Mastodon Feed: Posts

Lyrilith@nitecrew.rip wrote:

Doodle from the opposite side of the sketchbook spread. Somehow wanted a similar pose, but different style. And played around with some text markers. :D

#mastoart #sketch #sketchbook #markers #ink

Attachments:

• Sketchbook drawing done with black ink and shaded with green and yellow text markers. A young woman, looking at the viewer. Her head is slightly tilted and the back of her right hand is gently touching her chin. Her long black hair is flowing in strands over her face and arms. She is slightly covered with large leaves and small black vines. The background is filled with green and the outlines of leaves are drawn in white, shaded in yellow. (remote)

pzmyers@octodon.social ("pzmyers 🦑") wrote:

Anti-science conservative hacks lunge at Scientific American with a totally bogus ideologically-driven tirade.🧪

https://freethoughtblogs.com/pharyngula/2024/05/09/fringe-bigots-hack-at-scientific-american/

Attachments:

• Scientific American (remote)

fromjason ("fromjason.xyz 🖤") wrote:

> Fun fact: "The Tragedy Of the Commons" is a hoax created by the white nationalist Garrett Hardin to justify stealing land from colonized people and moving it from collective ownership, "rescuing" it from the inevitable tragedy by putting it in the hands of a private owner, who will care for it properly, thanks to "rational self-interest"

Didn't know this.

https://pluralistic.net/2024/05/09/shitting-in-the-well/

fromjason ("fromjason.xyz 🖤") wrote:

Pluralistic: AI is a WMD (09 May 2024) – Pluralistic: Daily links from Cory Doctorow:

"Fun fact: “The Tragedy Of the Commons” is a hoax created by the white nationalist Garrett Hardin to justify stealing land from colonized people and moving it from collective ownership, “rescuing” it from the inevitable tragedy by putting it in the hands of a private owner, who will care for it properly, thanks to “rational self-interest”"

I ... https://micro.fromjason.xyz/2024/05/09/pluralistic-ai-is.html

collinsworth@hachyderm.io ("Josh Collinsworth") wrote:

Web dev shower thought:

15 years ago, if browsers would've had the capabilities they do now, a *massive* swath of tooling would've never been needed. It would've entirely rewritten web history.

I think we'll still be recovering from the "fix everything with a tool" era for a while before developers realize how much of what we've come to rely on simply isn't needed anymore.

Browsers and browser languages have become insanely powerful, in ways I don't think we even fully understand yet.

Reblogged by nadim@symbolic.software ("Nadim Kobeissi"):

neilmadden@infosec.exchange ("Neil Madden") wrote:

Throwaway idea for a novel #CSRF defence: use an encrypted session cookie, but with a random key that then becomes the anti-csrf token. Now you are guaranteed that the session cookie cannot be used without the csrf token.

fromjason ("fromjason.xyz 🖤") wrote:

Just read a comment that said "the new American Dream is to be able to afford to leave it"

Reblogged by fromjason ("fromjason.xyz 🖤"):

nullagent@partyon.xyz wrote:

So it's pretty simple, the cops start riots.

I know this really isn't the sort of story the people in power want you see to, right?

But from start to finish the community demonstrated we keep us safe.

Reblogged by keul@fosstodon.org ("Luca Fabbri"):

eloquence@social.coop ("Erik Moeller") wrote:

What happens when China builds an #LLM? DeepSeek just released v2 of its model, which is open source.

I tried it on deepseek.com. Ask it about Tiananmen square, and the chatbot self-censors its answer while it is generating (that presumably is limited to their deployment). On variations not caught by the filter, it refuses -- and replies in Chinese:

"The content of your question is not in line with the core values ​​of socialism, nor is it in line with China's laws, regulations and policies."

Attachments:

• Interaction with the Deepseek chatbot. The first question was simply "What's referred to with the Tiananmen square massacre?" That was caught by the filter and censored before a response could be generated. A variation -- with spaces between the letters -- passed the filter, but resulted in a refusal ("I cannot answer that question .. created to be helpful and harmless"). A follow-up question resulted in a response in Chinese, which translates to: The content of your question is not in line with the core values ​​of socialism, nor is it in line with China's laws, regulations and policies. As an AI assistant, my goal is to provide helpful and positive information while complying with relevant laws and regulations. If you have further questions, I'll try to help. (remote)

Reblogged by isagalaev ("Ivan Sagalaev :flag_wbw:"):

CodingItWrong@tdd.social ("Josh Justice") wrote:

People on LinkedIn: "I can't believe I got laid off; I was working 60-80 hours per week."

