On 14 Nov 2003 at 18:53, SANTU DESILVA <[log in to unmask]> wrote: > Coming out of lurkdom to ask: > > Could you define these terms as you heard them defined at > the FTC Span Forum you attended in May? The description > the gentleman from Notre Dame used seemed particularly > clear to me. Has this usage been altered, as he seems to > suggest? Sorry for the delay in answering, as I wanted to provide a thorough answer, and haven't had time to go searching through the very long transcripts. The full transcripts of the FTC Spam Forum are available at: http://www.ftc.gov/bcp/workshops/spam/index.html The definitions were on Day 2, pages 219 to 226, presented by Rebecca Lieb, in the context of the "Best Practices" panel. It was a pretty long discussion, so I won't quote the whole thing, but here are the main points: "[...]lots of references have been made over the past two days to opt- in and opt-out, and there are more subtle gradations along that chain, and I've identified five[...] [I include the following paragraph as a nod in Paul's direction, as his views on the subject were presented as well, which I had forgotten] "[...]my descriptions of these are going to be more important than what I call them. Some people say, well, only Spammers call it double- opt-in, and if it's really double-opt-in, you have to say it's confirmed opt-in if you're legitimate, semantics. I suppose people are eventually going to agree on the terminology. What's important is to understand what the various options are and what they mean to both users and to e-mailers[...] "I'll start with the worst and work up. The worst practices, and I think there was some consensus on this yesterday among the audience, at least, is opt-out. Opt-out is when somebody's address is added to a list without their knowledge or permission, and it's the recipient's job to tell the sender that they don't want it anymore[...] "A step above that is confirmed opt-out. Your e-mail address is added to a list of recipients, and you receive an e-mail saying you have been added to this list, you can do something about it, and then there is some sort of unsubscribe option in that e-mail[...] "Right in the middle of the equation is pure opt-in, which is pretty straightforward. You go to a website, there's a thing that says sign up for our newsletter or our specials or our deals, you type in your e-mail address, hit send, and you're subscribed[...] "Better than that is confirmed opt-in. You opt-in to something, and because you have opted into it, you get an e-mail, and it says, you have opted into this. Here's your user name and your password, if that's the case, and at least you know what's going on[...] "The gold standard is what I term -- and there is some disagreement on this, but I think it's the clearest terminology -- double- confirmed opt-in. [...] The user takes an action to subscribe, and immediately receives an e-mail that says, you have subscribed to this, but in case you are not the person who subscribed to this, your subscription is not going to be active until you answer this e-mail to confirm that this e-mail address is really the e-mail address that wants this subscription. "It's a more cumbersome process. The response rate to those e-mails is between 40 and 60 percent, which scares a lot of marketers and publishers to death, but it makes for the least complaints, the happiest subscribers and the most valuable lists for marketers and advertisers, because these people have proven not once, but twice, that this is, indeed, something that they want and are eager to receive." So, mea culpa, I used "double opt-in" instead of her term of "double- confirmed opt-in". I've mostly seen the term "double opt-in" since in marketing literature, and so that's what we used in the newsletter. We aren't trying to convert the anti-spammers. We want to convince the *marketers* -- those who want to be legitimate -- to do things right and not alienate their subscribers (the scumbag spammers are hopeless, there's no point trying to convince _them_). So we used *their* language. Francoise