Only cloaking is spam?

[DanK]

OK, so I've seen it at least twice now, Alan believes cloaking is always spam, and only cloaking. My question is, if search engines didn't exist, would it still be considered spam?

Now before you say without search engines there wouldn't be cloaking (I'd have agreed a year ago but I know more now than I used to), there are a multitude of ways and reasons to cloak, the popular cloaking to an engine which would normally be spam even by conservative opinions... could this be used legitimately? Yes, for example google has it's own meta tags, these tags are google only and no other engine needs to see them, if google was the only one with it's own tags, this wouldn't be much of an issue, but there are several others out there and if you start figuring out the bandwidth involved in sending all these meta tags it's pretty excessive.

There is also cloaking to a user, there are a lot of reasons to do this, to provide a page compatible with their browser, to provide a sales pitch according to what you know about the visitor, etc. Can this be used legitimately... you know the answer is yes, and if you need me to defend this example, it's because you don't know a whole lot about engines... we'll go back to google, an engine I KNOW uses cloaking techniques, they use a cloaking technique known as IP Delivery to send users in other countries to different pages with different content on them. Would they do this if it weren't for search engines, sure, but it's still cloaking.

So... why is cloaking the only thing you consider to be pure spam?

[Alan Perkins]

An excellent question DanK. It's Friday evening here in the UK though - curry night :steaming:

I'll write you the answer your question deserves when I have more time. In the meantime, take a look at the White Paper (http://www.ebrandmanagement.com/whitepapers/spam-classification/) - all the answers are in there, but I'll decode it for you here later...

[ihelpyou]

Alan will have a good answer but I will quote Spider-Food's JK Bowman:
Here is my take on the whole argument though:

At the most recent Search Engine Strategies Conference (presented by Danny Sullivan) just about all of the search engine representatives came out and publically stated that as long as the content of the cloaked page was totally on topic, they had no problem with the technology. And the search engine representatives said that they understood that when the technology was used responsibly it could even help them improve the accuracy of their search results.

However, if this statment is true then we have to throw out ALL of the old arguments about what constitutes spam on a non-cloaked page.

Just think about it. On a cloaked page literally everything is invisible. So, why then would it be unethicle on a static HTML page to use:

Hidden Text
Hidden Links
CSS
Hidden Layers
etc....

If one believes that cloaking is ethicle, then we have to throw all the arguments about what constitutes spam on visible HTML pages out with the bath water.
I find it hard to believe that those of you who went to the Search Engine Strategies would even think something like that. JK's last response sums it all up nicely. Where would the engines or us possibly draw the line if "cloaking" was fine and dandy? You could not. NOTHING would be spam if you did. You can find that thread here (http://forums.spider-food.net/topic.asp?TOPIC_ID=2488).

I could not have said it better.

[DanK]

That thread illustrates what I'm saying, cloaking isn't bad just because it can be abused. HTML can be abused, images can be abused, pdf files can be abused, shockwave can be abused, and so on. Because the posibility of abuse is there does not mean the technology is wrong. It's not our fault if an engine can't properly deal with all the technology out there, it's also not our fault for doing our best to compensate for their inability. Spam is spam but technology is neutral. Stereotyping technology is just as ignorant as stereotyping people.

[ihelpyou]

That's not the point Dank. Who or what would police all the cloaking if everyone used it? There would have to be a company started with full time people to check every gazillion page out there to make sure the cloaked page was irrelevant or not. Impossible to draw any lines with cloaking. It is much easier to simply not allow it.

Besides, the challenge to optimize an existing site is much greater and much more rewarding. Cloaking seems to be a cheater's way out. Hell, how hard is it to make an invisible page well optimized? Not hard at all.

[DanK]

Originally posted by ihelpyou
Cloaking seems to be a cheater's way out.

That topic was also covered in the thread you posted the link to.

[ihelpyou]

Dank..... I understand where you are coming from. All I'm saying is that to allow cloaking would be to allow everything. Cloaking is "invisible".

I know you are ethical, as I know Bob is as well. Not the point.

[ihelpyou]

Boy, I sure hope someone discusses "cloaking" at the SES in Boston in March. I will have sooo much fun and make lots of enemies, I am sure. I am NOT shy by no means. :D




I certainly would not have backed down form the "cloaker"s" arguments at that meeting. I would have spoken my peace.

[Advisor]

Maybe you could be the anti-cloaker on their panel like Alan was in SF. Make sure you bring a bouncer with you, though!

Jill

[ihelpyou]

Jill,,...... why did you agree with the cloaking argument that "if it's relevant" it's perfectly fine?

[Advisor]

Where did I say that, Doug?

J

[ihelpyou]

You didn't. At that conference they said that everyone agreed that if a cloaker deliverd relevant content to the spider then it was perfectly fine. You did not speak up! :eek: Maybe you were not there?

[Advisor]

I wasn't there. Heather and I find it very difficult to get off our butts and go to the other sessions! We're bad. I'm gonna try really, really hard this time to go to some!

Maybe the cloaker one was at the same time as ours was...yeah, that must have been it!

Jill

[markymark]

Actually, Doug, the quote was 'all the search engine representatives agreed....'. Unless there's something about Jill we don't know, she is not yet a representative of a search engine.

