Forum Moderators: open
There are 5 types of cloaking:
All five have unique applications and purposes, yet all 5 can fit nicely within one program.
Cloaking is the gate keeper that serves your site in it's best light, and protects your custom code from prying eyes.
Search engine cloaking is just one aspect of a much bigger picture. This is why search engines can't even consider banning cloaking. It is so widespread and pervasive, they'd have to delete 1/4th of the domains in their indexes - those would be the best sites they have listed.
Any time you hear a search engine talking about banning cloaking, listen to them very closely -- and remember. If they'd bold face lie about something so pervasive, what are they doing with the really important stuff? They can't be trusted - nor can those that are out here carrying their water.
With the assault of rogue spiders most sites are under, the growing trend of framing, agents that threaten your hrefs (smarttags), I think cloaking has a very bright future. The majority of the top 2000 sites on the net use some form of the above styles of cloaking (including ALL major search engines).
If it is good enough for them, it is good enough for us.
Like most people I'm still thinking should I Try or Not. but maybe it's time to start hitting Air's forum more often to try and understand cloaking some more
Really we need a new word to describe customized content serving, I actually don't like the word cloaking, it makes it sound sinister from the get go.
Tigger you know I'd love to have you visit this forum more often:)
Please note: none of this is meant as "cloak bashing", at least not when the term "cloaking" is used loosely (as in this thread). To me, the word cloak (when not referring to a garment) means "covering or concealing"; I believe that much of the confusion around this topic comes from the fact that some people mean different things by it from others. This is often a problem on Usenet and in discussion forums.
Trying to conceal the real content of a Web site (often unseemly or irrelevant subject matter) by cloaking techniques is what Air describes as [webmasterworld.com] "The Bad" in an interesting his interesting WebmasterWorld post. I'm at one with Air on that.
I don't consider that delivering the same content in WML and HTML based in HTTP_USERAGENT is concealing, nor is delivering a different stylesheet (assuming it doesn't conceal content by setting words or links to background colour).
There's a grey area around (for example) language and geography based delivery. HTTP headers convey language preference better than IP guessing in my opinion, as country of residence (or Web cache) is no guarantee to preferred language. Geography based delivery could lead to potential spider blocking, which some people invest considerable time in avoiding (Fantomaster comes to mind). While the level of attention given is impressive, I'd rather spend my time trying to acquire quality links. Anyway, many of us in the UK find it irritating to be redirected to UK versions of search engines and would rather just have the choice.
Session based tracking can be a SE nightmare for some people, but I find that letting spiders _and_ human visitors crawl the flat structure, while allowing people to get into a session based area when needed (personalisation, shopping cart, etc.) to be a useful approach. robots.txt helps to avoid duplicate listings and visitors with duplicate CartIds.
HTML has its own mechanisms for providing client scripting, Java or Flash, so I wouldn't feel the need to use UA or IP cloaking there. NOSCRIPT, NOFRAMES and OBJECT contents are just more flexible versions of image alt text. There's a view (one that I share) that we'd all have been better off had something like OBJECT been used for images and frames, but Netscape (used to think that they) knew best on such matters. For CSS, @import certainly has advantages, but trying to predict every useragent combination is the road to madness. The same goes for Javascript, where document.all is generally far more effective than a bunch of "if"s.
Judging from Brett's Knowledge base excerpt [webmasterworld.com] and Air's the good, the bad, and the ugly [webmasterworld.com] from May, most of the reasons for using cloaking (as opposed to techniques) are to do with poor use of HTML concerning browser compatibility, Flash, client side scripting, "design an appealing page" and inaccessible use of tables.
Unless I'm still missing the point, anyone who aims to produce accessible [w3.org], complaint [htmlhelp.com] HTML that's spider-friendly doesn't have much use. For such an author, the only reasons left seem to be to hide the HTML source from competitors, and that converting IMG alt text to plain text can get slightly better ranking in some engines.
