Yes that’s what I meant thanks for clarifying. I think local models are the future but I don’t see anyone serious arguing it. If it were true these city sized data centers will be graveyards in 5-10years
Is there anyone serious who thinks that the future is local models anyway? All computers used to be the size of rooms like these data centers and then they got smaller and faster until the home computer came. Is that not a possibility down the line as we improve efficiency of the models and increase compute?
I think native apps are critical infrastructure in AI development particularly around agents. The truth is there’s no good native interaction layer for custom agents. If you want to wire up and self host an agent that has access to anything ever your only option is a janky port to telegram or Slack. I’ve been building vessels.app because I think it’s the missing piece to agent interaction. I need testers if anyone is interested!
I use agents to do most of the tedious admin for my hire business, and I built www.vessels.app to run them on the go because there was no native solution to talk to my agents. I’ve started working towards releasing this to the public because it’s so much better than setting up agents via telegram or slack.
It sucks and I hate it but this is an incredible steam engine engineer, who invented complex gasket designs and belt based power delivery mechanisms lamenting the loss of steam as the dominant technology. We are entering a new era and method for humans to tell computers what to do. We can marvel at the ingenuity that went into technology of the past, but the world will move onto the combustion engine and electricity and there’s just not much we can do about it other than very strong regulation, and fighting for the technology to benefit the people rather than just the share price.
AI coding IS SaaS. Claude Code is a subscription service I pay to create the exact, niche, I need you to build a form that submits a blog post to Wordpress here’s my api key, here’s the word doc format it comes in, I’ll paste it in the chat and you make a paste from word button so I don’t have to manually enter the form and it works first try. That is the definition of software as a service. And it’s mine and it’s exactly what I need. Tomorrow I’ll need something else…
“The regulator would need to assess whether platforms were taking reasonable steps. If they were not, it could take that platform to court to seek fines.
There would be an independent evaluation of the ban conducted by an academic advisory group examining the short-term, medium-term and longer-term impacts of the ban.
“It will look at the benefits over time, but also the unintended consequences,” Inman Grant said.
A 14-year-old boy looking at social media on his mobile phone. Tech giants Meta and TikTok said on October 28 they will obey Australia's under-16 social media ban but warned the landmark laws could prove difficult to enforce. Australia will from December 10 force social media platforms such as Facebook, Instagram and TikTok to remove users under the age of 16. (Photo by David GRAY / AFP) (Photo by DAVID GRAY/AFP via Getty Images)
How is Australia’s social media ban affecting you and your family?
Read more
“Everything from are they sleeping? Are they interacting or are they actually getting out on the sports fields? Are they reading books? Are they taking less medication like antidepressants? Are their Naplan scores improving over time?” Inman Grant said.”
I’m a professional graphic designer, people in the industry use font, type and typeface interchangeably. No one goes “Umm Actually…”
you should also tell that to who wrote css, because font-weight doesn’t make sense if a font is already a specific weight. Words mean something specific until they don’t and the meaning changes over time and that’s okay
They have a panel who are reviewing it’s effects across all aspects of teen life including sleep patterns, school grades ect and then nationally test results ect
My cousin suddenly has been very captured and obsessed by an area of opinion I didn’t have a name for, fixed money supply, all inflation inherently bad, Elon Musk is badly treated, longer government terms (which sounds reasonable initially until you actually think about just having LESS democracy), no minimum wage. After some research it’s definitely coming from influencers linked to the SV techno feudalists - it’s just such a strong change. But you realise real power is only useful if people can come along with you - if you can build support with the public…
The only AI tools that will ever be truly useful are the ones you build yourself. Basically in this world useful = dangerous. Moving files around, changing file names, deleting files, reading emails responding to emails. The AI’s can do it but it’s dangerous, safeguards like human in the loop aren’t feasible at scale. Yet I’ve built agents or used Claude Code in folders to do this manually and it’s amazing - but every application with an AI button now you just KNOW it can’t do the thing you want it to do.
Literally could only do basic html css before. Now I can build solid Sveltekit web apps from scratch. It’s still hard and frustrating. I’ve spent all day trying to get OTP auth working it’s very frustrating it misses things. But other times you can one shot whole features. Supabase MCP has been an absolute game changer. Database fluidity in development is really key. Trying to push migrations and generate types manually beforehand add so much time.
As someone who vibes codes web apps that are being used by people, they take way way longer than initially expected. The prototype can take one shot, and the full prod implementation can take weeks to months. If you aren’t able to stick it out it will die but if you are willing to learn how to prompt to integrate Auth, email, security, performance than you can create truly incredible apps with no coding experience. But it’s basically a new skill in and of itself. Claude Code already gets the prototype to final product gap down a lot because it’s so good and getting better but it still makes mistakes and needs lots and lots and lots of nudging and directing and reminders and writing docs and getting other agents to check work based on docs ect.
