CFFormProtect: invisible, accessible, automated spam bot killer
I just released a new version of CFFormProtect. CFFormProtect is a fully accessible, invisible to your users form protection system to stop spam bots, and even human spammers. CFFormProtect works like some email spam protection systems, in that it uses a series of tests to find out if a form submission is from a spammer or not. Each test is given an amount of points, and each test that fails accumulates points. Once a form submission passes the threshold of 'spamminess', the message is flagged as spam and is not posted. The points assigned to each test and the failure limit are configurable by you the developer.
Click here to download CFFormProtect.
CFFormProtect uses these tests to stop spam:
- Mouse movement-Did the user move their mouse? If not, it might be a spammer. This test is not very strong because lots of people, including the blind, don't use a mouse when filling out forms. Thus I give this test a low point level by default.
- Keyboard used-Did the user type on their keyboard? This is a fairly strong test, because almost everybody will need to use their keyboard when filling out a form (unless they have one of those form filler browser plugins)
- Timed form submission-How long did it take to fill out the form? A spam bot will usually fail this test because it's automated. Also, sometimes spam bot software will have cached form contents, so the form will look like it took days to fill out. This test checks for an upper and lower time limit, and these values can be easily changed to suit your needs.
- Hidden form field-Most spam bots just fill out all form fields and submit them. This test uses a form field that is hidden by CSS, and tests to make sure that field is empty. If a blind person's screen reader sees this hidden field, there is a field label telling them not to fill it out.
- Akismet-All of the above tests can be easily bypassed if a spammer hires cheap labor to manually fill out forms. However, Akismet attempts to stop that as well. Akismet is a service provided by the folks that run WordPress. The free service (for personal use) takes form contents as input, and returns a yes/no value to tell you if the submission is spam. This test is disabled by default because you have to obtain an API key. This is easy to do, and CFFormProtect is easy to configure if you want to use Akismet.
You can view the project page here at RIAForge.
Jake Munson
38 Yrs old
I just put this in place here last night. I was getting 2-5 spam comments a day, that I had to manually delete. Now, the way I set it up I get an email when a comment fails telling me what tests caused the failure, and the comment doesn't show up on the blog. Over night I had 3 spam attempts, and surprising ALL of them were actual humans filling out my forms (all 4 tests to make sure they're human didn't flag them, they had actually used their mouse and keyboard). But the Akismet test caught them! :D
It looks like riaforge is up again. Let me know if you need any help getting cffp configured on your server.
It wouldn't be super hard to port this to PHP (there isn't a lot of code in this thing). You can grab the latest source from here:
http://cfformprotect.riaforge.org/
The JavaScript parts will not need to be ported, of course. But all of the ColdFusion will have to be translated to PHP. Good luck.
Hunting across the web, I stumbled on cfformprotect and in short, it's been the perfect solution.
This is how we've used it:
* users can switch public commenting on or off on their own blogs
* all comments left by non members get run through cfformprotect (logged in members get a pass so far)
* members can choose to add a personal akismet api key, so this was moved out of the .ini file into the cfc.
* anything passing the system goes into moderation and blogger is notified via email. This might prove overkill now, but was previously necessary.
* users who have previously had approved comments get a pass.
* comments that don't pass the system don't get saved to database at all. User is shown a message letting them know so real comments can still get reposted.
* in addition we've let users set limits on how long public commenting is open on a post, which gets rid of a lot of the bots commenting as they can be slow to pick up on new posts.
I decided to have rejections emailed to me for a while to see if any adjusting was needed and in about a month of usage now, I'm yet to see a false positive come in. All in all, it just proved dead simple to implement and works incredibly well. Highly recommended.
Just thought I'd pass that on with our thanks!