4 replies to this topic

[grypster]

Something I just discovered that may cause other SEOs some grief.

We use descriptive names for our images. While handy for us, it also ensured that Google indexed all of them which in turn added several GB to our bandwidth to the chagrin of my ISP. So, we set the robots.txt file to block jpeg indexing.

We were perplexed to find that virtually none of our images were showing up in Froogle in spite of the fact that their feed requires the relevant images be included. It turns out that the robots.txt overrides the feed information. The computer whiz I delegate this to agrees this is silly but sees no obvious solution short of writing a script that reads the Froogle feed data & specifically adds them to the robots file.

Has anyone thought up a workaround that is better than this or is this SEO SOL?

[awall19]

I believe this should work (though I have not used it myself)
for more info go to robots txt protocol site

User-agent: (whatever froogles agent is)
Disallow:

User-agent: *
Disallow: /images/

[Randy]

Been awhile since I looked at this issue, but the problem when I did (last Sept) was that Froogle doesn't have its own user-agent. At that time it used the Googlebot-Image spider to grab those images. Which kind of defeats the purpose if you're trying to keep your images out of Google but allow them for Froogle.

If you don't care about that, then use the Googlebot-Image user-agent, allowing it the ability to get to your images, and you should be fine. Assuming they haven't changed anything since then.

It would certainly make more sense to have the Froogle bot identify itself separately from the normal Googlebot's, however that wasn't the case as late as last fall.

There are ways you can keep everybody but Froogle from hotlinking to your images via mod_rewrite, however that doesn't solve the spidering problem you've noted.

[grypster]

Hi Guys,

Thanks for the feedback. My tech did some preliminary research & passed the ball back to me. I am waiting for a masochistic mood to hit me before I tackle this.

But, Randy is quite right that the problem seems to be that Froogle doesn't have it's own image bot. However, Google/Froogle said they would have an answer in a few days. I will report once I know.

[Randy]

Cool. Thanks for the follow up grypster. That could be very useful information, especially if they have a solution to the problem.