Spam Rapidly Increasing In Weblog Comments

dsurber writes "BBC News has a nice article discussing 'flyblogging', the phenomenon of spammers leaving advertising-related posts on personal weblogs. The writer comments: 'None of the other blogs I contribute to or run has been affected yet, but I can only assume it is a matter of time before the spammers move in, as they did first with UseNet and then with e-mail. It depresses me to think that any open medium can be so easily undermined by people with no scruples, no sense of responsibility and no idea of the damage they are doing.'" It seems a little surreal that people are having to develop anti-spam weblog tools.

Comments

I have a quick and dirty solution.

[Ignorant Aardvark]

Use the same type of human verification system that Yahoo uses when signing up for an e-mail account. If you can't type in the mangled letters in the image, then your post to the weblog is ignored. This would only be required for anonymous postings - if you're logged in, presumably you've already passed the human verification test upon account creation, so you don't have to go through the hassle each time you want to post.

Re:I have a quick and dirty solution.

[GeorgeH]

That's called a CAPTCHA, and James Seng wrote a Moveable Type plugin to do this with MT. CAPTCHA stands for Completely Automated Public Turing Test to Tell Computers and Humans Apart, you can read more in this story

The article misses the point

[HealYourChurchWebSit]

The BBC article misses the point, as does a similar article in Wired. Seems the editors are more focused on name-dropping and doomsdaying than on focusing on some recent solutions. For example:

• Jay Allen's MT-Blacklist" plug-in;
• Simon Willison's Blacklisting Comment Spam;
• Scripty-goddess' Anti-Comment Spam Tactics;
• BurningBird's Comment Spam Quick Fix;
• Kalsey Consulting's ideas on dealing with Comment spam; or
• even my own observations and suggestions on the topic

Point is ... perhaps we'd all be better service if said articles spent less time on the hype and a bit more investigation on some of the solutions ... whether they succeed or fail ... as both are educational.

Just so long as no one attempts to use a rather evil solution I discovered here on /... ... that would be wrong ...

Re:I've Noticed

[brianosaurus]

I think the ads in the blogs are going for better Google PageRank scores, rather than for direct exposure. Most blogs don't get a whole lot of traffic, mostly just family and friends, if even that much. Only a very small percentage of that audience will click, and they surely won't fall for it more than once.

But google reads a lots of blogs. If a spammer gets their link onto a whole lot of blogs, Google PageRank would see hundreds or thousands of links to their site and bump up its rank. They exploit everyone's blog in order to improve their score on searches.

That's the theory anyway. Whether or not it works is another story.

Re:Google?

[Lagged2Death]

My hobbyist project was picked up by Google after a while, but it wasn't until I retroactively changed my comment signature here on Slashdot and on Kuro5hin (thereby creating many links to my project page) that it went to the top of the search results. It wasn't my intent to subvert Google in any way - I was quite surprised by the dramatic result.

There have been some less-than-scrupulous advertising companies in the business of that publishing dummy machine-generated web pages to exploit this trick. The dummy pages were typically filled with repitions of some nonsense paragraph, with self-links (to other dummy pages) and client-sponsored links interspersed here and there. The idea was that the self-linking would make the site look like a large, legit site to Google, which would mark it as relatively well-trusted and influential. Then Google would dutifully note the client-sponsored links and rank them highly. I believe Google has worked on ways to stop this; I don't know how successful they've been, or if the dummy-site makers are still around.