That’s fine but you would have to find the commit message to understand the context, whereas this is closer to an extended git blame per line with more metadata because if you store the conversation history locally you can link that line to its conversation context and that lives with the code not with the git commit.
At the moment I use just extensive docs to track decisions and business logic but its static and is constantly going stale.
I've built a CMS with Claude Code aswell and its working incredibly to create JSON proposals that my sveltekit website reads and turns into beautiful proposal pages. When a customer creates a booking for my mini-golf hire company they get emailed and they get their own booking hub where they can update their booking details, see the proposal when it comes through see any invoices ect.
The best part is and what i'm so excited about is we have created a daily business script 'npm run daily' that Claude Code runs and the script uses the business logic to move bookings along in the cycle by telling Claude Code what bookings have tasks. It will return, you have 3 bookings that need attention, run 'npm run get-booking [booking shortcode] THEN that script returns ALL data for that booking row from the db and it knows what task is needed to be done so Claude Code has all the context for that booking and it's prompted at the end saying NEXT STEP Claude code run 'npm run generate-proposal [shortcode] JSON output. (there was an example json output in there for claude to know the syntax ) Everything goes to an out tray in the admin web ui that i have to manually approve. I'm still in testing but I'm starting to realise that Claude Code can be an agentic platform for apps run from the CLI, like my automated crm assistant we've built.
It’s like something from a Guy Ritchie movie played for laughs. An Irish gangster gets taken down because of his habit of posting reviews of wherever he goes, and accidentally shows his reflection in windows and mirrors.
Anecdotal, but I’ve had an Apple Pencil replaced for free on the spot well outside of its warranty because it was defective and I hadn’t realised (don’t ask) I’ve had a Apple Store employee hand me a brand new iPhone when I brought in my phone that wouldn’t turn on one day - no questions asked. I’ve had friends with very similar cases. It’s always felt like “against policy” which might be their actual policy so it doesn’t get abused or something? They mentioned that the manager made an exemption for me with the Pencil. It’s very hard to track and very hard to advertise but their customer service in my experience has been exceptional.
I’d love to make a game, and attempted to make a start with something simple like Unity and Godot. For me it felt like implementing simple things that basically every game has (clicking and dragging objects, or moving a player character with arrow buttons, collisions) required a lot of coding straight up. Rather than boilerplate code blocks or something. It’s not like I expect there to be no code,I expected to use code to explain to the game engine complex ideas/systems, not figure out how to do the things that almost every game has. I’m sure there are inflexible No Code game engines but is there something in between?
Even if it released no carbon it’s still an environmentally destructive practice and weakens ecosystems that we rely on for food. It also does release carbon, I’m glad we are closer to knowing how much carbon - however I hope things like this don’t get upvoted as a bit of “See what’s the issue?” from people adjacent to climate skepticism.
Anyone who thinks we should just live in a “pure meritocracy” is a little bit delusional. Maybe they’ve played too many RPG’s where players have stats and you can sort by attribute and pick the best. If that was the case in the real world there would no job interview, just sharing statistics and picking the candidates with the highest numbers. Employers have to evaluate people on how they will perform BEFORE they perform, on subjective grounds. So for any number of reasons, probably subconsciously, people who are more unattractive, short, or possibly of a different race than the hirer can be perceived to be qualified but ‘not a good fit’. There are always ways around excluding people without it seeming discriminatory. That’s not to say there should be quotas, but to think there’s no issue because we should just pick the best candidate never stop to ask “how? How do you know they are the best?”.
Never before have I so often been convinced by the last thing I read about this topic on either side. Usually the smartest people in the room agree and eventually convince other people who are genuinely trying to seek the truth. In this case, I’ll see a HN comment laying out how it’s obvious it’s lab leak for xyz convincing reason and then a twitter thread from a virologist debunking everything with a thread about how the lab leak theory is ridiculous. At the end of it all I probably lean toward lab leak simply because it just feels like it makes sense (which is not great reasoning I know!). I don’t know if more evidence coming out will move the trenches that are already dug deep. People like me can be swayed still possibly…
Can anyone explain how much pressure and where it comes from for the “raise revenue at all costs” pre-IPO. Is it common for other companies to shoot themselves in the foot by layoffs and policy changes that alienate users? Just feels like no one was smart enough to say “we can try and raise revenue by doing this, but we could lose everything” - maybe companies just constantly forget how defiant and fickle the internet can be.