December 21, 2001

12:05 PM

Amazon spamming search engines?Looking around for reviews on Cooper tires, I came across 2 links in the top 20 search results on Google for sites that look legitimate, but are actually redirect pages to Amazon.com (which doesn't even sell auto tires!). What's going on here? Since when is it legitimate for large corporations to spam search engines!??
posted by yarf (43 comments total)

The illegitimate links are, on Page 1, the one entitled "Cooper tires" and begins a description "... such horrid weather." The second reference is #1 on Page 2, "Cooper tires - Interested? Your #1 source! Click here." This is not the first time I've seen these kinds of Amazon.com links, and I find them not only annoying, but downright deceptive (since by the descriptions and the URLs, they appear to be legitimate links to relevant content).
posted by yarf at 12:07 PM on December 21, 2001

You aren't really describing what's happening, though ... Amazon didn't make up all that text, it WAS there when google indexed it.

Then the site went out of business.

Then amazon picked up the domain (eg. this one you refer to) and redirected it to themselves.

In a little while, next time the google spider hits that site, the descriptions will be fixed.

Not amazon's fault, and while it may be a little off base for them to buy a ton of domain names, I wouldn't exactly call it spamming.
posted by malphigian at 12:12 PM on December 21, 2001

Doh, sorry about leaving the http off that url, was to mychoicest.com -- was trying to make the point it redirects at the top level.
posted by malphigian at 12:13 PM on December 21, 2001

Registrant:
4-quality.com
13426 Maxella Ave Suite 496
Marina Del Rey, CA 90292
US

Registrant:
mychoicest.com
2554 Lincoln Blvd Suite 277
Marina Del Rey, CA 90291
US

Registrant:
Amazon.com, Inc (AMAZON-DOM)
1516 2nd Ave
Seattle, WA 98101
US

Appear to be different companies. Although both mychoicest.com & 4-quality.com point to amazon.com
posted by riffola at 12:16 PM on December 21, 2001

Damn I was searching the whois and didn't see malphigian's comments.
posted by riffola at 12:17 PM on December 21, 2001

I don't think that's it. Type in the filename into Google:

http://www.google.com/search?q=cooper_tires.html

Leaving out the last entry, we see 4 entries (all with no cached pages available) that all redirect to Amazon.com, and all that have the same directory and naming structure. Since Google updates its db every week, how is that these files have remained indexed?

Look at the DNS record for 4-quality.com:

Record last updated on 08-Dec-2001.
Record expires on 23-Oct-2002.
Record Created on 23-Oct-2001.

A site was created and in about 5 weeks' time, it was taken down and bought by Amazon.com? Gee, the same thing is true with nothing-better.net:

Record last updated on 10-Dec-2001.
Record expires on 23-Oct-2002.
Record Created on 23-Oct-2001.

And gee, look a that, the DNS records all have people who enter in 000-000-0000 for their telephone number?

No, I think something more nefarious is going on here. This is no simple coincidence.
posted by yarf at 12:24 PM on December 21, 2001

The dates and phone number and shell addresses are also the same (or similar) for the other 2 domains. It appears temporary dummy sites are being created, submitted for indexing, then redirected after indexing has taken place.
posted by yarf at 12:27 PM on December 21, 2001

even loan-rates-auto-new-car.com does the same.
posted by riffola at 12:31 PM on December 21, 2001

I should add that loan-rates-auto-new-car.com appears to be a meta domain. Spam the search engines with keywords in the domain itself.
posted by riffola at 12:33 PM on December 21, 2001

I'm guessing it's people redirecting to Amazon to get the referral money.
posted by peterme at 12:35 PM on December 21, 2001

Are you suggesting that google completes an entire re-index of the internet each week? Where did you get that info?

Not to mention Google would have to be complicit in Amazons spamming in that case.

What if all the phone numbers and addresses were formerly the same because Amazon bought a batch of domains from some squatter in Marina Del Ray? That group could have been the one to put the Cooper tires site everywhere.

