Google Chief Architect: we only respond to specific orders about individuals

I can't tell you want to believe, but maybe a place to start is: the world is a confusing place, full of miscommunication and gray areas. Why should we expect the world to be as it is in the movies, in which bad guys are obviously bad and that there is an "ending" in which things are clear?

For starters, I just re-skimmed the Washington Post report and noticed that it has since been amended:

> It is possible that the conflict between the PRISM slides and the company spokesmen is the result of imprecision on the part of the NSA author. In another classified report obtained by The Post, the arrangement is described as allowing “collection managers [to send] content tasking instructions directly to equipment installed at company-controlled locations,” rather than directly to company servers.

edit: Here's the link: http://www.washingtonpost.com/investigations/us-intelligence...

I think I'm wrong that that was one of the actual changes since yesterday...either way, the Post is allowing for the possibility of a miscommunication/misinterpretation by its source. Business Insider also alleges of other hedging by the WaPo here: http://www.businessinsider.com/washington-post-updates-spyin...

At this moment in time, we still do not know two things: who the leaker is, beyond a "career intelligence officer" and the contents of this 41-slide presentation other than the 3 or 4 slides that the Post and the Guardian have published. The identity or motives of the leaker isn't necessary to know, but it's kind of problematic when we are missing context to the so-far published slides that explain the PRISM program. For example, it kind of changes things if the first two slides say "PROPOSAL" or the last slide says "APRIL FOOLS"

Ah, you are right on the PRISM distinction, I mixed up details with the "metadata" thing, which I'm sure is not accidental. A lot of the responses from government officials are to the effect of "these aren't a concern, because to get the metadata we need to go through FISA courts for every individual", which is just a red herring now.

So, there are:

1. NSLs, published on the transparency report

2. FISA orders, which are "through a court", but is by all accounts just rubber stamping, not on the transparency report

3. PRISM, which is intended to bypass FISA entirely, according to the Guardian article. Either Google is lying, the NSA is doing it without Google's knowledge, or there is subtle wordplay involved (so, lying). Or, that it's been grossly misreported.

I think he broke both stories.

From http://www.guardian.co.uk/world/2013/jun/06/us-tech-giants-n...

> The Guardian has verified the authenticity of the document, a 41-slide PowerPoint presentation

"Any program" would be broad enough, but he didn't say that at all. He said they are not part of "any program that would give any government direct access to their servers." I could drive a truck through the holes left in that wording.

Do they have indirect access? Some API perhaps? Do they have any means by which they can automate the export of data for whoever they want, perhaps after clicking a checkbox that says that the target is officially under surveillance? Is there some form of data sharing that is brokered through a trusted non-government entity?

He also goes on to say that they follow the law (meaningless if the law says to hand over the data), and they frequently push back (which orders do they push back on? Probably not the orders that they aren't allowed to legally push back on).

He states that they don't follow "broad orders for all data", but this is easily satisfied by the description in the Guardian article that says an analyst simply has to certify each request by saying they believe that there is a 51% probability that the request is legitimate. Obviously, no one at Google even could challenge such requests.

Two things that cannot be true at the same time: this Google+ post, and the idea that Google coughed up their TLS keys to the government.