla

OpenAI’s Altman and other AI giants back warning of advanced AI as ‘extinction’ risk. If Sam truly believes in the risks posed by AI, he should consider shutting down OpenAI. Power down the data center, delete software & models associated with it. I don’t get his motivation here. What do you want, man? First,you say OpenAI is the best. Then, the next day you say it will pose a significant risk to the human race. So, shut down the OpenAI & move on with your life. Stop making up bullshit #ChatGPT

—@nixCraft

@tommorris @hc If the execs didn’t want ad blockers, maybe they shouldn’t have shipped ads that broke the website or were malware vectors.

—@astatide

“Haha lol banks use 60 year old programming languages!”

Yes, young whippersnapper, that’s because they need things to be reliable and not change all the time, the code probably has to run for another 60 years.

“Modern” devs could learn a lot from not trying chase every trendy new framework and every shiny new programming language.

—@thomasfuchs

@aardrian my take on the letter here (also before verdict) https://www.lflegal.com/2023/05/higher-ed-digital-access/ #EdTech #a11y #higherEd

—@LFLegal

Dear journalists: please stop embedding tweets. The infrastructure supporting that has, it seems, started to fail, so articles keep reflowing as parts of tweet embeds eventually load or don’t.

—@silvermoon82

In a meeting at work today I was explaining #Protactile and I told a story about how I used it and said “I find it really helpful, for understanding where things are in space and how they relate to where I am… Like when I’m doing archery.” And there was this shocked silence because there is a for real deafblind person telling them about how they shoot arrows. It was so hilarious. And apparently if you want to get people off of a subject just tell them you can’t see or hear but you shoot

—@Pawpower

@TheBierFrau It’s a sign of our beautiful times that advertising ruined, and made near completely useless:

1. Answering telephone calls.
2. Voicemail
3. Text messages
4. E-Mail

—@PatrickoftheG

It’s amazing that Reddit, having watched the fallout after Twitter killed off their third party developers, would go on to say, “You know what, that’s an excellent idea and we should do it too!”

It seems their new API pricing scheme might very well be a deathblow to @christianselig‘s excellent Apollo client (along with tons of others, I’m sure).

https://old.reddit.com/r/apolloapp/comments/13ws4w3/had_a_call_with_reddit_to_discuss_pricing_bad/

—@bigzaphod

Uhg. AI-suggested code is fractally wrong. The function calls it gives me have the wrong arguments (like, it just uses the integer “128” instead of a memory buffer), and after I fix all that, I find that it’s calling the wrong function in the first place. It’s just making shit up.

AI gives plausible-looking answers, not correct answers.

—@AlSweigart

There’s a fediverse reddit version called Lemmy? And reddits API is about to cost a fortune and price out 3rd party apps like Apollo? It’s like the fediverse is just right there waiting every time now.
https://join-lemmy.org/

—@kittywifclaws

The worst part of the Reddit situation is that for folks training certain models, the data is actually worth that much and Reddit will make some money.

So we’re entering into a world where all content of any value to people will be even more locked down because of its value to AI companies.

—@brandonhorst

About ten years ago, the IT building for a local highschool burned to the ground.

My company donated a spare Juniper EX3200 switch to help get them back online. They had erected a small structure without any insulation, and left the switch inside unpowered for a day while the fiber splicers got them hooked back up.

They started the switch, and it immediately shutdown. They looked a gift horse in the mouth and bitched to us that our donation was garbage.

I was dispatched to assist. I connected to the serial console and powered the device up. It immediately indicated that the CPU temperature was 252 C, and shut itself down.

It was January, the outside temperature was -3 C. The Juniper device was storing temperature as an 8-bit unsigned integer value. It got so cold, it thought it was on fire.

Can’t say I ever expected to use a heat gun to get a switch to start before that day. Got it warm enough to power on, and then it kept itself plenty warm to continue operating.

Data types matter, yo.

—@nuintari

It looks like Reddit is keen to pull an Elon around API pricing.

Yes, of course it is risky to build a business based on access to someone else’s platform. They hold the cards. But one of the joys of the open web was the ability to use the client you wanted. A graphical browser, a command line browser, from a script etc.

Once again, what a mess corporatisation of the web has caused.

https://www.reddit.com/r/apolloapp/comments/13ws4w3/had_a_call_with_reddit_to_discuss_pricing_bad/

—@neil

BU President whines about dissent, calls it “cancel culture”

https://popehat.substack.com/p/speech-or-cancel-culture-at-boston

—@Popehat

New Post: Science Briefs – Edition 37 https://www.worldblindherald.com/science-briefs-edition-37

—@worldblindherald

If you claim the tech you are building is an existential threat, either you’re lying about the threat to sensationalize the tech’s capabilities, or you should stop building it. Or both? It’s probably both.

—@blakereid

If you really want to know why and how ChatGPT works, here is a fitting post about it: Stephen Wolfram explains the basic mechanisms, how the probabilities are calculated, how neural networks work and how ChatGPT uses all that concretely.

It’s a long post and you need some pauses to think about it. But I think it’s worth reading.

https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/

#ChatGPT #AI

—@robert

Honestly, the most alarming thing about AI isn’t so much about AI itself, but about how utterly hell-bent humans are to use it for things that it does a bad job at. H sapiens is bound and determined to use this chisel as a screwdriver.

Case in point: the recent news story about a lawyer (or pair of lawyers – finger pointing is underway) who submitted a filing in federal court which had actually been written by ChatGPT.

What is the one thing one can assume everyone has heard about LLMs? That they make completely bogus shit up, including inventing nonexistent citations.

It would be hard to overstate how unacceptable to a court it is for a lawyer to submit a legal argument which cites nonexistent case law. That’s the kind of shit that can get a lawyer disbarred. It’s a, uh, *career-limiting* move.

But apparently some lawyer actually did it: he took the output of a computer program famous for fabricating false citations and piped it directly into a court.

1/?

—@siderea

I’m hiring for a GiveWP Tech Support position!

$22/hr, full time. Benefits for US folks, heavy focus on maintaining documentation. Apply here: http://app.jobvite.com/m?3e0CTmwA

Boosts are appreciated. Boosts will be compensated via unrecorded happy dances and/or spirit fingers, in addition to the regular positive karmic returns.

#WordPress
#TechSupport
#Documentation
#Editing
#Developer

—@BenUNC