Tuesday, March 29, 2011

Of Monster Trucks, Blogger Blogs, And Spam Classification

Jerry (we will call him that, since we don't know his real life name) loves monster trucks. He drives a monster truck, he goes to monster truck shows, and he publishes a blog about monster trucks. One of Jerry's friends is Sam (no, we don't know Sam's real life name either). Sam publishes another blog about monster trucks.

Recently, Sam's blog about monster trucks was locked, for suspected spam hosting, by the Blogger anti-spam classification process. Jerry, who reads Sam's blog when he is not working on his own blog, became unhappy. Jerry added a second blog, containing a list mentioning Sam's blog, and a few dozen other blogs about monster trucks that have also been locked as suspected spam hosts, to his portfolio.

Jerry hangs out in Blogger Help Forum: Something Is Broken, watches for anybody reporting a problem with a blog about monster trucks, claims that Blogger is discriminating against monster truck blogs, and offers his new blog as proof of a conspiracy by Blogger / Google. Jerry's latest blog is, predictably, "blogger-hates-monster-trucks.blogspot.com".

Since Google is based in California, California contains at least half of all of the world's econazis, and econazis hate monster trucks, it stands to reason that Google is going to hate monster trucks - and they are going to discriminate against blogs, published in Blogger, about monster trucks. Right?

Wrong, Jerry.

Here's the current spam classification advisory note, which is subject to change:

Given the fact that automated spam detection is not yet a perfect science, Blogger Support is always willing to investigate any reports of false positive spam reviews, politely posted here in the forum.

That being said, before you request an appeal, it's worth describing a few examples of what Blogger regularly removes, as part of its zero tolerance policy to abusive / inappropriate content:
  • Affiliate marketing.
  • Content created with scripts and programs, rather than by hand.
  • Content or links referencing referral-based activities such as GPT ("Get Paid To"), MMH ("Make Money from Home"), MMF ("Make Money Fast"), MLM ("Multi-Level Marketing"), PTC ("Pay To Click"), or PTS ("Pay To Surf").
  • Content scraped from other blogs / websites.
  • Copyright Infringement.
  • Large blogs with multiple, unfocused / unrelated subjects.
  • Links to Illegal Downloads / Streaming / Torrents.
  • Links to Account Hacking / Password Distribution sites.

Please observe that the advisory note does not mention blog subjects, because - with the exception of "Affiliate marketing / referral based activity" - subject category is not part of spam classification. The fuzzy Blogger anti-spam classification process looks for blogs with abusive content, not for blogs about monster trucks - or even about trucks, or cars, at all.

The nature of fuzzy classification, unfortunately, means that there will always be some false negative classifications (some spam blogs will occasionally not be classified as spam) - and there will always be some false positive classifications (some non spam blogs will occasionally be classified as spam). Both false negatives, and false positives, are as inevitable as they are unfortunate.

The subjects of blogs produced by spammers are going to resemble the subjects of blogs produced by ordinary bloggers, because spammers want their blogs to look like ours. If ordinary bloggers publish blogs about monster trucks, normal trucks, or even cars, spammers will follow. Anti-spam classification false negatives, and false positives, will include blogs of those, and other, subjects.

If your favourite blog, about monster trucks, normal trucks, or even cars, is falsely classified as a spam host, that's unfortunate. If you see other blogs, of that subject, similarly falsely classified, that's also unfortunate. But seeing either should not mislead you to the conclusion that Blogger is actively persecuting you - and publishers of other blogs - as you publish your blogs about monster trucks, normal trucks, or even cars.

Having your blog classified as a spam host should not cause you to complain about unfair practices by Blogger. Blogger is trying to keep the level of spam content down, and help ensure that our blogs will be listed in search results by search engines other than Google - and not falsely classified as spam hosts simply because BlogSpot is known as a spam haven. The problem starts with spammers, not with Blogger / Google policies.

Get real, Jerry. Help keep the noise level down, in the forums, so we can help more people with actual problems.

>> Top

4 comments:

Jeff said...

This is Jerry. lol
A blog has not violated anything until it has been reviewed. When a blog is closed by a bot, it is under suspicion, but not guilty. All I want to do is get a review on Monster Truck blogs.

We all want to speak to employees. You are the closest thing to an employee of Google/Blogger that I can find.

BTW, I don't hang out in the Help Forums. These bloggers email me with their thread so I can follow.

Chuck said...

Well, Jerry,

If your blog has been spuriously classified as spam, you are entitled to have it restored - but you now have to learn The Blogger Four Step Spam Review process.

If Sam's blog has been spuriously classified as spam, Sam is entitled to have it restored - but in this case Sam has to learn The Blogger Four Step Spam Review process.

Sorry, Jerry, Sam has to deal with his own trial, on his own. You may attend the trial, as an observer. But you must remain silent, during the trial. Observers have no right to speak.

There is no "we", Jerry. You have your blog, and Sam has his. You don't speak on his behalf, and he doesn't speak on yours.

Hopeless said...

Normally blog post has been remove because of too many active links in the comment itself.Another reason is they are posting generic comment like " good job,great post and thank you ". Each post should be valuable and helpful.Agree?

Thanks,
Andea from abesthandtruck.com

Chuck Croll said...

Andea,

I called that "Nice Blog" spam, several years ago.

http://blogging.nitecruzr.net/2009/03/why-we-should-all-moderate-comments.html