Maybe those of us who like HTML and ordinary text (quick loading without accessibility problems). This isn't meant to disparage cloaking experts (who invest much grey matter on the subject). Rather I don't see that cloaking is particularly useful to people who are happy for their human users to access the same accessible, search friendly content as the spiders.
Cloaking reduces spam
-Cloaking the top 30 results should be a requirement of any SE, imposed by the SE.
-Spammers rely on code theft to build their sites and pages.
-Cloaking reduces spam by not allowing anyone but the SE to see the top ranking code.
Cloaking reduces the need for SEs to keep tweaking their algos
-Once an algo has become "known" is must be tweaked or changed. This costs the SEs huge amounts of money and time.
-SEs do not let you know their algos so, I see no reason why I should.
IF you use an SE to search and find a site that delivers content relevent to your search. You have no right to the code that gained the site ranking. You do have access to the source that produced the page you are viewing.
I find the only people, group or organisation that complain about not having access to the original code are:
-thieves
-spammers
-reverse engineers
The above is what I have learned the hard way from competeing in a highly commercial category.
Don't forget "His Majesty the Baby"...the perpetual sore loser that whines when someone else beats him in the rankings.
Group 2 has two parts, they are professionals.
Part1:
The whiner with a vested interest in a certain topic.
- This whiner relies on their ability to sway the ignorant and/or the gullible.
- This whiner doesn't represent anyone but themselves.
Part2:
- This whiner will use their voice to prop up any idea or concept, as long as someone is signing the check.
I'd rather steer clear of talking about rights, and whether cloaking reduces spam. I'm more interested in the reasons for cloaking than the ethics. For me, the ethics were covered nicely in Air's earlier post [webmasterworld.com].
toolman: I'm not convinced that non-cloakers spend their working lives getting beaten by cloakers. The number of revenue earning search phrases to work at seems endless (at least for the click 'n mortar businesses). Even some highly competitive pure-play fields seem hardly to be dominated by cloakers.
Google's top ten sites for this month under {online casino} don't show considerably different content when accessed via the cache. "Adult" sites may well different, but that's not an area I'm involved with anyway.
Again, please don't see this as 'cloak bashing', I'm keen to elicit peoples' reasons for cloaking. I'm left reeling more reassured that most sites don't need to cloak, most of the time. As for what % I mean by most... ;-)
Me either. But there are those that believe that cloaking is "bad" no matter what. In some cases ( and some keyword categories) it is used to gain an edge over the competition. You'll never know it because the scripts they use most likely will never let you into the cloaked pages. It would take a great effort to defeat most of the better cloaking scripts and it is certainly beyond the grasp or interest of the average webmaster.
I find one of the better uses for cloaking to be for linking to external sites without giving the visitor the opportunity to leave by placing these links in front of them. Why must I have external links? Because that's what the se's want...so I give it to them. Does that make my site any less relevent to the topic/keyword at hand? I think not.
Actually I'm using talent gained from many hours of study and experience to present my site to as many people as possible....that's my job...cloaking is a tool.
How many webmasters have read all those stupid whitepapers all these se guys wrote in college? Should I call them ignorant idiots if they haven't. Gimme a break. Nobody wants to read that boring stuff...unless you goal is to "be all that you can be". Once you understand some of the deeper points of seo, you'll understand the neccesity of cloaking in various forms.
It's one of the best tools in toolmans toolbox :)
I find one of the better uses for cloaking to be for linking to external sites without giving the visitor the opportunity to leave by placing these links in front of them. Why must I have external links? Because that's what the se's want...so I give it to them. Does that make my site any less relevent to the topic/keyword at hand? I think not.
This is why cloaking has a bad conoctation to it.(spelling?)
You are giving the browser entirely different content than the spider, only to manipulate search engine results. This is also one reason the engines talk the way they do about "cloaking".
Mind you, I completely understand ALL the reasons givin above and ALL the reasons a site might HAVE to cloak. I just do not feel that hiding exterior links to other sites is good enough.