Still not buying the conspiracy, sorry, need more evidence :)
posted by malphigian at 12:35 PM on December 21, 2001

Some more detective action. Both of those addresses listed are boxes.

2554 Lincoln Blvd == Marina Postal Center
13428 Maxella Ave == Mail Boxes Etc.

Also,

loan-rates-auto-new-car.com
1427 17th Street B
Santa Monica, CA 90404

nothing-better.net
827 Marco Place
Venice, CA 90291

But those don't seem to match up to any mailbox place. All the address are very close to each other.
posted by smackfu at 12:51 PM on December 21, 2001

www.career-jobs.com/jobs-cat/michigan_employment.html is your extra conspiracy evidence. I was looking for job information on google because I'm interested in switching majors. Anyhow, the above URL re-directs to Spiegel. I thought perhaps going to the root URL would yield more information: www.career-jobs.com however, surprisingly enough, redirects to amazon! And even more surprised was I when I logged on to metafilter and saw that it was actually being discussed! I saved the URL, (it's only been about a week or two) because I wanted to tell someone, but who am I going to tell? My friends and family don't care or understand what I'm raving about, and I don't know who can answer me why it's set up like that, but I am awful curious.
posted by banished at 12:57 PM on December 21, 2001

The way I understand it, Amazon.com would never be paying referral fees for this sort of thing, so its doubtful these are from independents looking to get referral fees. I can't find any redirect response code either being sent on to Amazon which would track the referring fee (which is typically how they pay those fees). It's a mystery to me, but I'm going to keep plugging away to try and figure it out.
posted by yarf at 12:59 PM on December 21, 2001

Since when is it legitimate for large corporations to spam search engines!??
since when is it required of large corporations that they behave legitimately? business is business, right? they have a right to innovate, right? god bless america, right?
posted by quonsar at 1:15 PM on December 21, 2001

If innovation means making search results less relevant to the end user, guess what that means? I will be boycotting any company that uses this methodology to try and get my business.

I mean, this is bad for so many reasons, but one is that it legitimizes all those "Improve your search results!" businesses that cropped up overnight a few years ago to help you try and "trick" the search engines into placing your site above others in results. Worse, the search I was conducting has nothing whatsoever to do with Amazon.com's business! I could maybe see it if they even sold auto tires, but they don't.

I'm getting a header response code of "200 OK" when I plug one of these "redirection" URLs into some perl code which spits out all the technical information in an HTTP request. If this were a redirection, that should be a 3xx redirection code, which makes the mystery even more intriguing...
posted by yarf at 1:18 PM on December 21, 2001

This is a fascinating discussion. Both from an educational standpoint and looking at the 'Online Internet Mystery'. Thanks everyone!
posted by thekorruptor at 1:20 PM on December 21, 2001

It is a referral. The first redirect goes to:

http://www.amazon.com/exec/obidos/redirect-home/interestingfa-20

That then redirects you to the main home page after putting some money in the referrer's coffers (in this case, interestingfa-20).
posted by dlewis at 1:23 PM on December 21, 2001

Shit - I shouldn't have linked that. Don't bother clicking that link above if you don't want to generate more fees for this crafty individual...
posted by dlewis at 1:25 PM on December 21, 2001

Blah... I guess mystery solved then. Stupid bastard referral idiots. I can't believe Amazon.com puts up with this kind of stuff. Thanks dlewis!
posted by yarf at 1:26 PM on December 21, 2001

[dang, dlewis beat me to it...]

All of the above domains redirect to http://www.amazon.com/exec/obidos/redirect-home/interestingfa-20

"iinterestingfa-20" is a referrer code. Whoever they are, they're making money off of this.

So yes, it looks like someone sets up dummy sites to be indexed by the spiders, apparently using Google's NOARCHIVE meta tag. They may have simply changed their pages after being indexed, but it's more likely they use search engine cloaking.
posted by whatnotever at 1:27 PM on December 21, 2001

"My friends and family don't care or understand what I'm raving about, and I don't know who can answer me..."