Cloaking ? I agree that it IS spam and I think Dank is being more than a little disingenuous with his Google example.

Using IP delivery to re-route users from different countries is not my definition of cloaking. And do Google do this - I'm not sure that they do. I can get to Google UK, .com, Japan, Germany etc with no problems. They aren't using IP delivery to re-route me, that's for sure.

It seems to me that advocates of cloaking like to confuse the argument by putting foward good and worthwhile applications for IP delivery when they know very well what everyone else is referring to when they talk about cloaking - and that is delivering one page to the search engines as spider food and another to the visitors. This is spam. Pure and simple.

[Advisor]

Actually, Doug, the quote was 'all the search engine representatives agreed....'. Unless there's something about Jill we don't know, she is not yet a representative of a search engine. Well, yes, actually. I represent Dougie's forums, which according to AOL are apparently a search engine! hehehehehe

As for the cloaking issue, didn't Alan explain the 2 different types of cloaking or IP delivery or whatever the heck it is, back in the thread about his white paper. I know he did, and I'm sure once he drags his lazy butt out of bed, he will point Dank and everyone else to it.

Jill

[ihelpyou]

Yes, he sure did. It is in his post at the top. Oh Dank knows what we are referring to. :)

[Alan Perkins]

OK DanK here we go...

First things first, some definitions. When we say "spam", we are talking "Search Engine Spam". Here's a couple of definitions of spam:

Search Engine Spam: Anything done for the sole purpose of increasing your ranking in search engines.
Search Engine User Spam: Any resource that a user would not expect to receive in response to their query.

In other words, when we talk about Search Engine Spam, we're not simply talking about when you search for Pokemon
and get porn. Most (but not all) people would agree that was spam. That "(but not all)" has to be there because you'll always find someone who will say its relevant, even when it's obvious to you that it isn't. So, who is to arbitrate in these disputes?

The answer is the search engine. That is its role. To arbitrate and sort the pages in its index into an order of relevancy according to the query. The user is effectively asking the search engine to arbitrate and rank when they enter their query. So, how can the search engine perform this arbitration and ranking? What can it base it on? It has to base it on what the end user will see. Nothing else makes any sense at all. Search engine spam occurs when you deliver information to the search engine designed only for the search engine to see, since then you are interfering with the search engine's ability to arbitrate.

I have never said that cloaking was the only form of search engine spam. If I implied it, I didn't mean to. It's far from true.

Search engine spam requires a technology "carrier". There are lots of carriers available. Alt tags, invisible text, tiny text, div text, to name but a few. Most carriers are subverted from some other purpose, usually one designed to make the site more accessible. When subverted for Search Engine Spam, these carriers actually make the site less accessible - a great pity. For example, jamming alt tags with keywords makes a site less useful for blind users or those accessing the site on non-graphical browsers.

I define cloaking strictly as IP cloaking - delivering data only to the search engine, based on the search engine's IP address. Since no end users will ever see this data, it must be search engine spam. Does that make sense?

Now to take your points one at a time:

there are a multitude of ways and reasons to cloakThe way I define cloaking, there is only one way to cloak. IP cloaking. All other ways (e.g. agent-based delivery) I do not define as cloaking, but they may still be "carriers" of search engine spam. IP cloaking is the only carrier guaranteed to deliver search engine spam, since no end user will see the cloaked content. Proper use ofagent-based delivery, or even alt tags, means that some end user will see that content and derive benefir from it.

for example google has it's own meta tags, these tags are google only and no other engine needs to see themAgreed, so deliver them using agent-based delivery. Why go to the effort of maintaining an IP database for Google's robots? (Caveat: Google may consider using agent-based delivery for <HEAD> content to be spam - I do not. But I do for <BODY> content if that content is not designed to be seen by any of Google's users. Or any other seach engine, for that matter).

There is also cloaking to a user, there are a lot of reasons to do this, to provide a page compatible with their browser, to provide a sales pitch according to what you know about the visitor, etc. Can this be used legitimately... you know the answer is yesAgreed, so use agent-based delivery, or any other technology, to do it. Do not use IP cloaking. Anything you do in the <BODY> content should be for the benefit of at least some end users of the search engine. When you use IP cloaking, it isn't.

IP Delivery to send users in other countries to different pages with different content on them. Would they do this if it weren't for search engines, sure, but it's still cloaking.This is a common argument used by cloakers to obscure what they do behind a perfectly legitimate technique (IP Delivery). To quickly highlight the difference, IP Delivery is delivering content according to the visitor's IP, but not specifically targeting search engine robots. IP Cloaking is specifically targeting search engine robots. If IP Delivery is deployed, a search engine robot should receive the same content as a human typical of that search engine's users. If IP cloaking is deployed, at least some of the content delivered to the search engine robot will be unique, and that unique content will be search engine spam.

So... why is cloaking the only thing you consider to be pure spam?Cloaking is the only carrier of search engine spam that is guaranteed to be carrying search engine spam. All other carriers may or may not be carrying spam.

Does that all make sense? If not, answer me this: why do you think search engines consider invisble text to be spam (no matter what is in that invisible text?)

BTW, Search Engines have not stated that cloaked, "relevant" content is OK. Point me to one search engine's Web site that says it. I have been at the SES conferences, and I maintain a dialogue with search engines, and I take a very keen interest in what they are saying. They are certainly not saying "Deliver content to us that will never be seen by our users, and don't tell us you are doing it" - and that's what I mean by cloaking.

