@timewarp ok, PGP … remember EFAIL… and all kinds of usability issues which inevitably lead to security issues by ‘wrong use’ at some point. And another *centralized* ‘web of trust’ (benign as it may be) is also not something I look forward to. O well, some genius will emerge at some point and deliver us 🥳 may he/she/it/them be FOSS-minded
It’s quite possible that privacy is too hard for you and trash talking open source makes you feel better about the money you’re paying to someone else to say they’ll do a better job for you.
Okay, well it’s just the vulnerabilities you mentioned were geared towards email client issues that among other things would automatically load HTML data upon decryption. Furthermore, primary vulnerable targets were 10 year old email clients at the time that hadn’t received any security updates. The SE data packet issue had been documented even in the spec since at least 2007 about its security issues and recommended rapid mitigation techniques. All in all, the EFAIL documented issues with mail client failures, not with OpenPGP itself.
Second, OpenPGP web-of-trust, or whatever you want to call it (public keyservers) is entirely optional. In fact, Proton relies heavily on this in from what I can tell actually enforces it in a more insecure way, but opting users into their internal keyserver automatically.
@timewarp ok, PGP … remember EFAIL… and all kinds of usability issues which inevitably lead to security issues by ‘wrong use’ at some point. And another *centralized* ‘web of trust’ (benign as it may be) is also not something I look forward to. O well, some genius will emerge at some point and deliver us 🥳 may he/she/it/them be FOSS-minded
It’s quite possible that privacy is too hard for you and trash talking open source makes you feel better about the money you’re paying to someone else to say they’ll do a better job for you.
@timewarp don’t know what you’re talking about, I love FOSS…
Okay, well it’s just the vulnerabilities you mentioned were geared towards email client issues that among other things would automatically load HTML data upon decryption. Furthermore, primary vulnerable targets were 10 year old email clients at the time that hadn’t received any security updates. The SE data packet issue had been documented even in the spec since at least 2007 about its security issues and recommended rapid mitigation techniques. All in all, the EFAIL documented issues with mail client failures, not with OpenPGP itself.
Second, OpenPGP web-of-trust, or whatever you want to call it (public keyservers) is entirely optional. In fact, Proton relies heavily on this in from what I can tell actually enforces it in a more insecure way, but opting users into their internal keyserver automatically.