tleyden5iwx

Can they change the name to Proof Of Corncept?

In my tests it's not working though: https://github.com/RecapAI/Recap/issues/18

Thanks for sharing! Here's another approach that might interest you:

https://github.com/RecapAI/Recap

I believe it uses ScreenCaptureKit.

AGI will happen, but we need to start reverse engineering the brain. IMHO LeCun and Hawkins have it right, even though the results are still pretty non-existent.

In the meantime, 100% agree, it's complete fantastical nonsense.

I built Arty: https://github.com/vibemachine-labs/arty

It's an open-source iOS voice agent that uses the OpenAI Realtime API (bring your own key).

Current connectors: Hacker News (check demo in readme!), Google Drive, GitHub, and web search.

I got frustrated with the limitations of the OpenAI Realtime Voice iOS app—for example, it can't even connect to Google Drive.

Arty is self-contained except for the OpenAI model and any third-party services you connect to. Uses local tools—no MCP support yet.

If you'd like additional connectors, feel free to open an issue.

Yeah but there are a lot of vibe engineers, and we care about the code, because we have to own it

Feature request: add a Github Action so I can generate a codemap for my repo and throw it on my README. Then update it when major PRs change the codemap.

Agree with LeCun that current ai doesn’t exhibit anything close to actual intelligence.

I think the solution lies into cracking the core algorithms used by nature to build the brain. Too bad it’s such an inscrutable hairball of analog spaghetti code.

API clients can already easily identify the user during the authentication process. What’s currently “missing” is a way for an API client to pay a service via the API call “on the fly”, without having a credit card on file or some prior account balance.

Maybe a better example is an image generation api. Your agent chooses one based on its research, and without any account or credit card on file, it buys you an image using a x402.

Looks really useful! Do you happen to have a gallery of apps using it? In particular, I'd like to see how desktop or mobile apps handle the oauth flows.

I think the purpose is so that your agent can do things like buy airline tickets for you. Using x402 it doesn't have to go through a typical credit card checkout process, which might have a lot of safeguards that make sure there's an actual human on the other end (captchas, etc)

I don't think like a C programmer, my problem is that I think like a Java/Python/Go programmer, and I'm spoiled by getting used to having a garbage collector always running behind me cleaning up my memory poops.

Even though Rust can end up with some ugly/crazy code, I love it overall because I can feel pretty safe that I'm not going to create hard-to-find memory errors.

Sure, I can (and do) write code that causes my (rust) app to crash, but so far they've all been super trivial errors to debug and fix.

I haven't tried Zig yet though. Does it give me all the same compile time memory usage guarantees?

It looks like RethinkDB is relying solely on revenue from training and support. Is that even feasible for a product like theirs?

I've never used their product, I've only looked at their website and was extremely impressed. I remember their bullet points were around clean architecture, testing, performance, etc, all the stuff that engineers/devops folks care the most about, so they can avoid getting paged in the middle of the night to rolling reboot every node in their ${name-of-distributed-database with-scaling-issues} cluster.

But, unfortunately, it seems like if you depend on revenue that derives solely from training and support, you're best bet is to make a product that has:

* Awful documentation (hence the need for people to pay for training)

* Full of bugs and performance issues (hence the need for people to pay for support)

Since from the looks of it, RethinkDB was pretty much the polar opposite of this, it seems like they were essentially a victim of their own perfection.

I wonder if any RethinkDB users out there paid for support just to try to keep RethinkDB alive?

Also, I wonder why RethinkDB doesn't change their revenue model if it's not working for them?

It's more like "one of the memcached contributors", rather than "all the memcached folks".

Couchbase was a merger between NorthScale and CouchOne. The CouchOne founding team were all CouchdDB project founders: Damien Katz, J Chris Anderson, and Jan Lehnardt. NorthScale was founded by Steve Yen, who was an Entrepreneur in Residence at Accel, and Dustin Sallings, who was a committer on the Memcached project.

The first product that NorthScale released was called Membase, and the point of it was to help companies run large memcached clusters without having to think very hard. Couchbase later evolved into a hybrid of Membase and CouchDB that can speak the memcached protocol but also has some of the JSON document database properties that CouchDB has.

[Full disclosure: I'm a FT Couchbase mobile developer]

Agree 100% -- Parse deserves huge kudos for that move.

Ex-Parse customers should definitely check out Couchbase Mobile, which has some functionality overlap with Parse and is already open source with several repos on github:

https://github.com/couchbase/sync_gateway https://github.com/couchbase/couchbase-lite-ios https://github.com/couchbase/couchbase-lite-android

At least you won't have to worry about getting "Parsed".