[ihelpyou]

That was explained clearly and concisely Alan. Whatever I might not have understood before, I certainly understand now. Good job.

[DanK]

OK, now I think I get it, so wouldn't that mean that SEO is also pure spam, everything an SEO does is in relation to the search engines, would SEO's even exist without engines? I have to say that all SEO's are spammers, and there is no legitimate reason to perform SEO services.

[Alan Perkins]

Hi DanK

SEO is also pure spamNo, I wouldn't say so. Look at the fundamental problem your client has with their site, and it probably comes down to one of the following:

1) Poor technical architecture - the hardware or software makes it difficult to market the site for some reason
2) Poor information architecture - the site hasn't been designed to be marketed
3) Poor accessibility - the site has only been designed to be accessed by a subset of the market
4) Poor copywriting - the words your client has used to describe what they do don't match the words their customers would use to describe what they do.

As an SEO, you should be looking to fix the fundamental problem. That way, you will create a better Web site that deserves to rank higher.

What do you consider the role of the SEO to be, DanK?

[ihelpyou]

Well, I have to say my most important function is to make sure the content is layed out to sell my client's visitor. YES. I also make sure the content is optimized with the targeted keyword phrases my client is going for, BUT, would I or someone also make sure the content sells if there were no search engines? Of course.

[DanK]

I consider the role of SEOs to be improving ranking on search engines, which typically involves making changes to the page to help it rank better, it doesn't matter what you are changing, you are attempting to manipulate the results of the engine. SEOs may offer other services but the service of SEO is fairly specific, otherwise it would be called something more meaningful like sales optimization. Saying SEO is about sales is just the kind of thing spammers do to confuse the issue and justify the use of cloaking, hidden text, link farms etc.

[ihelpyou]

Yes. I definitely would call myself a spammer. All I do is manipulate the search engines with no regard to ethics and morals. It's all the money. :eek: Hasn't everyone seen my hidden text, hidden links, and IP cloaking that I implement? :p

Actually, we or I did not ask for that term.... SEO. I think it should be "web site optimization", because that is actually what we do. We do not optimize the search engines, we optimize web sites. ;)

Just in case the industry does go with that term sometime, I made sure I am high up on it in Google. :)

WSO..... sounds great to me!

[Alan Perkins]

Hi DanK

I consider the role of SEOs to be improving ranking on search enginesAgreed that is the "end goal". The question is, what means do you employ to reach that end goal.

a) any or all available
b) ethical means that demonstrate respect for the search engine and its users

In other words, the ends do not always justify the means.

If you are improving a site's technical architecture, information architecture, accessibility or copywriting, that seems to me to be a worthwhile thing to do, whether or not it helps your search engine ranking. The fact that (if you do it right) it will help your search engine ranking should be because your site is now one of the best available.

Have I answered the original question yet? Have we moved on?

[MazY]

Originally posted by Alan Perkins
Have I answered the original question yet? Have we moved on?

LOL. I gotta love your style!

[Alan Perkins]

Sorry, I don't mean to be patronising. It's just that the original question in this thread was "Why is cloaking the only thing you [Alan] consider to be pure spam"

The debate appears to be opening out into the role of SEO. No problem, but I wanted to make sure that the original question had been answered to DanK's satisfaction.

[DanK]

I'd say I'm content to use your answer here, and to consider the implications of that answer (and probably in the process refine the white paper you wrote a bit). The problem with only cloaking being spam is that there is an entire industry out there that doesn't use the same definition as you, the only time cloaking is just spam is when it's used only for spiders, and even that is questionable when you consider the number of engines that are starting to add new requirements.

[Alan Perkins]

Very good answer DanK. I want to continuously refine the White Paper to make it more understandable and relevant. Any suggestions you can make to improve it would be appreciated.

In one sense you are right that there is an entire industry out there, but in another sense you're wrong. It seems to me that, for us, the "industry" is "what gets tarred with the same brush when something goes wrong somewhere". Think about what characterises other, more established industries and our "industry" has none of it.

I find it amazing that, after six or seven years, there is not even an official industry "dictionary". You are right about "cloaking". The term was originally coined to cover agent-based delivery of unique content to search engine spiders. I call this "Agent-Based Spam", because I don't see a great difference between it and using another "carrier", such as ALT tags. I can examine the source of both documents quite easily.

"Cloaking" then moved on to covering both agent-based and IP delivery, but both of these technologies deliver useful content to humans. Again, it's easy to examine the source of a document delivered using IP delivery.

"Cloaking", true hiding, is IP Cloaking, where no human (except a representative of the cloaker or the SE) can view the source. Cloakers would doubtless see this as a good thing, because it means I can't "rip off" their "highly optimised code". The sad thing is, this very feature means that cloaked content is spam. I'd like to use cloaking to simply describe the technology carrier that is IP cloaking, otherwise its meaning remains confused.

[DanK]

Here's a different thought (not entirely unrelated to the previous posts).

You say it is spam if you'd only do it because of the search engines.

So what is it when the only reason you wouldn't do something is because of the engines?

