In article <1179475.1028212096@[192.168.1.10]>, Daniel Staal <DStaal@usa.net> wrote:
--On Thursday, August 1, 2002 17:15 +0000 Tim Pierce <twp@rootsweb.com> wrote:
If I were doing this I'd probably skip having them enter their e-mail address in a box at all.
Yes, but... That assumes several things. First of all, it assumes the browser puts a real return address in there. (Opera for instance has the option of obscuring your return address.) Secound, it assumes they are doing this from their own computer, not a computer at a library or something. Thirdly, it assumes they want to subscribe with whatever the browser assumes to be their 'default' address, and not some secondary address on their computer. Overall, those are probably good assumptions 90 percent of the time, but I'd like to at least be able to check them.
Sure, but asking people to type their e-mail address into a form is not without its own risks. A depressingly high percentage of people are going to get their own e-mail addresses wrong. They'll use "earthlink.con" instead of "earthlink.com", # instead of @, their full name instead of their userid ("Joe.Schmoe@example.com" instead of "jschmo@example.com"), and so on. The return address in their e-mail program is likely to have been debugged already. It's not clear to me which is the more fail-safe solution. Letting people subscribe by just hitting a button is nice, but I would say on the web page that the preferred method is to send mail from their favorite mailer program to listname-request. :-)