Humans Fail Turing Test
I use a relatively unsophisticated but highly effective filter for blog comment spam. The system consists of several hundred keywords (mostly spam-related domain names which would never appear in a genuine comment), and a Turing test. The Turing test is a box at the bottom of the comment form, that says: “Do not put anything in this box if you are a human.”
The Turing test catches between 10-25 spam comments per day; the keyword filter about the same.
Every week or two, however, a human comes along, and fails the Turing test and fills in the “don’t fill in this box” box. I’m not quite sure what to do about that. Maybe the human thinks they are not actually human?
More likely they just don’t read the text next to the box, although it is fairly prominent.
Unfortunately, this seems to be a general phenomenon on the web. Many people have a tendency not to read text on webpages. For example, I’ve gotten a lot of questions along the lines of “How do I access my address book in SquirrelMail?” The answer, of course, is to click on “Addresses.” Similary questions arise with respect to my photo album software salonify — “how do I view a slide show of images?” (A: Click on “Slide Show.”)
My hope is that the next generation is better able to process web content, but I’m not sure that hope is well founded.
Getting back to the blog comment spam issue, it’s interesting to observe what sorts of topics appear most frequently in comment spams. Over the past several months, pornographic links referencing transsexuals seem to outnumber other topics. I wonder: do the spammers have a scientific method to their marketing?
Christoph Berg Jan 28
The Turing test is a box at the bottom of the comment form, that says: “Do not put anything in this box if you are a human.â€
This is not entirely true, the actual text is more complex and I had to read it twice to really get it. I imagine people commenting care about their comments and don’t care to read instructions that don’t seem relevant.
Adam Rosi-Kessel Jan 28
Christoph: fair point. I was definitely paraphrasing. Perhaps I should simplify the text, although I do think it’s pretty clear where the comment should go.
I put the additional word “comment” in the Turing test area to fool any robots that are looking for the word “comment” before a field to fill in.
Dylan Jan 28
Perhaps also you should make the entry box much, much smaller, so people are less inclined to use it. Are robots likely to look at the size of the text box?
Josh Triplett Jan 28
Along the lines of Dylan’s suggestion, why not try marking the spam-fooling text box and label with class=”spam-trap”, and adding “spam-trap { visibility: hidden; }” to your CSS? I seriously doubt that spambots read CSS. :)
Josh Triplett Jan 28
Also, for the benefit of any users who might have non-CSS browsers, I would also suggest simplifying the label to “Spam trap; do not use for comment:”
Jacobo Jan 28
Usually, behind such “stupid” users there’s a design error. A time ago, I was receiving bug reports because, in a reporting webapp, the wrong report was generated. In reality, the page design made it look like the wrong button was the right button (even I got fooled by it every once in a while). I changed it and the complaints stopped :-)
I think that you should at least change the current text in the box below to the one you used in the story (“Do not write anything in this box if you’re a human”). You can add “or your comment will be discarded as spam”. It’s definitely easier to read than the current text (so many negatives).
awksedperl Jan 28
because the message of not putting text in the box is unnoticable.
Øivind Jan 28
“Put some text in this box only if you are not a human and don’t want your comment to appear (comment):”
The major flaw is in the structure of your text it’s too long and should use wording akin to “…don’t put text in this box if you’re human…[blabla]”, but as pointed out by others a simpler text like “Don’t put anything in this box if you’re human”. The (comment) part of the text just adds to the users confusion too.
The box could also benefit from some css-love to make it smaller, less prominent. Right now it looks like a space for signatures, a private message to you or something like that.
Mick Jan 28
I think the problem may be the location of the text telling one not the fill in the box.
My suggestions:
1. Put the text above the “do not enter box”.
2. Make the “do not enter box” really, really small.
3. Change the text to something like:
“Do NOT write in the box below”. The current text is kinda vague, and too long.
We should patent this and make it free before someone like Amazon does ….
Simon Jan 28
Are you sure they aren’t just curious. Most of the members of our GLUG presented with the H2G2 “do not press this button” would instantly press it to see what happens.
Can I suggest “leave this box empty”. as a shorter label?
Anonymous Jan 28
Adrian Jan 28
What about making the form input invisible via a nontrivial CSS rule?
Flobi Jan 28
I am just curious what happens if I do put something in the box below.
Flobi Jan 28
(By the way, that would be nothing.)
anticapthcian May 18
I recommend http://akismet.com or spam karma 2 for filtering – these are non-intrusive and only make bots (not users) live harder!