I say it's spam if you refuse to design a site a certain way, only because the engines aren't able to handle the site properly. Websites don't exist because someone made a search engine, engines exist because we have websites. If we are not allowed to run our business to the best of our ability because an engine can not handle our content properly, is that our fault or the engines. Things change but the engines are always a step behind content.

[Alan Perkins]

DanK

Good thinking!

It's true that engines can't keep up with the latest techniques. It's also true that lots of other visitors to your site can't keep up with the latest techniques. You are free to ignore all of those potential visitors, if you like (although Disability Law may restrict you on this some time in the future).

If you choose to create a site that can't be indexed by search engines, fine. But don't then trick a search engine into delivering its customers to that site. Make the same content you make available to a search engine available to users of that search engine.

My approach, FWIW, is to make sites accessible to as many potential visitors as possible. That way, those sites are always accessible to search engines.

[DanK]

Originally posted by DanK
You say it is spam if you'd only do it because of the search engines.

So what is it when the only reason you wouldn't do something is because of the engines?

Don't forget to answer the point above, that was the important part of the previous post...

[Alan Perkins]

Could be lots of things, some of which are mutually exclusive. Here's three things it could be:

1) Any awareness of the importance of search engine marketing
2) Accrediting search engines with more importance than they warrant
3) Missing the point that, if you're not accessible to search engines, there's a lot of other potential visitors you are not accessible to.

[DanK]

Uh, could you re-write my question in your own words so I can get some vague clue what question it is you think you just answered?

[Alan Perkins]

I'll try...

"If I really want to do something a specific way, but engines wouldn't understand it if I did it that way, so instead I do it another way, what is it?"

To which any of the three answers I gave above could be correct:

1) You could understand that you would be better off not doing it that specific way because you would be closing off search engines as a marketing channel
2) You could be accrediting search engines with too much importance. Forget them and do what you want
3) You could be missing out on the fact that (for example) your whizzy Flash site wouldn't be accessible to a large number of Web users

[markymark]

DanK,

I've just caught up with this since my original posts, so forgive me if I go over old ground or have misinterpreted you.

It seems to me that you are arguing only semantics. It is true that search engines haven't caught up with the technologies employed in web development and design. Agreed. It is also true that sites that do not conform to the current limitations of search engines may be more relevant to searchers than some of those that do. I agree with this too.

Nevertheless, the solution you are advocating (cloaking) has only one intent - to deceive the search engines. There are other methodologies that will allow your or your clients' sites to be both search engine friendly and, for want of a better phrase, cutting edge.

The fact that there are limitations to what search engines can index does not mean cloaking is a valid solution. It isn't.

[Alan Perkins]

Well put, markymark

DanK, I said earlier

If you choose to create a site that can't be indexed by search engines, fine. But don't then trick a search engine into delivering its customers to that site. Make the same content you make available to a search engine available to users of that search engine.

i.e. why go to the effort of installing "cloaking" software just to deliver content to search engines. If that content is worthwhile (which it should be), deliver it to the users who don't have access to your "cutting edge" site.

[DanK]

Originally posted by Alan Perkins
I'll try...

"If I really want to do something a specific way, but engines wouldn't understand it if I did it that way, so instead I do it another way, what is it?"

To which any of the three answers I gave above could be correct:

1) You could understand that you would be better off not doing it that specific way because you would be closing off search engines as a marketing channel
2) You could be accrediting search engines with too much importance. Forget them and do what you want
3) You could be missing out on the fact that (for example) your whizzy Flash site wouldn't be accessible to a large number of Web users

So in essense, by choosing not to do something, or to do something else because of the way engines work, you are doing something specifically for the search engines, and no other reason. That sounds like you are spamming the engine to me.

So when you build a content page to increase your search engine ranking, regardless of whether the content is good or not, you did it to increase your rank, even though the content has value to other people you still did it primarily for rank, how you choose to increase your ranking in an engine is semantics, when in the end what you are trying to do is increase your ranking. I do not cloak to engines, and I don't advocate cloaking to engines. And naturally I don't advocate that cloaking is always spam.

[Alan Perkins]

DanK

What are you trying to prove? Your logic has lost me.

So in essense, by choosing not to do something, or to do something else because of the way engines work, you are doing something specifically for the search engines, and no other reason.No, you are not doing it specifically for the search engine. There should always be another reason. I've mentioned four possible reasons a couple of times in this thread - improving technical architecture, information architecture, accessibility and copywriting.

So when you build a content page to increase your search engine ranking, regardless of whether the content is good or not, you did it to increase your rank, even though the content has value to other people you still did it primarily for rank, No, DanK, I build content pages for people to read.

I do not cloak to engines, and I don't advocate cloaking to engines.Glad to hear it!

And naturally I don't advocate that cloaking is always spam.When isn't it? How can content be relevant to users if it can't be seen by users?

[Mel]

Hi Alan:

To restate my thoughts on this matter, I fail to see how anyone in this day and age cannot be aware of search engines and their profound commercial impact on websites. I trust that this is self evident and does not have to be defended or explained.

So when I write a web page I want it to include information which is useful and interesting to my desired readers, and I also want to be sure that these readers can find it and view it properly. Given that I (and fifty million other webmasters) am aware of the existance of search engines I will (if I want to stay in business) write it in such a manner that it is also useful to search engines, and I will also modify my natural writing style to sell the customer better. It is virtually impossible to isolate any one thing and say I did or did not do that because of search engines solely. There are too many parameters some of which may not even be conscious decisions.

