4 ms·
SSL/TLS is something that needs to be thrown away and start over (not that it would happen realistically without immense pressure after another spectacular fail
by fubarred 12y ago
SSL/TLS is something that needs to be thrown away and start over (not that it would happen realistically without immense pressure after another spectacular failure). The over-complexity of X509 and the ease of which one can acquire legitimate certs for domains one doesn't own is appalling. From recent revelations, it's even more troublesome the number and scale of exfiltration of private keys, making it possible for some state actors to MITM 10's-100's megaconnections. (One has to put on their tinfoil hat to estimate how many countries have successfully placed staff in core IT/webops positions of Fortune 100 that are then able to leverage that access... Not to mention high-level engagement. [The direct approach conversation might go like this: "gives your keys or we will send in agents to expose embarrassing details about your org and we will still get the keys anyway."])
Perhaps folks like 'cperciva would be kind enough to propose a single, simple TOML-based cert system that is extremely lightweight with the fewest of features. (Not that TLS/SSL would change without focused, sustained herculean effort immediately after yet another Heartbleed.)
- erglkjahlkh 12y agoExtremely light weight? It may be complex, but x509 is based on ASN.1 and one of the most compact and efficient way to represent and handle certificates. There are so many tools supporting it properly, and many of them are extremely hard to change (especially HSMs) because there are actual dependencies, that I wouldn't bother. The problems are mostly in SSL & TLS protocols, not in the certificates. We should get a new alternative that would be designed to be easily implementable, and it should get proper reference implementation (with proofs of correctness, which by the way are available for X509... That's the only part that is verifiable afaik).
- tptacek 12y agoWhat exactly does the ease of acquiring a bogus cert have to do with the complexity of X.509?