Spamming on a Wednesday Afternoon


I got a little inspiration from an article that I read yesterday.  7 Enlightening Infographics About Email Deliverability was an article containing some infographics to try to help you understand email marketing and how to ensure that your message gets delivered.  One statistic that was quoted in the article shed a little light on the situation

76.5% of commercial emails sent reached recipients’ inboxes in 2011, and email blocked and flagged as SPAM increased 24%.

It’s a delicate balance between legitimate marketing and out and out spam.  I will confess outright that I’m sort of the world’s best target – I don’t read email from people I don’t know or content that isn’t wanted or solicited.  So, a lot of the work that goes into successful mass marketing is wasted on me.  Whether my spam catcher catches it or not, chances are I’m not going to read it one way or the other.  Besides, quite frankly, these days email is about the last thing that I read anymore anyway.  I’m more disposed to read Twitter, Facebook, and Google Plus messages first and most people who want to contact me are aware of that.  Even to the point where someone who insists on me reading an email immediately will send me a Twitter message directing me to read my email.

In one of the infographics, “History of Spam“, there was a piece at the bottom suggesting words that should not be incorporated into mass marketing messages lest they get tagged as spam.

As I look my way through the word cloud, I see exactly why these words would get flagged.

So, I wondered…how would this stack up against what my system is flagging as spam.  I have quick and easy access to spam resource in two sources.  I get all kinds of spam that’s caught in my gmail account and also in comments to this blog.  What would my own spam word clouds look like?  They’re easily created.  Just head to the spam folder in one tab and open Wordle for creating a cloud in the other.

Let’s do it.

Not that these are statistically correct or not but I don’t take advice from a researcher anyway, so I set my own rules.

Email Spam
I’ll just grab the 10 most recent spam messages and see what turns up.

This shouldn’t be too difficult to do.  Gmail currently has 131 messages tucked away there but I’m going to change the rules just a bit.  I’ll look at the 10 most recent English spam messages…

OK, another rule change.  When the same spam message with exactly the same words appear, I’ll just take the first one…

I seem to have four types of spamish emails.  Gambling (don’t they know they closed the slots at Windsor Raceway?), Watches, Pharmacy, and Search Engine Ranking Promises.

Blog Spam
I still think that it really stinks that people have bots that send spam messages to blog posts.  How low can you go?

Here’s the 10 latest.

There is definitely a different tone to the type of nonsense that gets caught in the spam trap for my blog.  In this case, they seem to be playing on the desire to rank your blog higher.  Nobody ever mentions that contributing better content is probably the best way….

I think it was an interesting comparison.  At the same time, things that are caught as spam would be good in another context.  In the classroom, it would make for some interesting conversation.

In fact, I think that this whole exercise would have a great deal of value.  If you have a class blog, share the text (without the links) with students and let them analyze what’s there.  Sure, there is the simple word cloud like I’m doing here but charts and graphs and their relationship to an algorithm that would rank a message and either classify it as “good” or “spam” would certainly be a great activity.  Akismet can help the cause with an explanation of its service.

It’s a dangerous world out there.  On my computer, I have WOT (Web of Trust) installed on all my computers so that I can get a sense of the values of any embedded links.  It was working overtime as I dug through the spam folders.

Advertisements

OTR Links 05/17/2012


Posted from Diigo. The rest of my favorite links are here.