An example that comes to mind is header tags - Frankly I doubt that I would use them on my web pages if search engines did not give extra credit for them in their algos. So by your definition I am a spammer, but someone who does the same identical thing and says "I do this because it is better writing style' is not a spammer? Especially given that it is impossible that he does not know that this is going to help him with the SEs.

Agent based delivery is a similiar thing - it can and is used soley to help more of your viewers see (or hear) your pages properly, and It is also used to trick the search engines. So what if it does both - is that spam? Or is it only partial spam?

Perhaps the basic problem with trying to find a one-size-fits-all definition of spam is that it is in effect trying to legislate morality, and we all know what a dismal record the human race has in that area.

Spam, like morality, is in the mind of the user, and thus I cannot see that there is a need for a one-size-fits-all definition, that will work in all cases and in all circumstances. If there is not a need why discuss it further?

[DanK]

Originally posted by Mel
Perhaps the basic problem with trying to find a one-size-fits-all definition of spam is that it is in effect trying to legislate morality, and we all know what a dismal record the human race has in that area.

I like the way you talk :)

[The Ney]

Originally posted by Mel
Spam, like morality, is in the mind of the user, and thus I cannot see that there is a need for a one-size-fits-all definition, that will work in all cases and in all circumstances. If there is not a need why discuss it further?

I agree with you on this one Mel. I have always claimed that deep deep in our hearts, we all know when we are doing something wrong. Even the spammers do. :)

I have followed this thread with the utmost interest. However i don't agree with some of the definitions given.

I think that any kind of technology used on the net is valid, taken that nobody is trying to decieve anyone. Take a window pop-up for example. 99% of the sites that use this technology, use it in order to push commercial messages, that the surfer didn't ask for. That is spamming. Does that mean that the pop-up window technology is wrong ? No!. There are sites that use this technology, as a part of their design. They use the pop-up window as a place to put the map of the site, or a legal agreement that the surfer has to agree to prior to entering the site etc.

I understand the claims that say that ALL of the cloacking is spamming, since they come from an experience according to which every time that people encountered clocking, there was a case of fraud. However, what about cloacking that is ment to protect the meta-tags ?

I know that someone has mentioned this before, but the issue was brushed off. Well, how do i protect the results of my work from being stolen ? I don't know any other way but cloacking. I can sue the thiefs (if my metatags are considered copyright, but how am i going to prove that?), but that is only in case i get on their tail, and even then it will take me a whole lot of precious time.

So i think that in this case cloacking by IP delivery is not spamming. I am delivering exactly the same content to both the user and the spider/editor. The only difference between the two is that the search engines IP will in, addition to the content, see the metatags, a lack of which will not hurt the regular surfer.

[ihelpyou]

I disagree with that completely. Why would you or anyone want to protect your meta tags? They mean nothing. What people who cloak in this way mean is that they want to hide what they are doing from visitors. That is what they truly mean. We all know the meta tags mean next to nothing in the search results. Not a good excuse at all for IP cloaking. What these people really mean is that they want to serve the spiders one thing and the browser another thing for the sole purpose to deceive and achieve higher ranks because they are incapable of optimizing and designing a website that is both pleasing to a browser and pleasing to a search engine spider......... translation? = Spam.

You or anyone can copy my meta tags all day long and all night if you wish. That will not mean anything at all to your ranks. Why cloak? You wish to "hide" and wish to deliver different content. Period.

[ihelpyou]

hey the Ney,...... please don't think I was referring to you in my post. I reread it and thought I might have worded it wrong. I do not know if you cloak or not. All I am saying is that is what all the IP cloakers use as an excuse.... that they are protecting their code from those who may steal it. I just think that is a lame excuse, when the real reason is that they want to achieve high rankings and IP cloaking is the only way they can do this.

[DanK]

Alans definition covers metas, and basically the rest of the HEAD section (the page says everything outside of the BODY but that pretty much leaves the HEAD). What it looked like to me was it was OK to cloak there, or just about anything else you want to do. Of course, if you just cloak there you can only protect a few things, meta keywords, and how many spidering engines use those, meta description, the only valuable meta tag used in regards to engines, but it's only valuable because it's displayed on most engines, so hiding it in your source won't do you much good except on google... which probably doesn't add any points for the description (probably ignores it completely), and speaking of google, you *could* hide the nocache google tag and only display it to them, that is a lot of technology just to save 50-100 bytes of transfer per page, but I can forsee a time when a lot of different engines all want their own meta tag info on your page. And the final thing you could protect in head, the title, now that is something that can have a big impact on your ranking... ooh, but those are also displayed publicly by engines so cloaking won't protect it.

Something no one has mentioned in regards to cloaking is dealing with rogue spiders, ones that won't behave, you'd call it spam to send those spiders something different than what you'd send a visitor, but is there a reasonable reason to not send them an error or an empty response in hopes that they'll go away and stop hammering your machine. Or if you have a malicious user on your site being able to block their access to a resource (this can be accomplished without any fancy cloaking software, but it doesn't take fancy cloaking software to cloak, not based on IP address, but this would in fact be cloaking, and in one example it's cloaking to an engine).

