4 ms·

> where anyone with enough money can buy their way in to obtain that PII, track you down, and maybe cut you up with a bone saw. TBF, I think that the vast majo

> where anyone with enough money can buy their way in to obtain that PII, track you down, and maybe cut you up with a bone saw. TBF, I think that the vast majority of companies out there are vulnerable to this. I’ve worked for 8 tech companies in my career, none of which did anything beyond a basic background check. Truly mitigating the problem you’re touching on requires a level of vetting and surveillance that you’d typically see applied to intelligence operatives. I think this is similar to how we view infosec generally: those with sufficient resources will be able to penetrate a network, regardless of the design or execution of network security.

johnyzee 6y ago
This is letting Twitter off the hook. It is not impossible to protect users personal information, even within a company, to a very limited set of people who actually need it, with audits on when and how they are accessing it, and periodic reviews of everyone's access levels. Mature organizations follow specific standards for this kind of stuff. For a company like Twitter, where the privacy of this information literally can mean life or death, it is unforgiveable to not have a better grip on it (cue some non-technical regional bizdev guy having deep access, as per the article).
- save_ferris 6y ago
  That’s a really good point, reducing access would help here. If you’re a nation state with the resources of Saudi Arabia, it still wouldn’t be impossible to bribe or blackmail an employee who has prod access because companies don’t have much of an eye on employees private lives. They would have the resources and theoretically the incentive to really dig into the social engineering/coercion side of things. You’d be amazed how many people that make incredible salaries are in fairly significant amounts of debt. It’s pretty common in western culture, and that’s a prime opportunity for those kinds of operations.
  - sneak 6y ago
    “prod access” should not be a single large group or access boundary. Access to non-tokenized PII data stores or the small set of systems that require touching untokenized PII should be a very small compartment, with extremely tight min-two-person deployment/introspection controls and minimal change frequency.

johnyzee 6y ago

This is letting Twitter off the hook. It is not impossible to protect users personal information, even within a company, to a very limited set of people who actually need it, with audits on when and how they are accessing it, and periodic reviews of everyone's access levels. Mature organizations follow specific standards for this kind of stuff. For a company like Twitter, where the privacy of this information literally can mean life or death, it is unforgiveable to not have a better grip on it (cue some non-technical regional bizdev guy having deep access, as per the article).

save_ferris 6y ago
That’s a really good point, reducing access would help here. If you’re a nation state with the resources of Saudi Arabia, it still wouldn’t be impossible to bribe or blackmail an employee who has prod access because companies don’t have much of an eye on employees private lives. They would have the resources and theoretically the incentive to really dig into the social engineering/coercion side of things. You’d be amazed how many people that make incredible salaries are in fairly significant amounts of debt. It’s pretty common in western culture, and that’s a prime opportunity for those kinds of operations.
- sneak 6y ago
  “prod access” should not be a single large group or access boundary. Access to non-tokenized PII data stores or the small set of systems that require touching untokenized PII should be a very small compartment, with extremely tight min-two-person deployment/introspection controls and minimal change frequency.