Subnostr

One thing that would greatly increase my confidence in recommending nostr:npub17tyke9lkgxd98ruyeul6wt3pj3s9uxzgp9hxu5tsenjmweue6sqq4y3mgl to those looking for a hardware signing device would be a cold eyes expert review of the stateless claim.

I left this feedback with the team over nine months ago, and have not heard back that they are pursuing this route.

https://github.com/SeedSigner/seedsigner/issues/391#issuecomment-1599256650

cc nostr:npub1tv8gmfhalwnxxquxjzeh6gtdsdz6vg7vx0s3rt7s7uuw6aujh32qn77wn2

SeedSigner 1y ago

Happy Good Friday to you as well, have had several people DM me about this now.

While the CS principal of the volatility of RAM is widely accepted, and if false, the viability of Tails OS and many other forensic live environment tools would be completely destroyed as well.

That said, (1) would it be a good idea from an adversarial perspective for anyone invested in SeedSigner to perform or be involved in such an analysis? and, (2) we are a volunteer FOSS project -- are we expected to somehow fund someone else to do this work? Even if we did this, should they be trusted if we are paying them for the work? (the incentives would be all out of alignment)

With SeedSigner's growing popularity, if RPi RAM really isn't stateless, this would be a complete softball for a HWW company to demonstrate in a reproducible manner and remove us completely from of the cold storage ecosystem.

I don't know what more to say here, if you want to believe then it sounds like you should not trust me and and figure out how to independently verify. It could be a wonderful contribution to our project and could simultaneously allay your concerns.

Appreciate your interest and curiosity. Peace be with you.

Reply to this note

Please Login to reply.

Discussion

elsat 1y ago

> would it be a good idea from an adversarial perspective for anyone invested in SeedSigner to perform or be involved in such an analysis?

Sure! Although I think payment might be required to incentivize the few people in the world who could review this.

> (2) we are a volunteer FOSS project -- are we expected to somehow fund someone else to do this work?

I am painfully aware of the non-paid nature of FOSS contributions. As I suggested in the post, I think this would be a strong stand alone opensats/other grant candidate application.

> Even if we did this, should they be trusted if we are paying them for the work? (the incentives would be all out of alignment)

This is how the security industry works. You pay for a review. For reference, see SimpleX security review synopsis, and report links: https://simplex.chat/blog/20221108-simplex-chat-v4.2-security-audit-new-website.html

https://github.com/trailofbits/publications/blob/master/reviews/SimpleXChat.pdf

> With SeedSigner's growing popularity, if RPi RAM really isn't stateless, this would be a complete softball for a HWW company to demonstrate in a reproducible manner and remove us completely from of the cold storage ecosystem.

If HWW company is spending demonstrating resources trying to discredit seedsigner this is great marketing 🤠.

> I don't know what more to say here, if you want to believe then it sounds like you should not trust me and and figure out how to independently verify. It could be a wonderful contribution to our project and could simultaneously allay your concerns.

I think it’s an opprtunity for seedsigner to say you are the only FOSS security reviewed HSD - this will be a yuge win to offer against security criticisms.

Unfortunately I am not in financial shape to fund a security review, nor do I have the expertise to review myself. If you need help with pursuing grant money let me know, and I can readily contribute my time towards such an application/s. Just let me know.

I think it’s a worthwhile effort and noble project what yall are doing to provide an affordable HSD with off the shelf components to the masses. If I did not believe in or care about your project and motivations I would not offer my time, or bring up this near year old comment.

Peace be with you.