We have a new Metafilter tagline.
posted by luser at 1:30 PM on December 21, 2001

Oh, and yarf, they are 302's...

$ telnet www.mychoicest.com 80
Trying 209.132.252.32...
Connected to www.mychoicest.com.
Escape character is '^]'.
GET / HTTP/1.1
HOST: www.mychoicest.com

HTTP/1.1 302 Found
Date: Fri, 21 Dec 2001 21:15:57 GMT
Server: Apache/1.3.22 (Unix) PHP/4.0.6
X-Powered-By: PHP/4.0.6
Location: http://www.amazon.com/exec/obidos/redirect-home/interestingfa-20
...
etc.
posted by whatnotever at 1:33 PM on December 21, 2001

Send an e-mail to Google. They will determine if something is fishy.

One thing some web sites do is create "gateway" pages. Such a web site uses a server-side mechanism to detect whether the entity viewing the page is a person or a search engine. If it is a search engine, then a page specifically designed for that search engine is served up, with the intent of increasing its rank in search results. If the page viewer is a human then you see the normal page. Google does not look kindly upon such practices.
posted by fleener at 1:45 PM on December 21, 2001

It gets weirder.
posted by Mid at 1:51 PM on December 21, 2001

The netblock owner for all of those sites is Alchemy Communications. Based on their customer list, I'd imagine that they either condone or partake in this exercise.
posted by machaus at 1:55 PM on December 21, 2001

I think you only get a referrel fee off amazon if the visitor you redirected to them actually buys something.
posted by poodleircus at 1:58 PM on December 21, 2001

Talking of craftiness, perhaps someone can help me out with an Internet Explorer problem. Whenever I try to access a page that doesn't exist (like http://www.lkjasdklfjlisadilnasdvi.com ... does that exist?) I get redirected to a search page owned by LoopUp.com. It pisses me off because they earn banner ad revenue every time I type a URL in incorrectly (which for me is pretty often). I managed to change the default search page back to Google, but I don't know how to change this "not found" page. Anyone else having this problem?
posted by dlewis at 2:18 PM on December 21, 2001

I get redirected to a search page owned by LoopUp.com.

Err, you're getting what now? Are you on a network or at home on a standalone machine?
posted by yerfatma at 2:26 PM on December 21, 2001

dlewis: did you try checking/unchecking "Show friendly HTTP error messages" under Advanced in Internet Options?
posted by riffola at 2:31 PM on December 21, 2001

So, I'm not completly sure I understand this. Is Amazon involved? Or is this some guy with clever code?
posted by Greener at 2:31 PM on December 21, 2001

or both
posted by Greener at 2:31 PM on December 21, 2001

Some guy with a clever code
posted by riffola at 2:33 PM on December 21, 2001

Also, where did you find the redirect code...I looked but couldn't find it. In semi-toned down tech talk..what's going on. I mean, I'm a web designer...but I'm not fully understanding this.
posted by Greener at 2:35 PM on December 21, 2001

Greener: I don't think Amazon's involved in this scam.... but you never know.

Ok, now this may seem a little crazy but... if I try to load a page which results in a 404, for example http://www.google.com/a, then I'll get a 404 error page from Google. But... every five or six or so of these 404's, I won't get a page from Google. Instead I'll get a custom 404 page from some company called LoopUp.com that says that the document I was looking for could not be found and would I perhaps like to look for it with their search engine or buy a domain name off them or perhaps click through their banner ads etc. This only happens with IE.

I don't know how it got there, whether it was a piece of Javascript or a Gator-style module maliciously tagged onto some other piece of software, or maybe from just looking at too much porn. But I can't get it to go away. Not even by turning off the "friendly HTTP errors".

