Forum Moderators: open
What a different web I saw - this cloaking stuff is apparently much more popular than I ever realized. Among these was webmasterworld.com, there is an extra link to SearchEngineWorld.com on every page when I can see what Googlebot sees. Maybe I'm wrong, but I was under the impression that this sort of thing would get you banned from Google. Someone set me straight :P
Some cloaked sites will actually ban you if you try surfing using the googlebot navigator user agent coming from a nonvalid ip address. Webmasterworld has a very comfortable relationship with the search engines of course, so it's not a problem for them.
Other amusing things are that there are html errors on these pages, they don't validate, brett tabke puts performance and stability above standards, correctly so too of course, it's why this site is successful. One of those errors is currently triggering some kind of bug in Firefox on logon, but I haven't been able to figure it out.
I've been thinking of getting into a form of cloaking using certain different techniques re arranging content of the page more than delivering different content, that's quite easy to do, and almost undetectable if done correctly from what I've been able to gather.
We do not allow the running of rogue and unwelcome aggresssive bots on WebmasterWorld. That is internal policy [webmasterworld.com] and I can not see that it would ever need to change. Protecting the site from them is job one. That is part of what you are seeing and why people on some of the cable networks *must* log in. At that point, they see a different page - a members page.
Same is true for the logged in vs unlogged in member. One sees edit links and other member utilities, while the nonlogged in member (and bots), see a stripped down version of the page.
Then there are the private RSS feeds for every forum.
Then there are the PDA and low impact versions of every page.
Then there are the experimental XML, CSS, and Atom editions/versions of everypage.
Lastly, there is also and experimental on-the-fly 4 language translator available to the beta team.
Like all big sites, I don't see Geo Targetting as an opt out option. We have to do it in this envronment.
So ya, we can roll a page 7 ways to sunday based on any environ variable, but none of it is done on an IP basis (except for the banned ip list .htaccess).
> brett tabke puts performance and stability above standards
No I don't. I think you can have both and thanks for the tip on validation. <*shrug* hang head> - I've not run the validator in awhile. It is SOOOO hard to keep up with that in a dynamic environment. When we put it online two years ago - every page validated. I'll work on it this weekend.
Anyway, back to the topic at hand. Does google allow cloaking? Sure, cloaking in many of it's forms is not against google policy.
Pop Quiz, how many of the following sites cloak:
CNN,
RoadRunner,
Yahoo,
Google,
AskJeeves,
Microsoft,
WashingtonPost,
BBC,
Ebay,
Answer? They all do.
The only time cloaking will get you in trouble is if you are delivering vastly different content to the bot as you are the user.
Cloaking gone mainstream: [webmasterworld.com...]
I think you can have both and thanks for the tip on validation
I'm seeing a problem in the logon page, it's not letting me logon from firefox, it did the first time, but I deleted my cookies for a test, then when I tried logging on again it wouldn't let me, I checked the logon page code and there was/is a missing </font> tag, that was the only real coding mistake I could find, I've seen Mozilla have problems when it comes to missing inline tags. That's firefox 0.8, when I click logon button, nothing at all happens, zero. First I thought it was that somebody just didn't like me, but then I figured if that were the case my IP would just get blocked, so I had to go to choice 2, broken code.
Since I'm now surfing the site logged on and not logged on for the first time, I also saw how differently you deliver the pages depending on logon status, especially on the google news forum.
Logon page works fine on all other browsers. A quick check of this page, 10 errors, but I'm not sure it's the same page being validated. But there's errors and errors, no alt tag is a technical error but not a coding error.
Sorry, I spoke loosely, I meant your choice to stay with a layout that works on almost all user agents out there, using <font> etc, putting near perfect useablity above perfect standards compliance, same decision almost all megasite s make, for the same reasons, which I think are very solid, and respectful to all users.
[edited by: isitreal at 7:16 pm (utc) on June 11, 2004]
Maybe I'm wrong, but I was under the impression that this sort of thing would get you banned from Google.
You are.
[webmasterworld.com...]
Some sites have a legitmate need to cloak/geo-target/whatever you want to call it.
The litmus test should be: Is Google's index better or not without this cloaked page in it? I think you'll find that for most of the big sites that do cloak, the index is better with those sites in it.
My hat's off to any site that does not contain ERRORS but doesn't conform to W3C's standards. Afterall, most of the successful and/or high-ranking sites on the web don't validate ... why follow a follower when you can follow a leader? :)
The litmus test should be: Is Google's index better or not without this cloaked page in it? I think you'll find that for most of the big sites that do cloak, the index is better with those sites in it.
Excellent point. Although I can think of a ton of NON-CLOAKED sites that Google would be better without :)
Or is that by definition unacceptable?
It is possible to use cloaking to help your SERPS in a way that is acceptable to google?
I suspect it's okay to use cloaking to 301 redirect spiders to your canonical URLs, which are the most basic forms stripped of session IDs, tracking info, etc.
This would ensure, for example, that spiders do not perceive one particular page with a hundred different tracking params as 100 different pages. It would presumably help SERPs in that the PR from 100 different links with a hundred different tracking params would be consolidated to a single URL instead of 100 different URLs.
Can anybody comment on whether this is accurate?
Don't deceive your users, or present different content to search engines than you display to users.
Don't employ cloaking or sneaky redirects.
It's pretty obvious that Google has said no cloaking of any sort is acceptable in their webmaster guidelines.
But, they've publically accepted it in the article above. It sounds like Google's in a tough position - they know that some sites must cloak for search engine visibility, and that some cloaking is very helpful to their engine, but they're VERY afraid of telling joe-blow webmasters that it is okay to do.
If you do it, there is a risk. Who knows what the penalty for cloaking your site will be... it ranges from nothing to complete hand-ban from the index.
That said, sometimes it is necessary. If your site is so borked now due to an evil content management system and you're already not in the index, the risk isn't too bad, is it?
I continue to cloak where necessary.
Aren't individual SEM tactics just risk management in the end? Higher risk = higher payoffs, as with most things in life.