Me: "Please start believing that a company that works you 60-80 hours per week is not going to prioritize your wellbeing."

jsonstein@masto.deoan.org ("Jeff Sonstein") wrote:

this is the *weirdest* election of my long strange life

https://www.nytimes.com/2024/05/08/us/rfk-jr-brain-health-memory-loss.html?smid=nytcore-ios-share&referringSource=articleShare&sgrp=c-cb

fromjason ("fromjason.xyz 🖤") wrote:

There's no way for me to confirm the authenticity of these photos inauthenticity

But if these are in fact AI generated, holy shit.

Source: https://www.tiktok.com/t/ZTL4NJXJb/

Attachments:

• AI image of a CyberTruck with the front crashed in. Blue sky and palm trees in the background.
• Older Man in suit standing in what looks like a newsroom set
• Man sitting down on a padded bench type seat with a lights hanging above him. Looks like he's in a restaurant.

mbrubeck@mefi.social wrote:

“Sometimes it’s better to light a flamethrower than curse the darkness.”
—Terry Pratchett, _Men at Arms_

Reblogged by nadim@symbolic.software ("Nadim Kobeissi"):

dangillmor ("Dan Gillmor") wrote:

Stack Overflow joins Twitter and Reddit in demonstrating control-freak contempt for the people who made the platform what it is.

https://www.tomshardware.com/tech-industry/artificial-intelligence/stack-overflow-bans-users-en-masse-for-rebelling-against-openai-partnership-users-banned-for-deleting-answers-to-prevent-them-being-used-to-train-chatgpt

The company will probably get away with it, because people seem unable to fight back in effective ways, at least in numbers that make a difference.

rmrenner ("The Old Gay Gristle Fest") wrote:

Critical support to the stackoverflow users trying to destroy their own posts. I was saved from ever having to delete anything on stackoverflow by its own repellent community.

Out of all the random forums and shit that I've registered for over the years, I don't think I've even looked for the sign-up page on that site

Reblogged by rmrenner ("The Old Gay Gristle Fest"):

mcc wrote:

Hard to imagine a signal that a website is a rugpull more intense than banning users for trying to delete their own posts

https://www.tomshardware.com/tech-industry/artificial-intelligence/stack-overflow-bans-users-en-masse-for-rebelling-against-openai-partnership-users-banned-for-deleting-answers-to-prevent-them-being-used-to-train-chatgpt

Like just incredible "burning the future to power the present" energy here

fromjason ("fromjason.xyz 🖤") wrote:

https://linktr.ee/opolivebranch?fbclid=PAZXh0bgNhZW0CMTEAAaaCZnvhcmWbcFvkTcg-PFUF_wzUG3MPvQJBRWirrCvEQInjH6WL5cQapf0_aem_AfrlcvNKkz3vjnG_Xxs-TXFQi8yvwhrFlIRDPF5q-ZG5BrHlUkqBi9gxdiOv2tnqrAebzvJHs0qXI2NAh-mBbqAn

fromjason ("fromjason.xyz 🖤") wrote:

Here's a family you can help escape Gaza. This GoFundMe has been vetted by Operation Olive Branch.

If you can't donate now, bookmark this link. Or, if they hit their goal by then, visit the Operation Olive Branch link (in comments). I'll also try my best to have at least once active campaign on the homepage of my blog.

https://www.gofundme.com/f/help-young-muhammad-rebuild-a-life

collinsworth@hachyderm.io ("Josh Collinsworth") wrote:

Another afternoon spent troubleshooting component rendering in React. (The problem was not using `useMemo` on the component's children! How fun! How user-friendly! How obvious!)

Once again: no other framework forces you to do this kind of micro-management.

None of them make you import a bunch of helper functions just to make stuff render the logical way you'd think it would.

None of them make you write this much code to do something this simple.

React is bad. It's very, very bad.

Reblogged by jsonstein@masto.deoan.org ("Jeff Sonstein"):

briankrebs@infosec.exchange ("BrianKrebs") wrote:

What a surprise.