If the links are good enough for the spiders, they should be visible on the browser also.
To me, none of this really matters, as I tend to go for highly targeted keyword phrases that are most likely going to be a good, potential customer. This cuts out much of the sites that use this type of cloaking for results.
>anyone who aims to produce
>accessible, complaint HTML
>that's spider-friendly doesn't have much use.
We aren't just delivering HTML - there is XML/RSS Feeds, WML, Various Entity (Character Sets),
Page Languages, Microsoft wants SmartTag exceptions, and Google wants MetaTag exceptions. Most search engines don't report what language they are looking for and you have to guess and redirect based on Agent and IP.
As you said, producing html 4 code is nifty, but even so, you eliminate up to 20% of your users who aren't html 4 compliant.
>I don't consider that delivering the same content in WML and HTML based in
>HTTP_USERAGENT is concealing
Can't deliver the same content. I am generating content here that is available in certain formats for particular agents.
You can't see that without the right key to the lock. Part of that is specific generation for search engine agents and ip's.
As far as I'm concerned, cloaking equal custom or personalized content generation based on any and all available variables.
Search engine spider content generation is just one more aspect to the much bigger picture.
>Cloaking reduces spam
That is a very interesting angle Startup. Sure made me stop and think about. Required cloaking.
>SEs do not let you know their algos so, I see no reason why I should.
That is the perfect argument for cloaking.
>I'm not convinced that non-cloakers spend
>their working lives getting beaten by cloakers.
Case A: One simple page put online 3 years ago:
The page included some techniques I've never read about anywhere else that were learned the hard way (I'm still milking that 97 Infoseek "advanced degree in seo" education for all it's worth).
The page has been pulling in 300-3k referrals a day for the better part of 3 years. That code is the basis for the majority of my seo strategy. If that page 'leaked', it would removing the finger from the damn that unraveled my entire system.
That page walked through several engines due to se partnerships.
Other engines spidered the first engine and also ranked it high.
That lead to a hand full of quality (free I might add) directory listings.
Those in turn led to topic specific directory listings.
Other engines that base listings on links started ranking the page high.
Before the start of this year the whole site was 'running with the big dogs' under some envious kw's.
That I've used some techniques I've never heard anyone talk about before - little micro gems that work. If I'd not cloaked, and kept the page cloaked, I know this site would not be here today.
The overall success of one high quality page where the code doesn't get stolen shouldn't be overlooked.
Take that same scenario times 1 core client and 80 sub-clients and you can see why I believe in cloaking so much.
Case B: An uncloaked site.
Put the site online 3 years ago. In late 99, it was doing 100k views a day off 20k unique at $5-16cpm. The site attained remarkable rankings across 3 engines under top ten kw's.
Within 3 months time, the site had been stolen and replicated 3 to 4 dozen times. The big se's dropped the site completely because they couldn't figure out which was the core site and which was the stolen sites - all they knew is they had dozens of duplicate - baby with the bath water syndrome. One large site even replicated content and then claimed to the search engines that the content was originally theirs and got the se's to drop my site and keep there.
Today that site is lucky to generate 30k a day and scraps for every ranking it has.
Had I cloaked that site completely three years ago, I have no doubt the site would be in the top 100 traffic sites on the net today. As it stands, I am to the point of not caring what happens with it, and have virtually forgotten about it.
Cloaking techniques could have prevented all of that.
I've always felt those SEO's that bash cloaking are those that can't produce rankings in the first place and are jealous of those who still can.
toolman: there are those that believe that cloaking is "bad" no matter what.
That's not my view (some of Brett's motives above couldn't be considered bad, IMO), but some of us do believe that cloaking can be dangerous. Especially if the site gets ranked well. I would not want to have to start again with a new domain.
toolman: Why must I have external links? Because that's what the se's want...so I give it to them.
