Different scenario but it reminds me of when Missouri prosecuted a reporter who found that teacher's SSN numbers were exposed in the HTML of a webpage
> "Parson described the journalist as a “perpetrator” who “took the records of at least three educators, decoded the HTML source code, and viewed the Social Security number of those specific educators” in an “attempt to steal personal information and harm Missourians.”"
Using the plan mode in cursor (or asking claude to first come up with a plan) makes it pretty good at generic "how can I improve" prompts. It can spend more effort exploring the codebase and thinking before implementing.
Great comment. I'll add that despite being a bit less powerful, the Composer 1 model in Cursor is also extremely fast - to the point where things that Claude would take 10+ minutes of tool calls now takes 30 seconds. That's the difference between deciding to write it yourself, or throwing a few sentences in Cursor and having it done right away. A year ago I'd never ask AI to do tasks without being very specific about which files and methodologies I want it to use, but codebase search has improved a ton and it can gather this info on it's own, often better than I can (if I haven't worked on particular feature or domain in a few months and need to re-familiarize myself with how it's structured). The bar for what AI can do today is a LOT higher than the average AI skeptic here thinks. As someone who has been using this since the GPT4 era, I'd say that I find a prompt about once a week that I figured LLMs would choke on and screw up - but they actually nail it. Whatever free model is running in Github Copilot is not going to do as well, which is probably where a lot of frustration comes from if that is all someone has experienced.
Had to scroll way too far to find a comment like this. Unfortunately HN famously doesn’t understand what the average consumer wants. The Dropbox comment comes to mind.
I think it's the amount of embedded videos. I appreciate them because it really illustrates what each feature is, but they lag the page. If I recall this can be solved with animated webp images which are more lightweight than full on <video/>s. Or maybe just not autoplaying them
If the flight is delayed by 3 hours, you will get a refund if you cancel. This is great if the delay is long and there is a flight on a competing airline that would let you get out sooner.
You can set email notifications (or any app) to deliver quietly. I made this change years ago for email and some social apps. You can go through the notifications when you check your phone but aren't distracted with it vibrating on every email.
Right, I was hoping not to use them at a previous company that used AWS. One day we got hit by a DDOS (trying to get us to pay to stop it). Even with AWS WAF costing $0.60 per million requests, we ended up paying around $10k in WAF rules to block the attack. Yes, hundreds of thousands to millions of reqs/sec. Luckily the attacker had their entire botnet using an Accept-Language header from a specific (non-english) language, which made it an easy rule target. If it wasn't for that, I'm not sure what we would have done. Would actually love to hear what others do, I want the answer to be more than "use CloudFlare", but it's the only option I've found since then.
I've noticed this too. Why does amazon have aboutamazon.com and Google have developers.googleblog.com? They literally have their own .google TLD but still choose this weird domain.
Same with local governments. They love something really random like <countyname>proptaxpayment.org instead of treasurer.<countyname>.gov. It's exactly the kind of domain you are told to watch out for, but actually legit.
This always results in a ton of hacky JS to detect how the element should reposition itself if it overflows the screen (depending on the content and screen size)
I haven't tried Claude Code yet, but I extensively use Cursor and find it can do similar work (which makes sense). I think it's good you led with the length prompt/spec, which will maybe help the "AI can't do ____" crowd grasp that it really is a skill issue when they can't get any good results from LLMs. (Unfortunately it cannot help the "well, this isn't actually useful or making an impact" group)
Regarding some of the comments here: I found the article style fine, and I even like the "follow my journey" style writing as it helps the reader understand the process you went though. That kind of engineering and debugging workflow is something I enjoy about this industry.
Opus is too expensive and I find it goes way off the rails often (just writing way way too much. Maybe that could be controlled with a better prompt on my end). Sonnet gives me more realistic code that isn't too overengineered.
I don't either but unfortunately Cursor is better than all the other plugins for IDEs like JetBrains. I just tab over to cursor and prompt it, then edit the code in my IDE of choice.
I use one of these for two 4k monitors, sound, ethernet, mouse, webcam, AND charging. It's amazing having one cable to plug in when I work from my desk. Unfortunately requires one of those $400 docks though.
- Very little users know this is in the settings app (there are thousands of other settings that are two steps away, how many can you name?). How many people have ever clicked on their name in the settings app to see what is inside, despite it being the first item? Apple gets a cut of every transaction so it's in their interest to make it hard to stumble upon the subscription list by accident.
- Very little users understand that Apple is billing them. They buy something from X Corp and expect X Corp to be able to help them, the idea that everything is really through Apple is a foreign concept.
Well, the "one click" cancel is hidden deep in the settings app - and when customers contact us asking to cancel, they don't like to hear that we cannot cancel or refund them from our end. (Apple doesn't even provide a way to look up the customer. Most people don't understand that Apple is actually managing the entire billing process)
Webview apps are limited to 60fps so they feel sluggish on iOS. Running them in safari with the 120fps dev option enabled they can feel like native, but then feel laggy when bundled into their own app.
I've looked at a bunch of the popular JS libraries I depend on and they are all the same story, hundreds of open PRs. I think it's just difficult to review work from random people who may not be implementing changes the right way at all. Same with the project direction/roadmap, I'd say the majority of open source repos are like that. People will suggest ideas/direction all day and you can't listen to everyone.
Not sure for VSCode, but for .NET 9 they claim: "There were over 26,000 contributions from over 9,000 community members! "
This sounds kinda like what they did when they were forced to allow outside payments in the US. It could only be one link, with a big scary warning, and a 27% cut. They "comply" with the ruling by making another alternative deal that nobody would ever take. Fortunately this backfired in the US and they were actually forced to get rid of all the restrictions in May.