As per the request of one of the DanceNet readers, here are copies of the past ramblings of the DanceNet Webmaster.
I've been having problems with getting email sent to the havetodance.com domain. If you need to send me email at this time, please send it to benson@btwong.org until I get it fixed.Well, now I know how much spam I'm getting. I was wondering why it was so quiet over the last couple of months.
Whoops....here it is...Friday. I just happened to be testing the web browser on the Linux machine I just built and I realized that I forgot to upload the webpages for the week. It's been a while since the newsletter made it out before the webpages.
The third installment of the Lord of the Rings, The Return of the King, is still pretty good on the second viewing. In neither viewing did I notice the passing of (real) time. Must have been a good movie.
I realize that this website is visited by webcrawlers/spiders that harvest email addresses for spam. Occasionally, I get hit at all 60+ fake email addresses that I created for this website so I'm always looking for easy ways to get around that.
One method would be to use javascript to create a routine that would be called to "assemble" an email address within a webpage. Typically a "mailto:" string in an html file is a likely sign of an email address. Such a routine would break up the entire email address string in the file so that a spider program wouldn't recognize the email address within the file. If someone were to visually read the webpage in a browser, the javascript would assemble the address so it appears correctly in the browser and can be "clicked" to active the default email program and fill in the To: field. If you are interested in one example of this for your webpage, let me know and I'll send you what I have. (of course, some people shut off Javascript so this might be a problem).
Another method of interrupting spam is to use Unicode to disguise the source code in a webpage. (I made up a conversion table if you're interested in seeing it). In a HTML file, Unicode and ASCII can be interchangeable; whever you see ASCII displayable text, you can replace it with its corresponding Unicode. For instance, if I were to convert "mailto:dancenet@havetodance.com" to Unicode, it would end up being a Unicode string of 155 characters long and there wouldn't be any sign of an email address, include the "mailto:" or "@" strings.
While this latter method should be pretty effective and wouldn't increase the size of the HTML file by much, it's still rather clumbersome to implement, especially if the webpage has a lot of email addresses. While it wouldn't be that hard to write up a Visual Basic program to do the conversions for you, it still adds up to a lot of work for a text-intensive website such as this one but it it's easier to implement than javascript.
A reasonable compromise might be the introduction of "spam poison". Using the Unicode method, you can convert just part of the ASCII text address into Unicode; the email address would still be displayed correctly in a browser but the source code would not contain a valid email address. I do that by converting the "@" sign into Unicode (probably should do the "m" in "mailto:" also). Since only the email addresses use the "@" sign in the HTML file, I can do a global replace to make any email address on this website into "spam poison". I would imagine that the spammers might get smart about this and upgrade their spiders to take this into account, but I think it'll be awhile before they get around to this (because I don't think it's worth the effort to them). Meanwhile, you can take advantage of this small amount of effort to make it harder for these miscreants to use your website to build their mailing lists.