On 2/20/02 8:23 PM, "Stephen J. Turnbull" stephen@xemacs.org wrote:
Nor do the spammers need to deobfuscate all the obfuscations. They only need enough that they're getting a reasonable harvest rate.
A very good point. We want to make it tough on spambots, but adding complexity to the system is useful only if it actually makes it tough on the spambots. If, instead, it merely ends up adding complexity, we might as well not do it.
If the real answer is "well, it means they have to harvest our site five times to ge the address instead of once", we shouldn't bother. Does anyone know if the /. System actually accomplishes anything? Or have the spambots adapted already?
We can't do nothing until we get a 100% solution -- those don't exist. But we also shouldn't do things just to be seen as doing things, if those things we do don't really help or are merely minor inconveniences for the spam harvesters that are easily worked around.
Ah, the joys of design and architecture. Of course, if it was easy, it'd be written by now.