1) In general, it’s a solid brief. It’s going to be extremely unpopular in the IANAL computer nerd world, obviously, but it’s mostly pretty solid on the law.
2) DOJ brings up some provocative facts not found in the Lavabit brief that are not going to help Lavabit before the Fourth Circuit judges. For example, during the negotiations over whether Levison was willing to comply with the pen/trap order, there was an agreement that Lavabit could change its SSL keys after the pen/trap order expired, billing that cost to DOJ, so that DOJ would then lack the ability to conduct any surveillance beyond that time. Also, when Levison eventually made a counteroffer to comply with the pen/trap order under its own conditions, one of the conditions was that Levison would not turn over any data until the pen/trap order expired, unless Levison decided otherwise at his personal discretion. And earlier, when DOJ came to serve the subpoena on Levison at his home, Levison escaped out the back door and drove away to avoid being served.
3) DOJ argues on page 13 that the lawfulness of the subpoena isn’t before the Fourth Circuit, as the contempt was for failure to comply with the pen register order and warrant, not the subpoena. That’s very helpful to Lavabit, as it eliminates the easiest and most straightforward argument for why Levison had to disclose the keys. That means that the burdensomeness arguments that I focused on last time aren’t actually in the case, as those arguments are subpoena-specific. The only issues in the case are whether the pen/trap and warrant are valid.
4) DOJ also argues on page 13-14 that Lavabit failed to preserve several of its arguments below. In particular, Lavabit had moved to quash the warrant but not the pen/trap order, DOJ says, so it never challenged the lawfulness of the pen/trap order and can’t do so de novo for the first time here. Because DOJ wins if either of the underlying orders was valid, DOJ wins on the procedural point alone. It’s hard to know what to make of that argument without knowing the district court record in more detail, but it’s potentially pretty important. I’ll be interested to see what Lavabit says in the reply brief on that issue.