handsomeransoms

It seems like you are trying to force your user to remember a salt. Why not just use a proper salt and a strong password hashing function?

Also note that this protection is only useful in the case where an attacker can get a database dump but cannot perform an active attack on the server.

On the other hand, I have seen some sites (gandi.net comes to mind) do something similar to this. Wonder if they have a similar security reasoning?

The goal of this amendment (appears to me, a non-laywer) to be to allow judges to issue warrants for crimes that occur in their jurisdiction, for materials that may not be in their jurisdiction, when the location of the materials has been obfuscated with an anonymizing technology. I don't think this is an "automatic warrant" - they still have to establish probable cause, etc.

A more interesting sentence from p. 325 discusses the mechanism by which the search may be carried out: "The proposal speaks to two increasingly common situations affected by the territorial restriction, each involving remote access searches, in which the government seeks to obtain access to electronic information or an electronic storage device by sending surveillance software over the Internet."

A few initial questions:

1. Is any part of the communication forward secure? I can't imagine how the multiparty chat would be.

2. Using the avatar to verify cryptographic identity seems weak, mostly because I don't expect users will check it (it's only in the Contacts view, and it's unclear that it has that use). It resembles the placeholder avatars used on Github among other sites, which seems to suggest that it is not meaningful. So - can the Peerio server silently MITM my communications?

3. I'm not quite sure how the search works (still reading the code), but it seems like it must be searching the plaintext stored in the client's memory. How well will that scale?

This depends on your threat model, and I'd argue that this statement is untrue for the vast majority of people, even people who have the technical know-how required to run their own email server. Running an email server, keeping it up to date with the latest security patches, managing SSL certificates, blocking spam, blocking malware, and blocking phishing attempts are all things that Google is better at than you. Part of the reason for this is their access to incredible volumes of data, which lets them analyze trends and emerging threats across an relatively large subset of the email-using population.

The average user's threat model is much more along the lines of phishing, malware, or spam-related fraud. Google is incredibly good at protecting people from these threats (as is obvious when I compare the volume of spam, which often contains malware or phishing links, that I receive on my Gmail account against my other, non-Gmail accounts).

Of course, if your threat model is that you require protection from law enforcement or government surveillance, then Google may be a poor choice as they are legally obligated to turn over information about you that is requested by such entities. If that is your adversary, however, than you should have a lot more work to do to protect yourself than just quitting Gmail and setting up your own mail server.