Next I guess we could argue about how email can only be used to spam, or maybe how a mass mailer can only be used to spam. You can look at the bad of anything, because it exists whether you see it or not. And even if you don't see the potential good in something, it doesn't mean it's not there. Believing the world was flat was popular for a while, and anyone opposed to that idea knew to keep it to themselves or become persecuted. I'm not denying cloaking can be used to spam, I'm just denying the close minded theory that it has only one purpose. Try this one on for size... a small round sharp wooden stick... with a lead center... a cruel stabbing weapon intended to distribute slow acting poison!!!! Or maybe just an every day tool for writing, thankfully the use of quills was widespread before the invention of a pencil, so it wasn't so hard for people to understand.

[Spider Man]

Why would you or anyone want to protect your meta tags? They mean nothing. What people who cloak in this way mean is that they want to hide what they are doing from visitors. That is what they truly mean. We all know the meta tags mean next to nothing in the search results.

Just my 2c but why would you not want to "cloak" your meta tags. As unimportant as they are, the keywords tag is essentially the point at which the search engine is asking what your page is about. Keeping in mind that different search engines like it in different ways. I have a site that does this with SSI for the keywords tag...

When it gets a visit from a spider it - Looks up what the main keywords are on the page. Connects to a search engine which tells it related popular search terms. Checks which search engine is visiting and dynamically writes appropriate keywords in that search engine's preferred format and feeds it that. This way I don't have to bother about keyword tags, I just write in my ssi tag on each page and let it decide (just for the record this is the only content dynamically generated specifically for the spiders on my pages)

Technically I am cloaking the keywords tag right? I'm doing it because 1) I'm too lazy to write a keywords tag on every flipping page! 2) I'm too lazy to write a keywords tag on every flipping page! :-) However, side effect of my lazyness is I NEVER have a spammy keywords tag. It always has appropriate and relevant keywords (as determined by a search engine) and because those that do use the keywords tag have different preferences (opinions of what is spam) it customises to never submit a spammy keyword tag. I'd even go so far as to say it's less spammy than peope who use normal static keyword tags, and a whole lot less work.

With all this talk about meta tags I Just thought i'd help blur the lines a bit! Obviously I agree that title tags and meta description cloaking is looneyville.

[ihelpyou]

Well Chris, that is fine as I am under the opinion that the keyword tag is not used anymore by any engine so what you are doing would be just fine. I personally do not think that even if the tag is used by a particular engine, the way it is written does not mean much.

Where I have a big problem is when the cloakers go a little further and then claim they are hiding all their code from the visitors.

[Alan Perkins]

Mel said
An example that comes to mind is header tags - Frankly I doubt that I would use them on my web pages if search engines did not give extra credit for them in their algos. So by your definition I am a spammerMel, I am not calling you or anybody a spammer. I'm trying to guide you into the kinds of spam tests a search engine spamfighter would apply if they looked at your pages. If they see header tags wrapped around headings, they will have no problem with that. If they see header tags wrapped around the whole body content, they will have a problem with that. Nobody can tell just by looking that you chose to use header tags because search engines like them rather than because they made the structure of your document clearer. You are doing the right thing for the wrong reasons, IMO.

Why do you think search engines give more weight to text in header tags? Here's my answer : because the algo writers think that good Web designers will use header tags to structure and label the content of the document more clearly. The header tags should precis the document. Therefore, their content is more important than body text.

Mel said
Spam, like morality, is in the mind of the user, and thus I cannot see that there is a need for a one-size-fits-all definition, that will work in all cases and in all circumstances. If there is not a need why discuss it further?
What if the "user" in question is a spamfighter working for a search engine. Wouldn't you like an insight into their mind, to allow you to avoid spam penalties for things you had no idea were spam? We discuss itfurther because we wish to avoid spamming, not to discover after the fact that we have spammed.

The Ney said
I think that any kind of technology used on the net is valid, taken that nobody is trying to decieve anyone
By "anyone", do you mean search engine spiders? Because if you use IP cloaking technology to deliver unique content to search engine spiders (and there is no other credible use for it), you are deceiving that search engine, and, by implication, all the users that rely on its results. A search engine expects to base its ranking calculations on what its users will actually see.

The claim that all IP cloaking is spamming does not have to be made based on experience. It is spamming because the cloaked body content will never be made available to the search engine's visitors.

DanK described why using IP cloaking to protect your meta tags is a waste of time and effort.

He also said...

DanK said
Something no one has mentioned in regards to cloaking is dealing with rogue spiders

That's not cloaking, it's IP delivery. We have already discussed the difference earlier in this thread : "Cloaking", true hiding, is IP Cloaking, where no human (except a representative of the cloaker or the SE) can view the source. Everything other technology, including agent-based delivery, is not IP Cloaking and is therefore not necessarily spam.

[Alan Perkins]

Chris

It's only cloaking if it is IP cloaking. I assume you are using agent-based delivery to do what you describe. That is not cloaking. Using agent-based delivery to deliver unique content in the <HEAD> section of a document is a good thing to do IMO(*), as long as:

1) If the content is meta-data, it is accurate
2) If the content is designed to be read by humans (e.g. title, description), it is coherent

Your example passes both those tests.

