Was also wondering this the other day. Every harness I’ve looked into seems to be monolithic. I would really prefer specialized profiles/configurations instead of loading up every skill and MCP server imaginable into a big stateful works-on-my-machine mess.
I just want to run `pi --profile=./gamedev.toml “build a side scroller where you’re a turtle”` and have it do the thing.
There’s some likelihood the parent comment is astroturfing (not saying they are, just that it’s a possibility).
There’s less of a chance mine is, since I don’t offer an alternative and my review is more critical.
The SVP has the right idea as a leader. I’ve often framed this as adding “regression tests” to our process.
Ask yourself and the team, “what measures can we put in place to save us from ourselves and decrease the likelihood of the failure reoccurring?”
I like this idea. A simple bus route that delegates to waymos for the last mile seems like the best of both worlds. And maybe easier to develop public transit routes because they can be simpler.
Also hope it can applied to more rural areas eventually too, where driving is the only realistic option. DUIs, accidents, and whatnot are way higher than they ought to be in these places because there’s few or no alternatives. And you’ll never convince everyone to give up their imbibing.
For another datapoint, I took the plunge on a pair of xreals and trying to do real work in them felt like a joke. They’re not comfortable enough to wear as normal glasses, so you end up using them more like you would a VR headset. And at that point, they’re just a semi-transparent low-res screen that’s never quite in the right spot.
I feel like many of the comments are skeptical or dismissive but this seems like a really clever idea.
So many classic textbooks use pseudocode to get the gist of an algorithm across, so this isn’t an unprecedented level of abstraction to be working at.
It makes sense that there would exist a middle ground of abstraction between an English essay and actual running code.
Another interesting thought about this: it’s almost like a quick and dirty transpiler for your own programming language design ideas. I’ve toyed with writing my own ”perfect language” and this seems like a quick and dirty way to test out the ergonomics of various syntaxes and concepts via pseudocode.
Would be nice if there were a hn feature, userscript, or plugin to just filter comments from new accounts. Bonus if there was some sentiment analysis or llm-based analysis to filter out unsubstantiated inflammatory comments too.
Random naive thought, could I use this to backup and version all of my user data? E.g. just point Lore at my ~/Documents folder and call it a day? I'm sure there's some scalability constraint but I have music projects, game projects, markdown notes, photos, videos, etc. and have thought in the past it'd be convenient to treat all my files like a monorepo.
Thanks for the level-headed response. Agreed, "blame the kings" is also an overly simplistic take. There's rarely a single entity to blame. Just wanted to bring some nuance to the hate getting thrown around here.
Redirecting frustration towards the systemic issues that drive perfectly normal people to work for a terrible company like Meta seems more productive to me.
I hate what Meta does just as much as the next guy, but shaming friends who feel like they have no better employment option doesn't seem like the best way to make a difference here.
This is a pretty uncharitable perspective. Most folks I know working at Meta or Amazon aren’t morally bankrupt. They just have kids, debt, poor parents with health problems, etc. They work at Meta to support their loved ones. And it’s not like you can walk onto the street and just wave down a morally superior job with similar pay and benefits. Blame the tech oligarchs, not the workers.
This misses the point that parent was making. The conversation shouldn’t be “move your data to countries you can trust”. It should be “use protocols that don’t require trust in the first place”.
I’ve had the same thought and would love this. MacOS shortcuts are too deeply ingrained in my fingers.
But every attempt of mine to make Linux shortcuts Mac-like has had too many sharp edges to be useable.
Toshy didn’t seem to work well with Wayland and felt heavy.
Probably the best so far I’ve found has been keyd and custom configs for your most used apps.
A community effort might get us there. Distribute the hours of tinkering across many passionate users instead of everyone doing it in a vacuum.
It’s kind of wild to read through these comments and realize hn is still riffing on the same ideas. Is it e2ee? Does it run on Linux? Who would pay for something you can slap together in a weekend with a few bash scripts?
Really highlights this community’s values, skills, and blind spots.
Also a bit of a bummer that the privacy and open source situations today are even worse in many ways.
I did webdev for a long time, so yeah. If you want the story, I was looking into guix on asahi and ended up on https://www.asahi-guix.org/ which didn’t load anything, so I checked the page source and noticed the element.
Noticed a suspicious element called give-freely-root-bcjindcccaagfpapjjmafapmmgkkhgoa in the chrome inspector today.
Turns out about a month ago, the popular open source [JSON Formatter chrome extension](https://chromewebstore.google.com/detail/json-formatter/bcji...) went closed source and started injecting adware into checkout pages. Also seems to be doing some geolocation tracking.
I didn't see this come up on hn, so I figured I'd sound the alarm for all the privacy-conscious folks here.
At this point, I feel like browser extension marketplaces are a failed experiment. I can just vibecode my own json pretty-printer extension and never deal with this problem again.
Good points. What local models have you found work best for your use cases? I feel like if we get to opus 4.6 level intelligence running on local hardware, we’re in the clear for a lot of day to day use cases.
Definitely interested! I've reinstalled Fedora Asahi Remix several times on my old M1 after fiddling my way into a broken state. NixOS sounds like a tinkerer's dream but getting started is a bit intimidating.
This is smart and a good first step. Everyone can't be trusted to do the security dance flawlessly, though. We need sane defaults. Least privilege by default for 3rd-party code. Deno's headed in the right direction with this. But I think the solution needs to exist deeper in the stack. The surge in popularity of `curl -fsSL https://my-cool-ai-starup.ai/install.sh | bash` style installers is particularly concerning to me in this regard.
Totally agree with all of your comment. To solve for structured data instead of everyone writing parsers, I’ve enjoyed using nushell (not affiliated, just love the idea). https://www.nushell.sh/
It’s like powershell but not ugly and not Microsoft.
Huh? Zed's vim emulator is one of the best I've used in an IDE. Saying it's an afterthought feels disingenuous. I'm pretty sure some of the core zed devs are big vim enthusiasts.
I'd never heard of this effect before but it reminds me of a feature I've aways wanted for ios which is, in order to unlock your phone, you need to type in what you're planning to do (and maybe for how long), so you can refer to that or get notified when your memory inevitably blanks and you get sidetracked.
Is there an SRS app or Anki feature that takes into account your reaction time when answering a card? I've wanted to use Anki for things like speeding up mental arithmetic but Anki doesn't seem to have a feature for measuring and plotting progress on response times.
Does the local underscore variables feature solve this? Or the approach outlined in the plots tutorial? IMO, not allowing redeclaration is more valuable than supporting this use case. A slight paradigm shift away from your example gives you the significant benefits of a reactive environment with fewer edge cases/quirks. I'd much rather have a notebook error out instead of silently overwriting a value. You save so much time debugging.