I'd have absolutely no problem with that whatsoever. I love links...how could we have a "Web" without them. If we believe in "themes" (and to some extent we do), then those links will be relevant to the human visitors. Great!
ihelpyou: If the links are good enough for the spiders, they should be visible on the browser also.
That's my feeling, too. Part of Brett's reason for starting this was my nagging (elsewhere) for people to come up with reasons to cloak. Taken on face value, people around here don't want to use it for the bad reasons (eg. adult sites listed under family entertainment related searches, etc.). So if I'm not against you cloaking, what am I going on about?
My lingering doubt as to whether I'd ever be converted is due to my feeling that much of the time I'd rather see peoples' quick loading, link rich, accessible and browser friendly SE optimised pages than their Flash/Java/text-as-image laiden, browser incompatible, vision-impaired inaccessible, slow loading, confusing to navigate public pages. Maybe that says more about Web design than SEO, though. :)
As you can guess, I never have to to take the WWW unfriendly junk produced by graphic designers and 'multimedia authors' and make it SE friendly. That could be listed as another reason to cloak, I guess.
What I just do not agree with is giving the browsers different links than the spiders.
OR, giving the spiders a bunch of scrambled content taken from other sites with good link pop. I see this ALL the time also.
That site is only doing this to use my sites good LP or ranking for the sole purpose of influencing his/hers own rank. This I cannot agree with. I have seen my Url hidden in code too many times without it being a real link.
This is what gets me about cloaking.
Again, I definitely see all the good reasons there are for cloaking. But, along with the good, there is the bad.
Not to be obnoxious IHY...did you see this code in a cloaked page?
I guess it would not matter to me where the hidden link was. If the pages were not cloaked, you can bet that many cloaked pages out there have it also.
Please understand this is the perception that is sometimes given to sites that cloak. This is also why the search engines are very vague when commenting about the art of cloaking.
>>>You are giving the browser entirely different content than the spider, only to manipulate search engine results.
Why do people hire seo firms? To manipulate search engine results. Mind you this can be done with or without cloaking....so the only difference being whether an individual feels that cloaking is ethical.
Granted the spammer people who cloak, we all loathe them...that's a given in this discussion. I'm concerned with the professional seo's who use cloaking as a tool to achieve the goal of producing the highest ranking possible for their paying clients while protecting their strategies from the competition (it's not just a job, it's an adventure).
Manipulate? Perhaps, but often people just want to optimise a site to do well under it's core themes and subject matter (the 'O').
Beyond that, search engines want to send people to the content they've indexed. Cloaking (of several forms) has the potential to get the engines to send people to content they've not indexed. The word "cloak" makes people think you've something to hide. If it's highly a optimised page that you don't want stolen then fair enough, but I'd be willing to throw a few SEO cloaked-success babies out with the spamdex-fodder bathwater.
Brett:
Non-SEO alternate versions: There is no need for XML/RSS to share a URI with an HTML page, unless I've been completely asleep since March 1999 when Netscape brought out 0.9. I consider Smart Tags to be dead ducks for the forseeable future, but still don't see why the META tag would bother non-XP UAs or engines. Would the Google nocache META tag have effect anything but Google? I can't see it. I would never want to base characterset/encoding/language on IP, but I understand that the accepted methods are UA cloaking by your definitions above so I'll happily give you that one. :)
Hidden algo's: My lack of belief in the perfect argument for cloaking is probably just due to a stubborn view of what client-server interractions are all about. My background makes me biased, anyway I'm trying really hard to stay clear of moral issues...after all I've only been posting on WebmasterWorld for a week or two.
Case A: I almost always go for word spread rather than trying to find a popular phrase to target. If I was trying to sell a top10 keyword related product or service, I might need to wake up to cloaking, but at present I'm happy to slumber on without. Case A is an impressive tale; I'm sure I've never had a single page do anything like that well, but it can also take effort to make a whole site come close to that level of (relevant) referers when you're targeting subjects on nowhere-near top-ten search-word subjects.
