4 ms·

Are you guessing that there are secrets, or do you KNOW that there are secrets? Can you testify to Congress that there are secrets? Notice how the story chang

by mcherm 11y ago

Are you guessing that there are secrets, or do you KNOW that there are secrets? Can you testify to Congress that there are secrets? Notice how the story changed after Snowden revealed the fact that the US government was doing ubiquitous surveillance. Previously other people had reported that but not been believed. Some Senators knew about it but were prohibited from saying it was happening. Once it was revealed by a trusted source (trusted because his evidence proved he wasn't just making it up) the whole story changed.

whitegrape 11y ago
I know in the bayesian sense of knowing, and could testify to Congress that to the best of my knowledge, there are government secrets. My evidence is simply the existence of old secrets being revealed (in detailed form or not) indicating the government kept secrets, plus a steady stream of declassified secrets presently being revealed indicating that there are still secrets left. I have no idea about the character of these secrets, so this knowledge isn't very useful, compared to the knowledge Snowden leaked which not only gave the character of the secret but many of its details too (and his dump included still undisclosed material that the journalists have so far determined would be unwise to share, another piece of evidence that lets me know there are secrets). What sorts of factors besides "don't do it" could have been done up front to make the surveillance story less evil? Perhaps make Congress be aware of it and able to audit the activities at any time, even setting up alerts? Jeff Jonas has written a lot about the importance of immutable audit tables, which would help other government parties (like Congress) verify what the NSA is doing complies with the law and respects the common citizen's privacy. But having proper auditing in place probably would have made Snowden's leak much more difficult to pull off, so perhaps in this situation the better strategy is to hope for common morality to win out in the end and design systems to make it easier for whistleblowers and leakers to produce strong evidence about some covert activity at the cost of short-term potential abuse from employees using the data for their own personal gain?

by mcherm 11y ago

whitegrape 11y ago

I know in the bayesian sense of knowing, and could testify to Congress that to the best of my knowledge, there are government secrets. My evidence is simply the existence of old secrets being revealed (in detailed form or not) indicating the government kept secrets, plus a steady stream of declassified secrets presently being revealed indicating that there are still secrets left. I have no idea about the character of these secrets, so this knowledge isn't very useful, compared to the knowledge Snowden leaked which not only gave the character of the secret but many of its details too (and his dump included still undisclosed material that the journalists have so far determined would be unwise to share, another piece of evidence that lets me know there are secrets). What sorts of factors besides "don't do it" could have been done up front to make the surveillance story less evil? Perhaps make Congress be aware of it and able to audit the activities at any time, even setting up alerts? Jeff Jonas has written a lot about the importance of immutable audit tables, which would help other government parties (like Congress) verify what the NSA is doing complies with the law and respects the common citizen's privacy. But having proper auditing in place probably would have made Snowden's leak much more difficult to pull off, so perhaps in this situation the better strategy is to hope for common morality to win out in the end and design systems to make it easier for whistleblowers and leakers to produce strong evidence about some covert activity at the cost of short-term potential abuse from employees using the data for their own personal gain?