Not to be confused with John von Neumann

I didn't even know he was sick.

Does the writer understand that for every developer who burns all tokens, there are many people who subscribe just to join the AI revolution, but only ask a couple of questions a day?

No. Large co I work at everyone is like running at least 3 concurrent Claude sessions all day every day. Talking to friends in other companies it seems the same.

Big difference between professional deployments and personal ones.

Europe has laws against misuse of personal data. What circus?

In Europe, it’s not (and should not be) a “standard” thing.

It shouldn't be but it is. Otherwise we'd find most European sites don't have consent popups. In reality we have tonnes, many now requiring payment if you want to reject trackers.

> Perhaps you are right but what the duck does sales funnel mean in a suicide prevention website?

They’re not governmental, but an organization dependent on subsidies and to get them, they have to show they’re relevant.

In the EU, it’s illegal

It’s “just” a major corporation collecting very personal data.

Europe has laws against misuse of data from visitors, so it’s generally safe to visit such websites.

The fact that this is on hackernews means this case is the exception, not the rule.

The article started with showing how complex the frontend is. And then moves the complexity to the backend, with tools that aren’t well supported and putting extra load on the server. For some applications, this is a good solution, for most, however, it’s not.

> with tools that aren’t well supported

This is an odd thing to say about the backend choices. Spring is ... well let's just say there are a few folks in and around that area.

I think it's totally fair, but I would assume that most Spring projects make significant DX tradeoffs compared to a full JS stack or serve an API rather than html. 1. Spring + Handlebars: You can either write the html template in a string loosing syntax highlighting and other LSP stuff, load it from a file loosing colocation. 2. Handlebars + webcomponents. They simply bundle all the web components into a single file, which breaks down when they get large and you don't need every component on every route. 3. Tailwind: Looking online you can get it working with spring boot, but the route chosen here is a script to run the cli, which again means every route ships every tailwind class used anywhere.

> putting extra load on the server

This is precisely where the publisher has the most control over the user experience. Putting load on the browser makes a user's experience dependent on their hardware & software stack.

Part of what makes a good user experience is is working nicely with the users hardware and software stack and that's much easier on the client. The user would like the website to have native scroll physics, respect their system preferences, react to changing window sizes and with different input methods, screen readers and so on.

If the key to a good user experience was server side control, than the hallmark of a good website would be being an RDP stream and prefers-color-scheme wouldn't be a CSS feature but a HTTP header.

> The user would like the website to have native scroll physics, respect their system preferences, react to changing window sizes and with different input methods, screen readers and so on.

Yes, these are features implemented by every popular browser via HTML & CSS. Fancy front-end work frequently breaks these features.

> key to a good user experience was server side control

I think you're reading my comment from an extremely front-end perspective. I simply mean that where possible it's better to render pages and do logic on the server side versus on the client side. The same HTML + CSS is generated either way, the only question is what % of the work is done by the user's device vs what % is done in a data center.

Haven’t you heard about Codex?

its an SKU from OpenAI's perspective, broader goal and vision is (was) different. Look at the Claude and GLM, both were 95% committed to dev tooling: best coding models, coding harness, even their cowork is built on top of claude code

I'm not sure how this makes sense when Claude models aren't even coding specific: Haiku, Sonnet, Opus are the exact same models you'd use for chat or (with the recent Mythos) bleeding edge research.

Anthropic models and training data is optimized for coding use cases, this is the difference.

OpenAI on the other hand has different models optimized for coding, GPT-x-codex, Anthropic doesnt have this distinction

But they detect it under the hood and apply a similar "variant", as API results are not the same than on Claude Code (that was documented before by someone).