If the operative word is "sent" and "all" the likelihood is zero, as I can assert I've never sent a private key to NSA, and I've made quite a few over the past few years.
That said, carlosdp makes a great point as to how one should behave. Even though not all private keys get shipped to NSA (can't make claims about other teams), the government is very public about other data sharing programs (see https://www.dhs.gov/sites/default/files/publications/privacy... for example).
Even though all government agencies must disclose these types of programs, they are so numerous and often so difficult to decipher, the only rational response is to assume everyone has everything or could have access in the future.
The only way to avoid this is to build zero-knowledge systems on the server side, something I hope you'll see more of in 2017.
Ultimately, it comes down to how you construct the root origin of the problem facing government right now. For me personally (hence why this post is on Medium and not on a *.gov) is the lack of both faith and trust (two very different things) in public structures of all kinds, not just the federal government.
We've lost that faith and trust for very good reasons, to be sure. We may forgive, but not forget those indiscretions to put it mildly, and they go back to the founding of the republic.
The problem as I see it though is that faith and trust have fallen into such disrepair, it's become a self-replicating negative cycle. So few people believe govt can do anything well (in terms inclusive of ethics/effectiveness/efficiency) which leads to apathy, which leads to skilled people leaving govt or not joining, which leads to further degradation, which leads to a further undermining both in perception and in actual funding or authorities from Congress, and on and on it goes.
That cycle is not just observed at the federal level of course. Even more pernicious to me is the apathy it creates at the state/city (or even community) level, even if one person or one small group's potential for positive impact is exceedingly more likely at the level!
I certainly still have my ethical red lines, and I recite them to myself on a near daily basis so I don't normalize the "might" of the feds with "right". But I still see the one of the core origins that got us to this point being the cycle above, so trying to break the cycle is still my #1 priority for now.
OP here. Yep - the record keeping is out of this world. Due to the spending clauses of the Constitution, even a penny spent in excess of budget or even without all the proper authorities in place can result in a Constitutional violation, and you need to write a fulsome report to your agency head, Congress, the White House, auditors, etc. I've seen it done for what are effectively rounding errors for the US Govt (a few hundred dollars in the wrong spending 'bucket').
Echoing comments above, the laws/regs are strong, the problem continues to be implementation. Govt fiscal IT systems are some of our worst / most tricky legacy systems to modernize.
OP here. I totally grok your feeling on that, but we've adopted the opposite strategy from Day 1. Extreme transparency to the degree we're allowed, so the public can see the value of the work and defend it if necessary. At the moment, I'm still cautiously optimistic.
OP here. "In theory" - most of the stories don't realize we're funded by a fund with billions in revenue, and was net positive in 2016. https://www.gsa.gov/portal/content/152462#ASF
Congress designed the revolving fund with the purpose of spending any funds in excess of costs to help the rest of government be more efficient and effective, as those funds don't naturally return to the US Treasury as they would in other parts of government.
In terms of 18F's performance in both regards, the capital start up funding we've received from the fund has been extremely well spent IMHO, but of course I'm totally biased. I'd look at what we've produced with that $ and let the work speak for itself, especially given how much govt would normally spend on this much information technology: https://github.com/18f
That also only captures things which have an impact in software - we're still working on the best way to represent the fiscal or performance impact of our consulting work.
All that said, of course we can always be more efficient ourselves, and I think the next few years of fiscal reports will show that.
I’m Noah Kunin, the Infrastructure Director at 18F/GSA.
While the Department of Homeland Security (DHS) owns the Trusted Internet Connections (TIC) policy and controls (https://www.dhs.gov/trusted-internet-connections) we’ve been working hard with DHS teams to clarify and improve implementation guidance.
We hear you - loud and clear - and understand there’s a lot of frustration.
Check out our updates with one of our pilot partners, Amazon Web Services:
18Fer here. The main issue here is how the Gov actually gets $ to that entity. There are lots of laws and rules that we need to obey about who we pay $ to - Gov doesn't have a ton of flexibility here. There was a reason why bidders had to be pre-registered.
That said, many of us want to continue to expand our engagement with the community. When it comes to bounty/reward style programs, the one I'm personally focused on figuring out first is a bug bounty program. Once we've figured out solutions for a bounty type program, you'll absolutely hear from us.
18Fer here. Before I answer in greater detail, why do you think FIPS requires RHEL with mod_nss only? I don't see why an OpenSSL in FIPS mode wouldn't fit the bill too.
If the operative word is "sent" and "all" the likelihood is zero, as I can assert I've never sent a private key to NSA, and I've made quite a few over the past few years.
That said, carlosdp makes a great point as to how one should behave. Even though not all private keys get shipped to NSA (can't make claims about other teams), the government is very public about other data sharing programs (see https://www.dhs.gov/sites/default/files/publications/privacy... for example).
Even though all government agencies must disclose these types of programs, they are so numerous and often so difficult to decipher, the only rational response is to assume everyone has everything or could have access in the future.
The only way to avoid this is to build zero-knowledge systems on the server side, something I hope you'll see more of in 2017.