Comments URL: https://news.ycombinator.com/item?id=49099293

Points: 3

# Comments: 1

It has a pro over documented examples in that they are _guaranteed_ to be correct.

Next month, we're introducing new weekly rate limits for Claude subscribers, affecting less than 5% of users based on current usage patterns.

Claude Code, especially as part of our subscription bundle, has seen unprecedented growth. At the same time, we’ve identified policy violations like account sharing and reselling access—and advanced usage patterns like running Claude 24/7 in the background—that are impacting system capacity for all. Our new rate limits address these issues and provide a more equitable experience for all users.

What’s changing: Starting August 28, we're introducing weekly usage limits alongside our existing 5-hour limits: Current: Usage limit that resets every 5 hours (no change) New: Overall weekly limit that resets every 7 days New: Claude Opus 4 weekly limit that resets every 7 days As we learn more about how developers use Claude Code, we may adjust usage limits to better serve our community. What this means for you: Most users won't notice any difference. The weekly limits are designed to support typical daily use across your projects. Most Max 5x users can expect 140-280 hours of Sonnet 4 and 15-35 hours of Opus 4 within their weekly rate limits. Heavy Opus users with large codebases or those running multiple Claude Code instances in parallel will hit their limits sooner. You can manage or cancel your subscription anytime in Settings. We take these decisions seriously. We're committed to supporting long-running use cases through other options in the future, but until then, weekly limits will help us maintain reliable service for everyone.

We also recognize that during this same period, users have encountered several reliability and performance issues. We've been working to fix these as quickly as possible, and will continue addressing any remaining issues over the coming days and weeks.

–The Anthropic Team

Comments URL: https://news.ycombinator.com/item?id=44713757

Points: 609

# Comments: 705

But does anyone have thoughts on the security aspect. Getting people used to just running code like this that has full access to the system is slightly concerning.

On the other hand it’s no different than installing npm packages

The TikTok algorithm is good at figuring out what you do. Not what you say. So the content will be a lot more engaging.

As you scroll, it learns what you like and generates more videos.

https://www.home-assistant.io/voice-pe/

Comments URL: https://news.ycombinator.com/item?id=42703839

Points: 1

# Comments: 0

The forms can be dynamically generated within the workflow, and then call back with the response.

The docs still need some work and I plan on adding some video demos, but here it is so far.

https://humaninput.app

I will be supporting custom domains, so I could support `/ideas` only for those sites.

Other options include redirecting/rewriting your personal `/ideas` path to this product. Or, offering subdomains (e.g. moshe.example.com/ideas)

I'm building a platform to do this, so I can share my ideas.

My page is at https://ideas.moshegrunwald.com

You can see more details about this specific idea [here](https://ideas.moshegrunwald.com/ideas/i_mos2)

I would love to know if there is a better way to do this.

Now with LLMs it’s simple to extract structured data from emails.

I built [Orderling](https://orderl.ing) that is basically a CRM for your orders. It uses OpenAI api to extract the order information and automatically adds it.

For a long time, I’ve had this idea to scratch this itch and build something that can solve this. The primary blocker was that it’s not practical to manually add everything to such a system. There is nothing like Plaid for merchants, where I can access order information. Building out native integrations for each merchant is too much work to do for an unproven idea. Even scraping the data from emails is a lot of work. Each merchant has its own email format, and could even have multiple email formats (A/B testing etc.).

What changed now are LLMs. Now, it’s easy to extract structured data from unstructured HTML. When an order comes in, I send it to the OpenAI API with an expected data schema, and it’s able to extract the information.

I built [Orderl.ing](https://orderl.ing) to solve this problem. You can think of it as a CRM for your orders.

What is different about Orderling over other products is that the focus is less about tracking the order coming to your door. It’s about tracking the order after. When you confirm that you actually got the item, you mark it as received. You can then decide if you want to keep it, unsure about returning, or want to return it. You can then easily come back to see orders that you still need to decide about, or that need to be packaged up and shipped. You can also quickly see a list of orders that are close to the return-by date, so you can make the final decision, or actually return it.

There are still lots of features on the roadmap. I’d love to get feedback from users to see if anyone else finds it helpful, and to see if there are any other features that would be useful.

Comments URL: https://news.ycombinator.com/item?id=39251684

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=39205840

Points: 3

# Comments: 0

> Apple updated its App Store guidelines today with a change to its traditionally strict restrictions around push notifications. Apple has long banned apps from using notifications for “advertising, promotions, or direct marketing purposes,” but that changes today. Apps can now send marketing notifications when “customers have explicitly opted in to receive them.” Users must also be able to opt out of receiving the ads.

> The change follows a couple incidents over the past two years in which Apple bent its own rules by sending out push notifications that read a lot like ads. Since other companies’ apps could be banned or have their push notification privileges revoked for that behavior, the moves were criticized as another example of Apple getting away with special treatment because it controls the platform.

https://www.theverge.com/2020/3/4/21165087/ios-apple-push-no...

> The flight software is written in C/C++ and runs in the x86 environment. For each calculation/decision, the "flight string" compares the results from both cores. If there is a inconsistency, the string is bad and doesn't send any commands. If both cores return the same response, the string sends the command to the various microcontrollers on the rocket that control things like the engines and grid fins.

https://space.stackexchange.com/a/9446/53026