Mngh. I may have failed to talk myself out of a project.

Discussion
A while ago, I contemplated the idea of offering hosted iocaine, for a very small fee, about enough to cover the hosting costs, if a dozen or so people would sign up.
The idea was that you'd get a WireGuard tunnel, and through that, could configure which hosts the reverse proxy should front for, and the backend it should use to serve requests that pass iocaine. A simple CLI tool would be provided to manage that, along with an API.
I'd terminate TLS, do the fronting & filtering, and people would self-manage the rest. I'd fight the robots, and we'd collectively giggle at their futile attempts of getting through.
It wasn't a bad idea, there was even some mild interest, too. I eventually gave up, because I couldn't figure out how to build it and run it in a way that wouldn't drive me (even more) insane.

Meanwhile, I tried my best to make iocaine 3.0 much, much easier to self-host. How well that worked? We'll see! I have to cut a release first.
But there remained a problem: even if I make it easy to self-host, one still needs a VPS, and enough sysadmin chops, and enough spoons to set it all up, and run it. I enjoy doing that kind of thing, but a whole lot of people who'd benefit from any kind of scraper protection, do not.
This has been eating me away.

So I thought about a few other ways I can make the life of AI companies hell.
Because if I have to be honest with myself, the reason I'm trying to make iocaine better, easier to use and self-host is not that I'm a selfless person. I'm not doing this out of the goodness of my heart.
I'd love to pretend I do, because that'd help me become a better person. But no, the real reason is that I hate the Crawlers, and the companies behind them, and everything they stand for. And that hate is powerful enough to drive me.
Sure, sure, I'm delighted every time I see people fight back, whether with iocaine, or in any other way (even Anubis! I don't like Anubis, nor its approach, but I admire people's determination to fight back in whatever way they can). But deep down, below a few layers of genuine delight basks the little shoulder-devil in the frothing rage of my heart.
I can hear its laughter. Even now. And the bloody bastard has good ideas.

In a number of different variations:
I could provide a WireGuard tunnel to an iocaine instance, either shared, or dedicated - in the grand scheme of things, this matters little.
You get to terminate TLS, and control all aspects of your site, only the filtering (& potential garbage generation) is left to me.
I imagine this could be useful for people who already self-host, but don't want to maintain their defenses. The downside here is that there's some additional traffic between your server and mine. HTTP headers for every request, and iocaine's 3-4k garbage responses. I suppose that'd still be a lot smaller than having to serve real contents to the Crawlers.

The original idea: I terminate TLS, do the filtering, you provide the backend. I give you an API, a CLI and a web interface to configure stuff.
The advantage here is that there's less outgoing traffic for you, as I receive the requests directly, and I can serve them garbage directly, without going through WireGuard.
The disadvantage is that I terminate TLS.

And maybe more, if I can think of anything. The point is, that I wish to be able to provide different ways to take the burden of crawler fighting off of people's shoulders.
Because I want to see the fucking crawlers eat shit and die. The little devil in my heart has poisoned my mind, and I now want others to join on this goulish delight.
Watch the dashboards togther1, enjoy the calm breeze of the autumn wind, because we no longer need to worry about our servers burning down under the assault of the Crawlers.
As in, everyone watches theirs when and if they want, and I'll watch mine the same way. ↩︎
As for when? I don't know. I started to build the infra & tooling for the original idea. A lot of that can be salvaged. There's a lot to be built, still, too.
I also have to procure a server to run it on - likely rent a VPS, because my homelab is not equipped well enough yet for this. We have power spikes at times, usually when I'm asleep, which usually result in 4-8 hour outages until I turn power back on.
That's acceptable for my own stuff, not when I'm providing service for others.
Yet, I want to have a rough estimate, one that's far enough so I don't feel pressured to crunch, but close enough to matter.
I will be aiming at launching this service on February 17th, 2026.
A dozen years before that day, my life changed, it's an important, and very dear anniversary - maybe that will even out the hate that drives this idea right now.
There's lots of things to work out. I'll start by turning this thread into a blog post, expand on it a bit, and publish it. I'll aim that for December 8th - a few weeks after iocaine 3.0, giving me enough time to ease into this new project, and present it in a... better way than this little toot storm.
@algernon just be sure to accept donations; I think that the whole powered-by-spite angle is very inspiring and you might even get people chipping in who don't use the service
I'm looking forward to this!
A space for Bonfire maintainers and contributors to communicate