Over 12% of links posted to Hacker News are Show HN projects now(orangecrumbs.com) |
Over 12% of links posted to Hacker News are Show HN projects now(orangecrumbs.com) |
¹ https://mikelovesrobots.substack.com/p/wheres-the-shovelware...
And it's almost certainly hitting other fields too. Games on Steam, mobile apps, websites and web apps, etc.
Quality could be questionable but quantity-wise I can definitely see the start of the exponential growth.
No you can't
I'm struggling to see what the point is to looking at Show HN projects these days. It's like inviting your family in to see your latest bowel movement before the toilet gets flushed.
It was pretty cool. Nothing at all like a bowel movement.
I’ll take it.
There was one she liked but she wanted to change the rules.
Of course it took Claude Code about 10 minutes to do it, and she was delighted.
My point was: that cannot be said about 90% of what gets posted in 2026 with a "Show HN" prefix because it is minimal-effort LLM-coded work that can be recreated easily by anyone with a Claude subscription, and we should not pretend that software developed in 2026 is the same as software developed in 2021. Everything has changed, including incentives to post, and that has changed the level of attention I can devote to that category of HN post.
You can resist.
Upvote quirky projects, call out fake hype, make your disapproval visible, call AI slop what it is.
The term AI has become a liability in some circles, so let that be proof that you don't have to just accept whatever AI companies tell us.
I screamed in people's faces - alienated half my relationships. And absolutely nothing I did had any impact whatsoever. And so I learned to accept the things which I cannot change.
You need to know - the hype isn't fake, my friend. Those stories are real. If you genuinely want to resist, you need to shed your bad information. Good luck.
AI seems to break a lot of peoples’ brains. It’s like it totally bypasses nuanced thinking and understanding and normal curiosity and goes straight to some kind of overflow state.
And you still seem to be exaggerating - I suspect that the number of projects that can be recreated easily is much lower than 90%.
And you sound extremely angry about it. Are you sure that the number of good projects hasn't increased? Or is it just the signal to noise ratio that troubles you?
And... as a final point - I suspect that bird detector wouldn't be terribly difficult for a Claude subscriber. Yet you already agreed it's not in the shit category, right?
My apologies, I assumed that it was implied that I was speaking generally. I will try to be clearer going forward.
I stand by my 90% figure based on what I observe from the full corpus of daily "Show HN" submissions.
> And you sound extremely angry about it.
I can't even begin to imagine what part of my very plainly written comments are communicating "extreme anger" but I guess it's difficult to communicate tone online!
> I suspect that bird detector wouldn't be terribly difficult for a Claude subscriber. Yet you already agreed it's not in the shit category, right?
It depends on what you mean by "terribly difficult".
To provide some clarity on my view:
Just because something can be recreated easily does not automatically make it terrible or bad, but the fact that we can now generate software as easily as we generate a bowel movement (to borrow from my previous comment) means the bar for quality and attention has changed.
> Are you sure that the number of good projects hasn't increased? Or is it just the signal to noise ratio that troubles you?
Yes, signal-to-noise.
I think it was your comparison to fecal matter.