janrakete
janrakete on Hacker News
Posted by
janrakete on Monday September 07, 2026 @04:43PM
from the show-hn dept.
https://github.com/janrakete/healthcore-bulp.io
Ever since my father passed away, my mother has been living alone in our family home. She’s now over 70 years old. I live in another city, and so does my sister. Of course, we call her regularly, but at the same time, she doesn’t want to feel like she’s being watched. She’s still too “young” for that. So our best way to check on her is to see when she was last online on WhatsApp. But in the...
It reminds me of how, back in school, we used to replace the sound that played when Windows shut down with fart sounds.
That's a nice idea, but it's probably quite easy to do in ChatGPT as well.
It gets interesting with words whose translations into another language have a completely different meaning.
Example: "Gift"
German: Toxoid
English: Present
Danish: Marriage
A simple translation can therefore backfire. Maybe your tool could warn users about this?
For my alternative rock band and our groupies (there were only a few of them), I programmed an online community aimed at punks, just for fun. I was in my early 20s at the time. The platform later grew to 100,000 members and provided me with a good monthly income.
That was a coincidence. Or rather, I was naive and started the project without a business model. But what I realized later was that, without even knowing it, I had connected two things: a technological trend (social networks) and a target audience with a specific need.
When starting your own projects, it makes sense to focus heavily on your target audience.I often check out global trends or the opinions of futurists and try to develop ideas from them. But I always try to avoid creating a one-size-fits-all solution. I’d rather have a small but loyal target audience than a large but disloyal one.
Right now, I'm working on https://github.com/janrakete/healthcore-bulp.io. For now, it's aimed at a small target audience (tech enthusiasts who have to care for people, such as their parents). But I have no idea if it will be a success.
My advice: Keep at it, keep trying. Eventually, you’ll have enough code and experience under your belt that you’ll be able to whip up a solution in no time when the next trend comes along.
I use Remarkable 2. It's not exactly an e-reader, but more of a notebook and sketchbook with a paper-like surface. It is, however, quite expensive. It's definitely too much money just for reading.
I think it’s generally a good idea to use a standalone, local solution. I also like the approach of combining multiple models with different focuses under a single interface. On the other hand, I usually use very specialized models (for programming, for example) and rarely end up generating music or a video on the side. So I can’t think of a reason to use the system right now.
What target audience do you have in mind?
Well, that's a good idea. Do you plan to add more languages? And what's your business model? If everyone puts together their own personalized feed, you'll have to generate a separate podcast for each person. That could get expensive.
It would be even better if your tool could predict every bet correctly. If everyone could win every bet, then the problem of gambling addicts and the companies that take advantage of them would be quickly solved.
But your solution is cool, too.
That's a really cool solution. I had no idea the AI could do this so well and stick to it so consistently. How do you determine the difficulty level?
It’s not a bad idea in itself, but as others have already said, it can be handled more securely using a local file. Especially now, with all the discussion about sovereignty and local hosting, a service like this is rather counterproductive.
But perhaps this could be used to build a local solution with a little more added value. It might be interesting to use the context to determine which AI solves the problem best or most cost-effectively.
I know people who work in urban planning, and they find this also generally interesting because the data and its visualization can be used to identify areas where action is needed in terms of local transit planning.
I couldn't find a direct option to adjust the font size. Could you add that?
I'd be very interested in that, too.
For me, it's more of a gimmick, too, since I find it tiring in the long run.
Since many words in a text are redundant, I read the text by skipping words anyway, and I'm just as fast that way as I am with a speed reader.
btw: Cool blog.
It sounds a little mysterious, especially since you have to sign up. Maybe you should explain more clearly on the page what users can expect.
Cool. I'm thinking of wedding photographers who take several hundred photos in a single evening and rarely know the people in the photos personally. Since not all the photos of everyone turn out well, this helps them pick out the best ones while still making sure all the guests are included.
Yes, that would be very interesting. Depending on the prompt, the result will probably be something completely different, but it would be interesting to see the level of gameplay and immersion.
Haha, nice idea. But it only works if the recipient is over 25 years old.
That's really cool, and it's a great use case for WebAssembly.
I'd suggest making the site multilingual, especially in German. Germany has some of the strictest data protection regulations, and your tool is very useful there,
That's a nice idea. But I would definitely adjust the colors. The colors on the buttons, in particular, are really hard to make out.
It's actually a little addictive, though it wasn't immediately clear how it worked. A tutorial would be helpful.