iEntry 10th Anniversary
RSS
Feedback
RSS
Bookmark
Twitter
Facebook
Digg
Stumble
Del.icio.us
Reddit
Furl
Google
Yahoo!
It’s always good to hear what Rand Fishkin of SEOmoz has to say and the conversation at SMX Advanced was no exception. Rand talks about the latest tools from SEOmoz including a new analytics tool that not only allows you to track your own statistics, but also those of your competition. SEOmoz has also updated their version of Page Strength. With the economy being the way it is, Rand explains how you, now more than ever, need to create different content for the different visitors you receive. Rand also discusses white hat cloaking or rather, conditional reader editing (Sorry Matt Cutts!). Check out all that and more in Mike McDonald’s interview with Rand Fishkin on WebProNews.
(No Ratings Yet)
Loading ...
i lovet his so much!
Hey Rand, I’m happy to have run across this post - I have been pushing these “GEO-SEO” issues for over a year now, and you articulated very well
one of the problems.
I made this post about 7 months ago in goog webmaster forums:
http://groups.google.com/group/Google_Webmaster_Help-Indexing/browse_thread/thread/15bd4448a7c04395#f0c9c721938be642
At the time, could find little to no discussion of the problem,
which has other dimensions. For example did you know that
google’s KML indexing engine seems to be entirely decoupled
from their regular spider? That has several interesting implications.
My site (gruvr.com, local concert maps) was seemingly being penalized for ‘cloaking’ - or at least goog thought it was all about
concerts in Mountain View! They didn’t get that they were polluting their own
index by simulating a surfer based in one location, or that they were forcing sites to sacrifice user experience to conform to a ‘flat-earth’ textual layout.
Worse, the AdSense spider uses the google location metadata… so even if I was mapping concerts for London, you’d see all ads for San Franciso nightclubs!
This last bit was losing me adsense $ and of course google would be losing much more over this issue, and still is.
HOWEVER the GOOD news is - about 2 months ago, I started to see more spider footprints that seemed to originate from places besides Mountain View… and my
traffic has been slowly building since then. So I think they may have patched it up a bit… did Cutts say anything that would confirm this?
thanks
I think that on camera, when speaking, I don’t always explore a topic with the same depth and finesse that I should, but your criticism is still fair. There are lots of ways to detect bots/non-humans without using the specific user-agent and that’s typically a better way to do it. I wrote a little more about this today on SEOmoz.
I don’t, however, appreciate the false insinuation that we do user-agent cloaking on any of our client sites (or maybe you meant SEOmoz, but we don’t have ads there). I’d say that’s unfair and unwarranted.
Umm yeah… not cloaking… that’s content delivery and you can do that without detecting google in fact even for Rand’s purposes there is NO GOOD REASON to test for specific engines. In fact the proper implementation and content dselivery “rule” should apply to any “agent” response that has the same shortcommings in it’s capability to use the site… it’s just stupid to do that checking for specific engines because you are risking the sites “trust” with SEs NEEDLESSLY. There is no need to detect by SE. You only need to test the capabilities of the requesting “agent” so…. no such thing as whitehat cloaking… I agree with Matt if you have a clue how a crawler works you can achieve the same outcome without doing it by detecting the engine… but then I’m not trying to promote content delivery as a cloaking technique to sell ads on a questionable blog! Rand.. If you are wondering about those weird bots… perhaps that’s what they are checking for… your clumsy attempt at cloaking!