No, the threat model was outlined in a previous post. Consider some "agency" that has lots of resources and technologies, but also doesn't particularly want local authorities or (for instance) hotmail to know what they are doing. In general, this is going to make their operation much less intrusive, lower cost (ie, due to not having to physically send people) as well as avoiding a lot of legal hassles due to paper trails. So I guess what I'm looking for is way to be quite certain that someone (aside from Hotmail admin) is opening, reading, and closing my email 'unobtrusively'. Of course, once such an effort is detected, said agency may decide to follow a more intrusive investigative path, but this has practical consequences. My home alarm system is probably a better example. If NSA, for instance, is going to bother entering your house and setting up whatever, I'd bet they'd LOVE to not bother with the local security/alarm company, because then there's a paper trail, people who might be a friend of the surveilled, and other 'local' issues. They're definitely going to use their fancy gadgets, etc..., to bypass the alarm system while making the alarm company everything's going just fine, or perhaps a battery has expired. In this case there'd be nothing to subpeona. Therefore, if you suspect you're being surveilled, even if you can't secure anything you want might want to secure, you can at least force them to commit legally actionable acts, or else force them to give up their 'phishing' expeditions. -TD
From: Bill Stewart <bill.stewart@pobox.com> To: "Tyler Durden" <camera_lumina@hotmail.com> CC: cyphrpunk@gmail.com, cypherpunks@al-qaeda.net Subject: Re: Email Certification? Date: Wed, 27 Apr 2005 16:04:54 -0700
I'm still having trouble understanding your threat model. If you're talking about somebody who can get Hotmail's cooperation, e.g. cops or sysadmins, there's no way you can prevent them from doing anything they want to your incoming mail. If you're worried about crackers guessing your password, then some web-based email systems automatically mark mail as read, some don't, some let you mark it, some let you remark it as unread. (I haven't ever used hotmail, and my cat stopped using it when the Child Online Protection Act required Hotmail to cancel accounts for anybody under 13 years old who didn't have parental permission, so the interface has probably changed since I last saw it.)
Are you worried specifically about Hotmail? You're mentioning using gmail to pre-filter your hotmail messages - gmail's going to have similar potential threats, except that it's probably better managed, and if you're going to send the mail to gmail anyway, why not just read it on gmail? In general, if you've sent unencrypted email to an untrusted system, then you've got no way of knowing that it hasn't been read.
At 01:09 PM 4/27/2005, Tyler Durden wrote:
Oh...this post was connected to my previous one.
Sorry...my ideas along these lines are still a little foggy but I'll try to articulate.
Basically, let's assume someone with some resources has cracked your email and wants to monitor what you send and receive. let's also assume they don't want you to know it. Let's assume they also are not particularly thrilled about having hotmail know what they're up to (if needs be they can obtain a warrant, etc..., but this is clearly less than desirable compared to more direct techniques). It seems fairly easy to me to (for instance) create a bot that duplicates all of the email and resends it to your hotmail account so that when you log in everything looks fresh and new. (There are probably easier ways to do this via direct hacks of hotmail).
Is there some way to make it evident that someone has opened your email?
Right now, I can't think of anything you could do aside from suggesting that hotmail (or whoever) offer some kind of encryption service.
BUT, it occurs to me that you might be able to have gmail forward your mail to hotmail via some intermediate application you've set up that takes the timestamp and whatever and creates a hash.