Bealers

@0xCygaar Owning your own stack. Even if still using remote API for models, you keep context and orchestration local

What are the main reasons to keep using OpenClaw with all these features that Claude is shipping?

My first actually useful offline personal AI assistant case study. I pasted in three recent bookings i'd made to chat and it was told to add to calendar and do some basic research on the surrounding areas. It noticed a booking mistake then, in Gmail, drafted me an email to the place to change booking. Running on Mac Studio at home. Local model.

@BLUECOW009 coolify.io

what is the best solution to host a application that runs several things such a python, redis, sql, a webserver, an api, and I also need to buy the domain name

Oh no, poor Nonce Andrew.

@LiveOverflow “I’ve found the issue”

"Now I see the critical pattern. [...]. Now I have the complete picture. [...[ The real mechanism is now clear." When agent tells you it figured it out multiple times in a row, you know it has no fucking clue.

@SwiftOnSecurity Early phone adopters went on the website and a) lol, massive fixed width b) lol, images on mobile bandwidth

@intptr88 @voideng @SwiftOnSecurity tables and 1px shims.

@voideng @SwiftOnSecurity You did not need flash for this, at all. Just simple HTML with <img> and <map> or however it was called 26 years ago.

@moving_charlie Negative equity for the second time in my lifetime. Ouch.

“We paid £600,000 with Help to Buy for this London new build 4 years ago. We’d like to sell it. We realise we’re not going to make any money on it, but we’d like to get our money back.” Agent: “I’m very sorry I don’t think we can help.” rather than be the ones to tell them it will struggle to get £400k now. Today, first hand. 4 years of mortgage payments. Now add in a £200k loss. Utterly brutal. Renting really would have been cheaper. This is the harsh reality for many that all the average “House Price Indexes” and accompanying news stories are glossing over. If they’d been staying for the long term, it wouldn’t have mattered so much. But if you’re moving within 5 years? Never buy a new build.

@frankdegods I think they make fun of vibe coders who conflate being able to cobble together a prototype or local-only app with producing a production application.

The fact that people make fun of people starting out with vibe coding is so lame to me. Making a fitness tracker or daily journal is actually so cool. It’s actually the first step in you learning how to express your opinions with software. That’s the opposite of slop.

@pravijn Literally by first thought 🙈

wait, are you all entering your credit card details into a website frank vibe coded?

@ZanonGalt @thegameverse Cheers Captain obvious

@thegameverse The animations simply are not there. They were never created. OBVIOUSLY, if CDPR was to add 3rd person, they would create and add the corresponding animations.

This is why there’s no third-person camera in Cyberpunk:

@0xRodo Cool. I still have my Fewo collab pair. Maybe one day I’ll wear them.

BREAKING: Nike has sold RTFKT According to the article below, Nike sold RTFKT on December 16, after previously shutting it down. The terms of the RTFKT sale, including the buyer, were not announced. RTFKT NFT collections are all up 100%-1000%.

@moonpay @icobeast @inversebrah w ratio

@icobeast thanks for sharing without the ticker your actually goated for that. it’s probably the coolest thing I've seen in my life, you posted a crypto chart on crypto twitter but you cropped out the name of the token. heroic. I love solving riddles thats fire bro thanks

This could be the greatest cup and handle chart ever created

Mood but unironically

Homelab upgrade. Repurposing my old validator with an external rtx3070 and 64gb of RAM

@natealex @MrEdogtagnft Apart from mine, obvs.

@MrEdogtagnft i told a few friends early on and thought it led to some fun discussions about the metagame. im sure some others will be disappointed to learn that their 100 bubbles will not turn into an xcopy 1/1 though

sorry for the dramatics but i think it was a way for me to procrastinate sharing my findings the oversimplified TLDR is that if you have the most bubbles, you will win the reason i find this interesting is because it is no longer really a game of chance, but rather an accumulation game. if you have 30 bubbles today, you will not win in 10 years. to take it a step further, if you have 30000 bubbles today, you will not win in 10 years [both examples above are assuming the holder is passive and not planning on making future moves beyond holding and praying] a team that can gather more than half of the bubbles can guarantee victory in 10 years. a 51% attack, if you will. this suggests that 51% of the bubbles are worth an XCOPY 1/1, and the other 49% are doomed to pop with no chance for glory the reason for this is because the pop selection is by wallet rather than by bubble. this is a key distinction. anyone with 10 bubbles will be long popped and dead before anyone with 1000 bubbles is at risk of being 'out' at this point some people are rolling their eyes saying "DUH" while missing the point entirely, so here's a more specific example: in a proportional system (selection by bubbles), someone with 60k of 100k total bubbles would have a 60% chance to win. intuitive. in a by wallet system, someone with 60k of 100k total bubbles would have effectively a 100% chance to win one more example, since i know there are at least 3 people still reading and struggling to understand it. the last 2 players are standing, one with 9 bubbles and one with 1 bubble. the underdog would have a 10% chance of winning in a proportional system. instead, they need to win 9 coinflips in a row (2^9, 0.195% chance) to beat the favorite you can see how this scales dramatically when you get into thousands of bubbles surplus this is a lot of words to say it works differently than most people think it does. i still find the experiment interesting because gathering 51% is non trivial given liquidity. that is, in essence, "the game". also worthy of note is that you do not need to pool together at this point. there are 130k holders and less than 700 are selected daily for a pop. this means shadow groups may exist until the great consolidation a few years from now i really enjoyed thinking through the dynamics and trying to understand how the metagame could play out. hopefully others will as well. happy to discuss more in comments may the odds be ever in your bubbles favor (although odds have little to do with it)

Pinch, punch, first of the month (no returns!)

Elf and safety bruv

this is absolutely terrifying - also goes some way to explain the source all of these chatgpt-4o 'romances' we've been reading about.

Shaw has entered the chat