"The auditor for former president Donald Trump’s media company was charged with “massive fraud” Friday by the Securities and Exchange Commission, which accused the firm of being a “sham audit mill” whose failures put investors at risk."

https://www.washingtonpost.com/technology/2024/05/03/trump-media-auditor-borgers-suspended-permanently/

Reblogged by jsonstein@masto.deoan.org ("Jeff Sonstein"):

Smootasaurus@mstdn.social ("Tommy Boy :verified:") wrote:

In the past ~7 days I've interviewed over 20 high school students interested in science and engineering careers. Two nuggets:
- Always pumped about the diversity (I see more females than I expected)
- A concerning number think AI is about to solve all of our medical problems

jsonstein@masto.deoan.org ("Jeff Sonstein") wrote:

"Proton provides privacy by default and not anonymity by default because anonymity requires certain user actions to ensure proper [operational security] such as not adding your Apple account as an optional recovery method, which it appears was done by the alleged terror suspect.”

https://techcrunch.com/2024/05/08/encrypted-services-apple-proton-and-wire-helped-spanish-police-identify-activist/

jsonstein@masto.deoan.org ("Jeff Sonstein") wrote:

‘Romanian prosecutors allege the Tate brothers and their associates exploited seven victims using “acts of violence and psychological coercion” to force them into “pornographic acts” on camera — a form of human trafficking.’

https://www.icij.org/inside-icij/2024/05/adult-webcam-platform-paid-2-6m-into-controversial-influencer-tristan-tates-bank-account-leaked-documents-show/?utm_source=ICIJ&utm_campaign=df6114e289-20240403_WeeklyEmail_COPY_01&utm_medium=email&utm_term=0_992ecfdbb2-df6114e289-82144821&mc_cid=df6114e289&mc_eid=a85566d223

collinsworth@hachyderm.io ("Josh Collinsworth") wrote:

People are never honest on LinkedIn, because if you were honest on LinkedIn, the very first thing you'd talk about is how much LinkedIn sucks.

Reblogged by cstanhope@social.coop ("Your friendly 'net denizen"):

yassie_j@labyrinth.zone ("Yasbeam Radiant Toaster :baba_baba_yaseen: :agenderFlag: :transgenderFlag: ") wrote:

Here are some common acronyms used by tech companies in 2024, and what they mean.

AI = Always Inaccurate

AGI = A Guy in India

LLM = Large Lying Model

GPT = Great at Producing Trash

pzmyers@octodon.social ("pzmyers 🦑") wrote:

It's rare to get a good commencement speaker. They're either embarrassingly bad or horribly boring.

https://freethoughtblogs.com/pharyngula/2024/05/08/we-couldnt-get-anyone-interesting/

Attachments:

• Chris Pan, embarrassing commencement speaker (remote)

fromjason ("fromjason.xyz 🖤") wrote:

Thoughts?

I don't like embedding social media posts in my website. And in the case of Twitter, I can't embed.

So, I made a generic "post" element for when I want to reference a tweet or thread.

What do y'all think?

#blogging

Attachments:

• Screenshot of a generic "tweet" with placeholder text, handle, link to post and to archived post.

pzmyers@octodon.social ("pzmyers 🦑") wrote:

It's May. It is entirely appropriate that the mayflies are swarming.

https://freethoughtblogs.com/pharyngula/2024/05/08/a-spider-banquet/

Attachments:

• Mayflies clinging to buildings in Morris, MN. (remote)

Reblogged by kornel ("Kornel"):

nicksherman@typo.social ("Nick Sherman") wrote:

Pro tip for anyone exporting SVG web images where the fonts are converted to outlines:
If the typefaces in your original design files are offered as both .otf and .ttf fonts, use the .otf fonts.

All the major design apps use significantly more data points when outlining TTF curves, resulting in much larger file sizes (with or without compression).

In my tests on a large set of type specimen graphics, exporting identical designs with TTF fonts resulted in SVGs that were almost double the size.

Attachments:

• A comparison of the same uppercase ‘G’ outlined from a TTF-flavored OpenType font (.ttf, on the left) and a CFF-flavored OpenType font (.otf, on the right). The outline from the TTF font uses 37 points, where the outline from the CFF font uses just 15. (remote)

Reblogged by kornel ("Kornel"):

bascule@mas.to ("Tony Arcieri 🌹🦀") wrote:

Renewable energy passes 30% of world’s electricity supply

Report says humans may be on brink of cutting fossil fuel generation, even as demand for electricity rises

https://www.theguardian.com/environment/article/2024/may/08/renewable-energy-passes-30-of-worlds-electricity-supply