We've had a couple concerns, and we've sent them in in separate documentation.
The first one is about forwarding to a friend. It's similar to the idea that you were just talking about in regard to introducing a friend to a product or to another individual. The way the “forward to a friend” traditionally works is that if I'm on a website, and I like the news article that Mr. Geist wrote, and I want to send it to my privacy person at work, then I enter his e-mail address and name on the form, and I type, “you should read this article” and I hit “send”. There's no clarification concerning who the original sender is because the message didn't originate from my local computer. It originated from another server, which is managed by a website.
I think the idea behind this is that if the message is delivered and the address is not kept or recorded--it's just used to deliver a link that says go read this article with my message included--then it's not really clearly defined who the sender of the message is. I initiated the message, but it was delivered through another network. That's something we have mentioned that should potentially be clarified in the act.
There's also the identification of a sender. What's clarified as an identification? Is it the logo in a message? Because if it's an image, a lot of e-mail clients block images. That's something that needs to be looked at. You can't spam-address stuff by using a postal address, by putting a clear text postal address in a message.
Regarding the idea behind anti-spam filters, there was wording around changing the content of the message during transmission. A lot of spam filters will put headers in that say “we suspect this is spam”. Those types of things need to be clarified as well.