Anders wrote regarding 'Re: [SLE] html symbols question ..' on Fri, Nov 12 at 18:48:
On Saturday 13 November 2004 00:58, Randall R Schulz wrote: [...]
Nonetheless, all the bot needs is an embedded JavaScript interpreter to thwart that scheme. It's still not as robust as something optical.
Somehing as simple as inserting a nonsense word in the address and giving instructions to remove it to get the real address will more than likely be enough to beat any bot. I find it difficult to believe that spammers would be sophisticated enough to write web crawlers with advanced parsers
Yup, there's that whole return-on-investment thing to worry about that. If you're scouring the web, looking for email addresses, are you gonna want to download linked javascript files and waste processor time parsing javascript to see if anything's document.write()'d that looks like an email address, etc? Probably not. I'm partial to adding a removethis subdomain to addresses, myself. It's pretty easy to write an output filter (with Apache) to take any email addresses and put "removethis." after the @, when the part after the @ is one of our email domains. That gets the added benefit of allowing the rejection of any messages sent to removethis.mydomain, which rejects the message for all recipients when there are multiple recipients specified (using my MTA's setup, anyway). --Danny