Google Toolbar = Alternate 'Spider'?

A question on whether the Google Toolbar could be a 'spider'

[qianxing]

If you guys haven't downloaded the new Google Toolbar, it's pretty cool, with lots of little do-dads and such. One feature is the use of PageRank in seeing the rank of any page that you're at, at the expense of information on your surfing being sent to Google.

Could this be part of why Google is spidering less? If people start using the toolbar and sharing with Google their surfing, Google could check out sites without ever having to send Googlebot. Why not let the surfers do all the work for them?

What could this also mean about stealth and cloaking when Google sees exactly what the surfer sees? It's pretty interesting stuff.

What do you guys think?

[robrob]

If they aren't doing it, they should be. Does anyone know if the page info gets sent to google every time? This would really make things hard for cloakers.

[littleman]

If they are, it's brilliant.

[robrob]

What I should have said is that it will make it hard for cloakers to fool the engines. Cloaking should still be O.K. if it is topical from what I hear. Can anyone comment on Googles attitude towards non-deceptive cloaking?

[qianxing]

I don't know of any official releases by Google on cloaking specifically, but they haven't seemed to care as long as it's relevant. It would be interesting to see if they have an official opinion.

-qianxing

[seoboy]

I'd say that if they *are* doing it and not telling us, its a *huge* invasion of privacy.

Maybe they could send something basic like "last modified date" or "file size", but sending the whole page would be overkill and a *huge* bandwidth suck.

My vote is no chance they are doing it. The only thing they *may* be watching is browsing patterns (ala Alexa) but if they are, I'd expect them to clue users in on this too... otherwise - privacy concern.

Just my $0.02

seoboy

[NFFC]

This paper, SearchPad: Explicit Capture of Search Context to Support Web Search [www9.org] by Krishna Bharat is worth a read.

Although it talks about a different product [a type of bookmarking system] there are some elements that have strong parallels with the Google Search Bar, in particular the collection of usage data without the server overhead of traditional click tracking.

[rzfree]

I've put two test pages on different domains, crosslinked, no other incoming links, and opened them with toolbar on. See if this will trigger googlebot visit. Can't determine what exactly it sends to the owner, but at least not page contents - i would notice on my system when large amount of data is transmitted.

I like the toolbar's page-rank feature, performing search is little faster too - no need to open Google page.

Rytis

[sean orourke]

>>> NFFC's good stuff >>> ...in particular the collection of usage data without the server overhead of traditional click tracking.

I've often wondered if the search engines might leverage something of a SETI variety. With all the CPU cycles attempting to find intelligent life on outer space, why can't we throw some distributed processing power into finding more intelligent life on earth, too?

I have high hopes for Google's toolbar. The current state of click-tracking from search engines is inadequate. No matter how much they refine their algos, places like DirectHit will never know that I type ClickZ.com into my browser bar 5 days a week, 52 weeks a year. With a little bit of tracking it should be entirely feasible to start floating the very best sites towards the top.

<speculation>Google just might be the search engine that turns some more "pure" SEOs into SEO/Marketing/Content/Strategy jugglers </speculation>

[qianxing]

I dunno. I have a bad feeling about it. Call it a hunch, but I think this will get used against SEO.

[BrettLaCroix]

I found this tool bar two weeks ago...

Installed it... thought it was cool!

Surfed around my online mall... watching the page rank bar light up.

This week... my mall is no longer listed in Google!

Bummer...

Coincidence, probably.

But now, this cool tool bar is now a "slap in the face" anytime I view a page of my mall and see the PageRank bar now completely dark.

:( Brett
<edit: no signature urls, please. -rcj>

[NFFC]

sean_orourke,

Glad you found it interesting, even more so is that Krishna Bharat's employer is Google Inc.

[Fusioneer]

The paranoid among you will have noticed that when you are using the Google toolbar TCP sessions are opened to:

services.google.com:80
www-su.google.com:80

There is a "intentionally blank" page at services.google.com and www-su.google.com points to their front page.

Speculating that the toolbar sends pagerank info (ie. registering page visit) to services.google.com and gets search results via www-su.google.com - perhaps they use different URL so they can track who is searching from toolbar as opposed to straight through the web.

I have also noticed that even after I have turned the toolbar off there is still a TCP session open with www-su.google.com. Not confirmed but make your own conclusions.

[Brett_Tabke]

Wait a sec, are you guy suggesting that the Toolbar is tracking you even when you AREN'T using it?

I just threw google in my hosts file and setup a packet sniffer and surfed a bunch of sites. The only time it tried to get to google was when I used the toolbar.

To me, it is no different than click tracking.

On a related note, where did the option on Google.com preferences go that you could turn on the little "bar" for the same PageRank display as in the toolbar? It used to be there, but I can't find it right now. Did they take that away?

[jeremy goodrich]

Google only spiders less on pages which are dropped, and or banned from it's index.

The toolbar has nothing to do with it.