PS Greener, I used the wget command line utility to request the redirected page. It prints out some debugging information about what it's doing when a redirect happens. Whatnotever used telnet - another way of doing this.
posted by dlewis at 2:57 PM on December 21, 2001

dlewis: i had the same problem with my IE (5.5, Win98), though I got redirected to a different search engine. A reinstall of Win98 got rid of those shenanigans.

I was wondering if this is something more serious. I can't remember any script changing the settings on that (and I would, cos theoretically it should ask me if a site wanted to make changes to my settings).

Very odd...
posted by slater at 2:58 PM on December 21, 2001

The 404 redirection issues you guys are talking about seem to be most likely related to spyware... there are a number of them that do things very much like what you describe.

lavasoft ad-aware is highly recommended!

Also, on topic, whatnotever, good job solving this one!
posted by malphigian at 3:32 PM on December 21, 2001

Since when is it legitimate for large corporations to spam search engines!??

What rock've you been hiding under Yarf?
posted by xochi at 3:54 PM on December 21, 2001

Doesn't Amazon alter its cookie when you go there via refer, and if you buy anything at Amazon, within a certain specific time (days? weeks?) of that cookie being planted, pay off?

If you figure that 1% of your people hitting your engine-spammed refer page will buy something at Amazon with an average purchase price of $25 within the specified time ... and the refer fee is $2, then 10,000 hits a day is $200 a day in revenue. Not great, but not bad, either.
posted by MattD at 8:28 PM on December 21, 2001

This is somewhat curious and amusing so I decided to check the sites using the Google's "link:www.domain.com" option, and here is what I found:

link:www.mychoicest.com

isalesmarket.com
Description: New Brunswick online photo listings. Buy, sell, find houses, cottages, land, commercial real estate,...
Category: Regional > North America > Canada > New Brunswick > Business and Economy
www.isalesmarket.com/ - 6k - Cached - Similar pages

link:www.career-jobs.com

isalesmarket.com
see description above
www.isalesmarket.com/ - 6k - Cached - Similar pages

So I decided to visit the site:

unavailable

So this made me curious, so I decided to check DNS information:

Record last updated on 09-Dec-2001.
Record expires on 29-Oct-2002.
Record Created on 29-Oct-2001.

As has been noted above, I would have dismissed this as a freak occurence, but the consistency is un-nerving.

Also, there may be some people (like myself) who were unable to figure out how to disable IE's choice of a search engine for the longest time, Googlify your Browser!.

posted by sarosh at 9:00 PM on December 21, 2001

I think 1% is an order of magnitude or more off. I would think the number of people who purchase after following that link is probably closer to 1/10 of 1 percent-- if that--

Amazon doesn't pay for page views, just purchases, right?
posted by fncll at 10:26 PM on December 21, 2001

Amazon's referral program only pays for purchases.

These Cooper Tires links would not be so high in Google's rankings unless the sites (and pages) were legitimate at one time. What's happening here: A company surfs Google (and probably other engines) looking for domains that are (a) still showing up in the database, and (b) no longer registered. They register the domains and set them up to refer all requests to Amazon, using their referral ID.

It's not the best money-making scheme in the world, since Amazon referral pay has been decimated over the last year, and the search engines will figure out quickly that the pages are no longer present. If enough people complain, Amazon will probably delete the affiliate membership of user "interestingfa-20".
posted by rcade at 3:20 PM on December 22, 2001

To answer this question about IE search pages, the answer is in the registry:

From Mark Hensler, an admin at webmaster forums.

"It's a registry hack...

HKEY_CURRENT_USER\Software\Microsoft\Internet Explorer\Main

Set 'Search Bar' to the page you want displayed in the side bar.

Set 'Use Custom Search URL' to 1. (0 uses default)

[...]

I did a quick google and found this page:
http://support.microsoft.com/suppor...s/q171/8/53.asp

-Mark"

hth,

:) S
posted by Suzanne at 5:09 PM on December 22, 2001