Pricing page is hidden behind a registration form. Why?

I also wanted to see how/if it handled semantic data (schema.org and Wikidata ontologies), but the hidden pricing threw me off.

Thanks for the feedback. We are definitely not trying to hide it. We actually do have pricing listed in the API section regarding the different operations, but we could definitely work on making this clearer and easier to parse.

We are simply in an early stage and still finalizing our long-term subscription tiers. Currently, we use a simple credit model which is $1 per 10,000 credits. However, every account receives 50,000 credits for free every month ($5 value). We will have a dedicated public pricing page up as soon as our monthly plans are finalized.

Regarding semantic data, our JSON extraction endpoint is designed to extract any data on the page. That said, we would love to know your specific use cases for those ontologies to see if we can further improve our support for them.

It only works for languages and frameworks that are already in the training data (duh). It still is mostly useless when you need to create something from scratch in an unstable language.

That, and you can’t also get the amazing results if you’re poor or have bad internet.

Good thing almost all of programming falls into the former. Most of the economy runs on well defined languages. Billions and billions of dollars.

Not true. I built some tools in Hare, which almost certainly isn’t in the training data to any significant extent. It was more work than having it build Go or Rust, but it got it done. It had to curl the docs a fair bit.

Opus 4.5 and update your priors. This was certainly true >6months back and is no longer the case

We are using the latest stuff. Our experience is still not great.

Why do you guys always assume we don't as though the oldest models are easy to use accidentally

I have a feeling that the HN hypebeasts have a lot of overlap with the folks that previously used to copy/paste blindly from StackOverflow.

It’s an easy deflection. Dismiss any opinions because you’re using it wrong or not the latest.

Good for anything >= 1 month old.

Use other nonsense fear inducing argument in the mean time, continue gathering gobs of VC money, get your bag, continue till the bubble pops.

In all fairness, and putting hype and anti-hype aside, I’m really interested to see the actual value of LLM/agent services after the VC money subsidies dry out. Would people we willing to pay for services at 10x the current price?

I read the same exact thing 6 months ago.

Yeah bro thanks for the tip and few shillings to you good sir. I was here still using GPT 2 because they said GPT 3 might be too dangerous.

> The so-called web of trust is meaningless security theatre.

Ignoring your comment’s lack of constructive criticism, I’m going to post this meaningful implementation that an excellent cryptographer, Soatok Dreamseeker, is working on: [1].

You may also search for his posts in this HN thread, his nickname is “some_furry”.

[1]: https://github.com/fedi-e2ee/public-key-directory-specificat...

Keyservers already “solved” this problem without needing federation because we only needed one keyserver anyway. Federating them isn’t going to do anything. Web of trust is a broken system that sounds super cool until you try to really use it. It has so many flaws that there’s really no way to revive it. Keybase tried to do something about it and also failed.

To be clear, this is not Web of Trust. It's using Key Transparency as a means to distribute public keys more securely than TOFU.

If people want to build WoT on top of ny design, I won't stop them, but it's not a goal of mine.

Keybase was doing great until it got acquired by Zoom and people felt uneasy about the implications, IIRC

And where most of CPU time will be wasted in, if you care about profiling/improving responsiveness.

At that point you're just io bound, no? I can easily parse json at 100+GB/s on commodity hardware, but I'm gonna have a much harder time actually delivering that much data to parse.

What's a better way?

> I would never pay for search.

I’m curious, why not? The amount of time and frustration saved seems insane, which seems to be a net positive, unless you can’t afford it of course (which is a good argument, but I’m not seeing any in your comment).

Because there are free and privacy respecting alternatives. I use DuckDuckGo, but you can also use meta-engines that anonymize traffic like SearXNG. None of these require accounts or payment methods that effectively de-anonymize you. Finally, if you really like Google for some reason but don't want the side effects, use an adblocker and a VPN.

My headcannon is that it’s a money thing, and that there must be intense lobbying by the internet surveillance manufacturers (like Thales).

Really? How did you manage to get past the Outlook blocks? Those were always the problematic ones for me.

After setting up dkim, dmarc, etc. I've had no problems in the past decade except for one person using aol. I told him that his email was broken and if he wanted to receive my email he needed to fix it. I don’t count such things as deliverabilty problems, but as receivability problems on the other end.

I’ve never sent any kind of bulk email and I suppose my host has a good IP. Everything I do depends critically on email deliverability, often to addresses I’ve never sent to before, so if I had a problem I would certainly know about it.

It’s like Vim, you learn it once, and you keep using it forever once you’re used to it.

I’m so thankful to see a flake.nix file in every single cool project on code forges.

To be honest I don't know what to say, you can use nix in many ways, and you don't even require to know the language.

The easiest entry-point is to just use it like a package manager, you install nix (which is just a command...) and then you have available the whole set of packages which are searchable from here: https://search.nixos.org/packages

nix-shell is just to download&add programs temporary to your PATH.

I don't feel that this is harder than something like "sudo apt install -y xxxxx" but for sure more robust and portable, and doesn't require sudo.

If at some point you want to learn the language in order to create configurations or packaging software, it may require to check a lot more documentation and examples, but for this I think it's pretty straightforward and is not harder than any other package manager like aptitude, homebrew or pacman.

Nix with Flakes never randomly break, I still have projects from 3 or 4 years ago that I can still run `nix build` and getting it running. Yes, if you try to update the `flake.lock` this may introduce breakages, but this is expected if you're pining `nixos-unstable` instead of a stable branch.

I’m not sure what you mean by “a nightmare to set up”. You install Nix on your current OS with the determinate.systems installer, and you enter `nix run github:johndoe/project-containing-a-flake-dot-nix-file` to try out the project and have the full reproducible build taken care of by Nix.

Sure, installing packages the proper way requires a little bit more setup (Home Manager, most likely, and understanding where is the list of packages and which command to build to switch configuration), but as trivial as other complex tasks most of us hackers are capable of doing (like using `jq` or Vim).

How do you plan on maintaining the tool if you’re out of funds to run Claude Code?

Ghostty already supports shaders and effects like this.

It can only apply shader(s) to the current frame I think. To produce the crt ghosting you'd probably need access to the previous frame (not an expert).

I've tried the shaders in the following repo with ghostty. They definitely work. I ended up keeping a cursor trail shader. https://github.com/0xhckr/ghostty-shaders

Yes, correct! If you check out https://ghostty.org/docs/config/reference, the iPreviousCursor is available, so it can be used against the iCurrentCursor to produce a fading effect. But I think the entire previous framebuffer isn't there (yet).