[spicybright]

Just some observations:

- I don't have any info on how this works from the repo. 99% of people will click on the repo and click back because of the lack of this. I went on google and found a [this](https://sideprojects.net/posts/NB697xnBrWQioywoP/rentmycpu-r...) link, but still no information.

- How are payments processed? Do you need to trust the other party a good amount? As a cpu cycle seller, how do I know I'll get money for time I process? How does a cpu buyer know they aren't getting scammed?

- How are computations done? If I have to re-write my existing software to leverage this, I might as well rent a VPS to save a lot of time.

- Do I need to select user by user at a granular level to run my tasks, or can I just throw a task + $ at a pool and have the system figure it out?

I love the idea though, it's really been a bit of a pipe dream for the past 20 or so years. It's just a very tough problem to execute correctly. Best of luck.

[throwawaypls]

Don’t want to be cynical (and appreciate your more constructive feedback), but i’m surprised by the kind of stuff that makes up to the front page these days.

[spicybright]

I tend to agree. I don't expect every project to have a landing page with flat trendy graphics giving you the bullet points like a start up, but just a few paragraphs/link to a white paper would go such a long way.

[geek_at]

One or two screenshots of the web interface would have been enough

[throwaway894345]

This reminds me of homomorphic encryption: you can perform computations on data while it remains encrypted, which would solve the trust problems you mention afaict. This seems like it would have enormous potential to change the cloud landscape. https://en.m.wikipedia.org/wiki/Homomorphic_encryption

[hn_throwaway_99]

My understanding (which is admittedly very little) though is that homomorphic encryption requires many times more computations compared to running on unencrypted data, so that could dampen it's usefulness.

[throwaway894345]

Perhaps, but there is a LOT of unused capacity out there.

[spicybright]

It's less of a resource problem and more of an allocation/networking problem though.

Most problems worth sourcing out requires a lot of data, and piping that through a home network connection to a "meh" CPU and back isn't that useful.

Even if you spend time divvying it up, I can't imagine it would be easier/cost effecitve than just spinning up an AWS node to get it done.

[Jhsto]

It indeed seems to be: https://arxiv.org/pdf/2005.01945.pdf