(*) I hear some search engines disagree with this. Agent-based delivery is a whole can of worms that I think those SEs would rather keep the lid on. But that's tough because agent-based delivery is an entirely appropriate technology for people to use, invented before even the first Internet search engine. It is up to search engines to adapt to agent-based delivery, not the other way round.

[Spider Man]

It's only cloaking if it is IP cloaking. I assume you are using agent-based delivery to do what you describe. That is not cloaking.

Bad assumption. I use IP cloaking just because I felt like using that, I could use either for the same effect. Surely when we define cloaking it is the "effect" we're looking at and not what's being used to implement it?


Now, for the body text question! Assume I want to use a combination of these three methods of visitor detection..

1) IP
2) User Agent
3) Crawling pattern match

to hide my email address from automatic extractors (getting fed up with all those emails!). I'd also need to hide them from caching search engines (coz email spammers could have a field day pointing their extractors at something like google's cached pages). If I combine the three methods to feed email extracting spiders an email address at the domain they came from and feed search engines an address like nospam@please.com, I am essentially changing body text. Now I'm pretty sure this isn't spam, but i'm also pretty sure it is a form of cloaking?


Oh and just for fun, how about image cloaking? Am I allowed to feed Google's image search thing images that say "Quit Stealing my Images" whilst sending my normal images to every other viewer (or do you think they might prefer a picture of Larry Page or the google logo or something)?


Sorry, just pushing the boundaries of your definition :D

[Alan Perkins]

Sorry, just pushing the boundaries of your definitionTesting, not pushing...

I use IP cloaking just because I felt like using that,IP Cloaking is delivering unique content to search engines. That unique content is spam.

You either want a search engine or other robot to crawl and index your site or you don't. If you don't, use the robots standards (robots.txt, robots meta tag) to prevent it crawling and indexing you. If it doesn't obey either of those standards, use any or all means available to you to stop it.

If you do want a search engine to index your site, don't trick it into delivering its visitors to your site. A search engine expects to base its ranking calculations on what its users will actually see. If you don't fulfil your part of the "contract" by delivering content that the search engine's users will see, then don't be surprised to be labelled as a spammer and excluded from the index.

The concept of "removed" content is, mmm, interesting... [retires to think some more]

[Spider Man]

Yes, testing!!!


IP Cloaking is delivering unique content to search engines. That unique content is spam.


Now here's where I'm getting lost. From your answers to my keyword example I presume you're saying it's only ip cloaking if the unique content is spam? Because my keyword example is unique content but isn't really spam. Or is it the quantity of unique content? If the former is the case - what's delivering different, non spam, content specifically designed for search engines?


A search engine expects to base its ranking calculations on what its users will actually see.


I feel a devil's advocate mode coming on! A search engine doesn't "expect" (unless it's a pretty naive search engine). It may well wish, what they expect is that in all probability the on the page content that it is reading has been manipulated or influenced in some way. That's why we have the explosion of search engines that rely heavily on off the page factors. 'course this makes spamming with cloaking less and less viable.


If you don't fulfil your part of the "contract" by delivering content that the search engine's users will see, then don't be surprised to be labelled as a spammer and excluded from the index.


I actually sit on the fence on this one. I don't much care if people cloak or not, I think its significance will decline shortly. I do find it hard to understand there's any suggestion of a "contract". Maybe I've sat through too many boring law lectures :) Or to put it another way, the search engines make advertising revenue through a medium that involves borrowing small pieces of your content. Not only did you not choose to give them this content but you have to opt out (with the robots.txt) if they'll let you. They use up the web site's bandwidth and the vast majority of sites will receive no or negligible increase in targetted visitors. I'm pretty happy with the deal but I can see why the guy on page 263 of the results might not agree with you if you use "contract" in your terminology!!

[DanK]

It's OK to cloak to a spider, or use any other technique, if you decide a spider is 'rogue', now maybe I wasn't clear about it but rogue just means to me misbehaved, a rogue spider might be run by an engine (several are). If it's OK in this one case, doesn't that mean there is a legitimate use for IP cloaking to an engine and that it's not always such a bad thing.

There is a reason to cloak meta keywords, maybe not because engines value them, but because they can be an indication of what keywords you are targetting, I'd consider it paranoya to worry about it that much but in a highly competitive market it could be benificial to have keywords no one else has thought of. There is also a down side to using cloaking to hide meta tag information... you are essentially keeping an unknown engine from being able to index your page in a reasonable way (yes most engines ignore some or all meta data but not all engines do, and not all engines are known to everyone as new ones pop up every day, you never know which one will take off). There is another reason to cloak your meta tags, such as if you are trying to rank under trademarked names like google yahoo altavista msn excite looksmart hotbot Google Yahoo AltaVista google yahoo.com, someone could hide this (from users) by using cloaking, and since this is illegal it would be abuse of a technology. The site I got those meta tags off of also seemed to be placing the word search engine in the keyword list several times. I don't know if it's spamming but it is surely unethical, as many of us here know it's illegal to use trademarked names in that way.

[Mel]

ALAN SAID
What if the "user" in question is a spamfighter working for a search engine. Wouldn't you like an insight into their mind, to allow you to avoid spam penalties for things you had no idea were spam? We discuss it further because we wish to avoid spamming, not to discover after the fact that we have spammed

I am talking about a one-size-fits-all definition which to my mind is not a likely outcome when search engines are concerned, and at any rate, based on their past practices, they are not going to willingly divulge to us their preferences or rules.


