It's easy to trust a corporate overlord with your pictures or your email, because the immediate damage doable by somebody who has compromised those things is relatively low. Privacy is important I guess, but not when compared to things like whether your car or your insulin pump does what it needs to to keep you alive.
Eventually, the bad guys will be sophisticated enough, and the tech will be integrated enough, that it's no longer safe to trust economic incentives alone. You're going to want your sysadmin to share your interests (in a more specific way than you get from they-also-like-money).
Not all corporate overloads are equal. Or rather, if you and your buddies get together and pay the $350+fees to legalzoom to start a corporation, you too, can be a corporate overload. There's still miles to go before you're Facebook, but congratulations, you're now... still the same person you were before you clicked that button on legalzoom's webpage and spent $500 or whatever.
Where is the problem of people turning into corporate overloads for you? Is it at 10 employees? 100? 1,000? 10,000? If we're too stupid to differentiate specific corporations because their legal structure means they're all exactly the same, then yeah, I guess there's no hope and we're all doomed.
Really, it's easy to get sysadmin types interested in this; the problem is that most people aren't sysadmins and don't know any. If you really wanted a business model out of this, it'd probably be a managed service that lets non-tech-savvy users spin up their own versions of this without learning the details.
> Without this, it will work --- in the same way that open source works --- without any guarantees or commitments whatsoever.
There are plenty of successful economic models around open source, and plenty of open source software is used in high-reliability contexts. What comparison are you trying to make?
I’d do it for free. I’ve long been frustrated that I have more reliable infrastructure in my homelab than most companies I’ve worked for, and that none of them have any interest in shifting out of the cloud.
I don’t see a market for it, though. Most people are generally happy with Google, Apple, etc. to host their stuff, and I get it - it’s quite reliable, integrates with the rest of their respective products nicely, and Just Works. Add to that the economies of scale, and it’s a non-starter unless you find a niche group of people.
Google One is $99/year for 2 TB of storage. For me to have confidence in uptime to offer public storage, I’d need at least 4U of colo rack space, and ideally 6U (2x 2U for HDD servers, 2x 1U for hosting applications in HA-ish). That would cost a few hundred USD/month, not to mention an initial outlay of tens of thousands of dollars for servers and drives (mostly the drives… high capacity enterprise-rated HDDs aren’t cheap). And that’s only for one site - ideally, of course, there are at least two, or at the very least, off-site backup like rsync.net.
And if you get hit by a car? Or worse --- maybe you get married and have kids<g>?
One big reason people *buy* service is sustainability/longevity/redundancy.
There are no absolute guarantees but I think most commercial endeavors nowadays would bet on AWS/Google/MS/Apple over "Hosting by Joe and Friends".
Personally, I'm betting on those who are highly incentivized and have the resources and structure needed to sustain reliable service.
I maintain 3 copies --- no hosting required. One copy is strapped to my wrist at all times so it is always just as safe as I am --- if not safer.
This is the "1" part of my 3-2-1 backup strategy.
Can you say more? What solvent would I use? What material would I choose to make the parts from? (I would be using a service to print the .stl file.)
Also, FWIW I am married and have kids. Hasn’t stopped me from homelabbing.
Co-location is still readily available. Which service and reliability improvements are you looking for that competent sys admins couldn't provide with multiple co-lo's? Not everyone made the cloud jump.
I moved to AWS and haven't had that problem since.
What we need first is incentive for smart, dedicated, part-time sys-admins to devote time and effort to community hosting.
Without this, it will work --- in the same way that open source works --- without any guarantees or commitments whatsoever.
In other words, you're on your own for the most part. So it really is just a variation on self hosting. By the way, we've already been there, seen that and done that --- it was called "co-location".
When you need something more with service and reliability, well --- you're right back to paying corporate overlords.
But thanks for the round trip thought experiment.