Stephan Spencer's Scatterings

The Scattered Wisdom of a scientist turned web marketing virtuoso

December 2008
S M T W T F S
 << <   > >>
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      

BMW spams Google and gets slapped for it

Google has raised its mighty hand against automotive giant BMW AG by banning the bmw.de site from Google's index. You can verify that this is indeed the case by conducting a site:bmw.de query. BMW was employing doorway pages and sneaky redirects, two no-nos mentioned in Google's Webmaster Guidelines. To add insult to injury, Google engineer Matt Cutts publicly chastized BMW on his blog.

This proves yet again that even big corporations who spend lots of money on AdWords are not immune to the ultimate penalty from Google: a total site ban.

Now let's see if BMW heeds the warning and completely cleans up its act, removing all the spam from all its domains (such as these two pages: http://www.bmwgroup.com/spider.html and http://www.bmw-motorrad.at/at/de/sitemap-optim.html -- yes, I've nofollowed both of these links so I'm not passing any of my link gain to these doorway pages). Hat tip to Philipp Lenssen and Netprofit (the latter is in German).

Posted by Stephan Spencer on 02/06/2006 | Permalink

Comments (0)| Comments RSS | Filed under: Search Engines , , ,            

Mechanical Turk – not exactly new

It is interesting that Amazon.com has launched a new program called the Mechanical Turk where people in the world can earn small amounts of money doing tasks that are not suited to computers/artificial intelligence. Amazon bills it as "artificial artifical intelligence."

So, if you wanted to have a whole bunch of images organized and tagged, you could pay Amazon for the use of this system and they, in turn, would pay the people that are actually doing the work. Human robots in other words. Or if you wanted your blog posts submitted to Digg or Slashdot. Or various other Turk tasks (as imagined by Richard McManus of Read/WriteWeb).

The Turk is an interesting idea, but isn't totally new. The business model is kind of unique, but spammers have been using a "Mechanical Turk" of sorts for quite a while. It is the way they get past "captchas."

A captcha is a graphic with some letters and numbers in it, and is basically a tool that tests whether you are a human being or not. You type the letters and numbers as you see them into a box in order to gain access to a certain web page or function of the website, such as posting a comment on somebody's blog. Here is a captcha taken from the comment-posting form on Matt Cutts' blog:

example captcha

Captchas are only a moderately effective way of stopping blog spammers, because spammers have figured out a way to get humans to help them out for free. The spammers create porn sites and require the visitor to fill in a captcha form in order to access the site. But the captcha the visitor has to solve isn't some random captcha. It is a captcha taken from a site that they are trying to spam. Then, all the spammer does is take what has been typed in and submits it to the blog to post their spam.

So it's really the spam and porn people that are the innovators in this space of artificial, artificial intelligence. Ironic isn't it!

Posted by Stephan Spencer on 12/29/2005 | Permalink

Comments (2)| Comments RSS | Filed under: Search Engines, Blogging , , , , , , ,