for an entire website though of many pages I can see this can be useful.

Comments URL: https://news.ycombinator.com/item?id=48499993

Points: 59

# Comments: 22

- at the first layer, people (even mainstream folks) hear about openclaw and what it "can" do, gives them amazement and think they want to try (somewhat like chatgpt itself did a few years ago). the github stars on the main repo can come from that. people think of the potentials, want to do something with it

- most people who actually set up an openclaw, don't actually have legitimate and substantial use cases of it. I'd wager >90% of them end up setting up something basic. the gmail/search/etc "pedestrian connectors" you mentioned is a good indicator that people are doing pretty basic stuff they could've used other tools for. they use openclaw for it for a variety of reasons, maybe they haven't heard of the other tools, or they think openclaw is novelty tech, etc.

- the skills has way fewer installs and stars than the main repo because setting up an openclaw by itself is hard enough and most people probably don't understand why they need skills or what they use it for. I actually barely understand it myself either.

- also, the high number of stars indicate most people see it and think it's cool and want to try it out, but never end up doing it. as an example, in a tech co of 200~ people, maybe 10 people will have tried actually installing openclaw and using it. most will have heard of it and maybe starred it but never had time or bothered trying.

- most people try openclaw for a bit and realize actually how expensive it is to run. to do some basic cron job you're easily using a few dollars a day in API costs. $3/day doesn't sound like a lot and sounds like coffee money but you soon realize that's equal to a $100/mo subscription (assuming you're using raw API per token pricing and not some flat monthly plan). it really isn't a toy but most people who try it treat it like a toy and it's way too expensive for that. properly set up, it really should be a virtual assistant for a busy business owner. you have to get enough value out of it to be worth its price in tokens.

all in all, at this point in its ease of use, it really is niche tech, that somehow made it into mainstream consciousness, but mainstream is realizing it's not for them, after trying it.

For OP though who is a Turkey-based company and want European region servers anyway, it might make sense.

But either way, nowadays it's trivial to ask Cursor/Claude Code to write you such a shell script.

During initial setup it even asked how you want its personality to be, I said upbeat and cheery. I know, I know, cliche for AI chatbots, but I kinda like it that way. But after that setup, it was nothing like it. Everything it says is just matter-of-fact-ly/stoic. That wouldn't be so bad if it didn't also have this quirk where if you point out its mistakes, it keeps saying it'll fix it, but it still does not work right the next time. It just keeps reassuring you it'll be fine, in that stoic way, and then it's not fine. That's enough to drive me nuts.

I got used to it now though and I don't get mad at it or scold it. I just learned to work with its quirks and still get tasks accomplished.

The category of task I have it do is basically the pattern of "scrape some certain websites on a regular schedule, do some light data processing/understanding/analyzing, report the result to me [all or sometimes only when there is something worth mentioning]".

You could simulate the same things with cron jobs on a server and some scripts and LLM APIs. But having Openclaw do it does make it a little bit easier to set up and make changes.

The initial setup was a bit more time consuming than I thought it would be. I set it up on a VPS, I already have scripts to set up a server and tighten security normally, so I could just use that, but people who don't already have that stuff would have to do that first. Then the Openclaw setup and configuration was like a 20~30 step situation, lots of API keys to get, etc. I opted for Slack over Telegram or Discord (I don't use either of those regularly) and you pretty much have to set up a new slack bot app yourself (you follow their listed steps, but there's still hiccups here and there), you have to debug and solve issues etc. to get through it all.

Then even after all the initial setup is done, it takes some time to learn and get used to its quirks and behaviors, at the beginning there's just a lot of frustration about things it can't do, or things it says it can or will do, but doesn't.

For example a possibly trajectory might be that many years in the future because human thinking has degraded due to AI-assisted cognition, most people will get a chip implant and AI-assistance becomes integrated with the brain. Basically same pattern as most everything else -- technology augments solve the new reality. I'm not saying this will happen, but just a possible outcome of this.

I tend to do extensive research (that process in itself would involve LLMs too, sure) in a tech plan, a product spec, etc. and usually end up with a really solid idea in my head and like say, five critical key points about this tech plan or product spec that I absolutely must convey in this document.

Then I basically "brain dump" my critical key points (including everything about it, background/reasoning, why this or that way, what's counterintuitive about it, why is this point important, etc.) in pretty messy writing (but hitting all the important talking points) to a LLM prompt, asking it to produce the document I need (be it tech plan, product spec, whatever) and ask it to write it based on my points.

The resulting document has all the important substance on it this way.

If you use LLM to produce documents like this by a way of a prompt like "Write a tech plan for the product feature XYZ I want to build", you're going to get a lot of fluff. No substance, plenty of mistakes, wrong assumptions, etc.

I like karaoke and I grew up with the Asian style karaoke with the music video behind and the karaoke lyrics at the bottom.

Sometimes I want to do a song and there is no karaoke version video like that.

A few years ago I came across ML models that cleanly separate the vocals and the instrumental music of a song. I thought of the idea to chain together ML models that can take an input music video file, extract the audio (ffmpeg), separate the tracks (ML), transcribe the lyrics (ML), burn the lyrics back with timing into the video (ffmpeg), and output a karaoke version of the video.

This is an early version of the app, Mac only so far (since I use Mac, despite it being an electron app.. I do eventually want to make a Windows build), I've only let a few friends try it. Let me know what you think!

Comments URL: https://news.ycombinator.com/item?id=47328060

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47289328

Points: 1

# Comments: 0

For devs who currently think this way, I suggest thinking about it more deeply from the perspective of a developer: Let's say you want to start a company/startup from a passionate idea you had. What do you think happens when you build it? In reality, do you truly expect "build it and they will come"? What happens when you bought a domain, put up your product on the web, or the app store? I can tell you what will happen: there will be zero people signing up to use it. Posting it on a Show HN or Product Hunt is an illusion of ease to publicize a product. A PH launch is a carefully planned and curated process involving hours and hours of marketing work to prepare for. A Show HN post will go unnoticed with no clicks 99.9% of the time.

And if you just work in a bigger company, as a non-founder, and say "this isn't my problem, I just build stuff for a job", what do you think the founders did to build their company so there are users who sign up and pay?

I'd argue this is true for social networks like Facebook actually. There was a magical period in Facebook between 2005 to 2010 or so where it was mostly college friends, high school friends, some work friends, and we all actually shared what we thought on our posts, shared links to interesting stuff, etc.

When all the relatives started being added to your network the vibe became decidedly different, and then acquaintances, people who aren't close, etc. and everyone has that one experience where one time they post something and someone who isn't close get offended, whether it's political or not, and they gradually share less and less.