Case B is a real eye-opener, thanks. A nice variation of the "if I submit my competitors sites a zillion times a day will I get them banned?" scenario, and probably with much greater chance of success.
I'm not quite sure how a feeling that "most sites don't need to cloak" translates to "cloak bashing". A person doesn't need to be an F1 basher to feel that most car engines don't need spin to 18krpm. I rather like the fact that my 6k limited engine doesn't stall as often as McLaren Merc's do - for the purpose I use it it's better for the job.
I guess I'm getting rather defensive. Time to go home to a chilled bottle of gin. :)
Thanks to everyone here. I know that cloaking can be contentious, this thread has ironed out quite a few loose ends for me (a cloak ignoramous who somehow managed to fret over frequent 'holy grail' style comments and felt left out).
Case A: One simple page put online 3 years ago:
The page included some techniques I've never read about anywhere else that were learned the hard way (I'm still milking that 97 Infoseek "advanced degree in seo" education for all it's worth).
The page has been pulling in 300-3k referrals a day for the better part of 3 years. That code is the basis for the majority of my seo strategy.
Yes! The original secret recipe pages. They are the best and for some reason those pages never fail! Except with some of the newer engines or re-vamped engines that changed their algos for links and click-thrus. Though, some of them can still beat that out. If you had the chance to have a top ranking site years ago, you can still use those pages as a foundation to build on. And for that reason I will cloak them.
When I optimize, I do not USE other's content or link pop to manipulate the search results. I do not hide anything I do from anyone or anything. To me, that is the difference.
If I add another site's link somewhere on the site, the link is read and viewed by all spiders, all visitors, all browsers.
You may find this unbelievable and not true, but No, I would not try to figure out anyone else's code to use as my own. No. Brett's? NO. I do my own thing based on trial and error, tweaking as need be.
My way has worked for me for a couple years now. Have no reason to do anything else. Have no reason to look at anyone else's work. Don't see the time coming when I would.
Many of you know me as definitely having my own opinion on things. Being this opinionated, do you really think I might steal code? oh yes, I have copied some javascript from time to time, but nothing to do with optimizing for search engines.
And no toolman, I am not picking on you either. :)
The hidden link thing just always kind of gets me going......also 1x1 pix or 5x5 as links only.
I am not advocating the changing of content. The SEs are given the same content as the visitors.
You are not cloak bashing, your questions and view points are more than welcome.
Now where did I put my car keys.
Added:
Smart tags are going to be used. Do you think MSN is going to list and or rank a site that is disabling these tags.
I do tend to lock my door, but right now the car is almost certainly unlocked. I may be only 40 miles from the nearest city, but the Scottish Borders is a long way in terms of crime.
Althogh I don't lock the car here (only 2 neighbours within a couple of miles), I do at home (just outside a medium sized town).
If my sites lived in "real estate", "casino" or "adult" fields then I might be more inclined to lock them. Problematic industries seem to be the biggest reason to cloak. We've not suffered from the goings on reported with those kinds of search terms, but then we don't use "secret SEO tricks" either.
I don't carry a firearm, but if I was a Moscow debt collector I probably would.
I do check for stupid passwords on our server, and I like to know that Linux patches are applied speedily. It's a matter of concentrating on getting maximum return. The danger of sendmail getting compromised is considered high; the danger of getting crucified by people stealing our HTML is considered low.
I am not advocating the changing of content. The SEs are given the same content as the visitors.
That's nice to know. I'd like to think that a robot would compare the contents of the two versions (something similar to the 'lynx -dump' output would do), but they won't because it would be computationally expensive. An alternative would be to get a human to check each 'varying' page. Again, too expensive. If they do anything, it will be to ban or penalise cloaking.
Maybe they won't, maybe they will. Not being in the business of disposable Web sites I daren't gamble.
SmartTags will be with us soon, but M$ have backed down and they're not going to alter our Web pages in the foreseeable (apart from the betas).
