I always wondered why Splitwise is the go-to when it's a huge pile of crap app with stupid paid features for such a simple concept.

network effects and brand recognition :(

Now why and how the f-- do you own stwipe.com ?

They registered it like any other, I imagine.

A clear case of phwishing!

It could have been yours

While I empathize a lot with the issue of agents building super generic "vibe-coded" UIs, the webpage linked from the Github provides a Showcase with presets in the style of Notion, Vercel, Linear etc. which just change the color / theme not the actual design to match that of these companies which I thought was the point of the design skills?

I'm glad to hear people share this sentiment. I think the human quality will be part of what people will start to enjoy more about music now that AI has improved so much in terms of it's ability to produce highly polished songs.

This is my hope as well but I'm doubtful. I'd expect it's more likely that the generations growing up with this stuff will enjoy being able to annoy their elders with it and fully embrace the human-less aesthetic.

I find it strange the websites some of these HN articles are hosted on... going through the effort to write a post before getting a domain outside of .vercel.app is interesting.

> the effort to write a post

This post was pretty obviously written by llm. You can assume the llm also picked the hosting platform.

It’s a slop article based on a tweet.

Yeah I find this article hilarious. Especially since maybe less than 1 in 10 visitors will actually want to share the article? So 1 in 476 is actually pretty decent usage.

For which the other 475 get saddled with a bunch of extraneous downloads and invasive tracking.

The cohort that is concerned about this almost certainly runs some sort of a blocker.

What about the cohort that would care about it if they were aware of it? Or the cohort that cares about it but doesn't know such things can be blocked?

I think it's best to leave these educational matters to tech journalists or Internet commentariat. UX is the priority.

"The people who care enough about the quality of the water do tend to test it, and will refrain from drinking it if it's polluted. But the vast majority of people do not care enough to go through the hassle. So polluting that drinking water for our own interest is a no-brainer."

Yes, that’s pretty much literally what’s happening with a lot of industrial plants or pig farms, for example.

Note that I didn’t endorse this stance, but it’s very much the calculation made in many cases.

Haha I love the super out of touch takes on here sometimes. Not using WhatsApp or Facebook where I live absolutely introduces major difficulty to communication and even every day interactions where almost 100% of communication in various social groups is happening on these platforms. Should I ask them to move?

Where there's a will, there's a way. If your friends want to, they'll find you.

Maybe I am misunderstanding but I feel like https://www.jan.ai/ is similar and works great for me

hard agree, there's already "voice ai" companies that use the normal models and have this "interaction" engine on top of them to produce better results than I've seen in these demos. idk why people are impressed

Have you tried this task using an actual OCR model like Google Cloud Vision AI? I am not sure if this is what Gemini uses under the hood but multi-modal LLMs are not designed to extract text like this so it should be no surprise it's not good at it?

Google Cloud Vision AI is a specialized model built on CNNs frameworks which is part of the Interfaze architecture which is an hybrid so you get best of both worlds. Google cloud vision was pretty far behind other specalized models like PaddleOCR etc anyways so if you're looking for a pure CNN, check them out.

You can find the explanation and the comparison in the article, which we benchmarked pure CNN models, pure LLM models and a hybrid architecture like ours.

I don't think I've tried Google Cloud Vision on that particular image, no. In my experience, based on some tests from a year ago or so, Azure Document Intelligence impressed me the most in terms of OCR - out of the big three players: GCP, AWS and Azure.

I should retry the experiment because there has been a lot of progress since then and I could imagine that GCP improved there vision models since then.