We already know, more or less, through experience and discussion, what each search engine considers spam and while there are some similarities, there are perhaps also as many differences.

As a matter of interest, if you examine laws passed by humans to regulate their citizens behavior, these are invariably complex and verbose, and require a system of courts and judges to clearly define who is guilty of infractions, and who is not. This is usually done in an attempt to cover the many eventualities that any situation can generate. This does not augur well for a one sentece definition of spam which would cover all eventualities.

Perhaps there is a possiblilty that the definion of spam could be an exception to this rule, but I fear not, given the many different scenarios that have been brought up here in this fourm already and the many different opinions.

In the unlikley event that we did get all the search engines to agree to and publish a common definition of spam, this would then create a market niche for someone to jump into saying we have different rules which makes our search easier and better.

I fully agree with the principle that we should all be "ethical" in our dealings with search engines, but each search engine makes its own rules, and each webmaster makes his own decision as to how he uses the technology available to him. This is a constantly changing scenario, and will continue to change as new technology and opportunities become available.

By all means discuss what is etheical and what is not, but don't try to fit everyone into a pair of size socks that the manufacturer says will fit everyone. As the owner of a pair of size 11 feet, I can testify that the results can be painful at best.

[Alan Perkins]

Great thread!


From your answers to my keyword example I presume you're saying it's only ip cloaking if the unique content is spam?You wouldn't go to the effort of IP cloaking to deliver non-unique content. People IP cloak to deliver content to search engines that will never be seen by the users of that search engine. By definition, that content is spam. It doesn't matter whether or not that content is "relevant" to the search term. Who's to say whether it's relevant or not anyway...that's what this whole thread is about. It's only relevant to users if it can be seen by users. Otherwise, it's spam.


It may well wish, what they expect is that in all probability the on the page content that it is reading has been manipulated or influenced in some way. That's why we have the explosion of search engines that rely heavily on off the page factors. 'course this makes spamming with cloaking less and less viable.This is my problem with spamming and cloaking. It is ultimately self-defeating, not only for its practitioners, but for everybody. Witness the move to the ultimate off-the-page method - paid placement. It's not really surprising when you think about it - as a search engine, why put effort into developing an easily tricked algorithm, when you can make cash out of a difficult-to-trick algorithm?


Not only did you not choose to give them this content but you have to opt out (with the robots.txt) if they'll let youNow this is really interesting. We're getting into the ethics of search engines themselves. In principle I agree that the most "ethical" method would be opt-out-by-default. The problem is that search engines, and therefore the Web, wouldn't be nearly as useful as they are if this were the case. The opt-in-by-default method is benefical to search engines and Web users.

To play devil's advocate myself, how unethical is opt-in-by-default? As a Web site owner, you have chosen to publish a Web site and make it publicly available. Do you have a right to expect that robots will not access your Web site? Where does it say that? What is a robot, anyway? Is Internet Explorer a robot? What if another Web site owner created a frameset that pulled in every page from your site into a separate frame, causing Explorer to generate hundreds of quick-fire hits? What if a browser pre-cached every link on the current page to reduce time-to-load? What if somebody employed a thousand people to do the task of a robot, to avoid having to comply with robot "laws"?

All I'm interested in is search engines having the opportunity to get better and better. I don't see how spamming and cloaking can help this.

DanK
It's OK to cloak to a spider, or use any other technique, if you decide a spider is 'rogue', now maybe I wasn't clear about it but rogue just means to me misbehaved, a rogue spider might be run by an engine (several are). If it's OK in this one case, doesn't that mean there is a legitimate use for IP cloaking to an engine and that it's not always such a bad thing.
IP cloaking is identifying a spider by IP name or address and delivering unique content to that spider. I don't see why, just because a spider is "rogue", you need to deliver unique content to it. Content that will never be seen by a search engine's users is not relevant to those users.

If, in dealing with a rogue spider, your actions prevent you from being crawled and indexed by that rogue spider, I don't see how anybody could label that as spam.

DanK
There is a reason to cloak meta keywords, maybe not because engines value them, but because they can be an indication of what keywords you are targetting
The keywords in the keywords tag should also be on the page! So it should be pretty obvious what keywords you are targetting without looking at the keywords tag. A search engine would not give much weight to a word in the keywords tag if it didn't also appear on the page.

Mel
If you examine laws passed by humans ... don't try to fit everyone into a pair of size socks that the manufacturer says will fit everyone
I don't think laws or socks are an appropriate metaphor. Let's not go there, though. I'm interested in how you do SEO. How do you know whether something you are considering doing is spam or not?

[Mel]

Hi Alan:

I beg to differ with the idea that there is no similiarity between laws and regulations or rules, In fact I can't think of a better metaphor. What do you classify your definition of spam as if not a rule or proposed law of SEO?

Regarding Socks, I think that everyone can understand the metaphor of what happens when you try to fit a large subject into one sentence, and my point is that I don't think you can find a one-size-fits-all definition, which is perhaps why there have been so many what ifs brought up in this thread.


I'm interested in how you do SEO. How do you know whether something you are considering doing is spam or not?

Well I don't have a rule book by my elbow - My first action is to determine what the most relevant keywords are and then begin to understand the customers business enough that I can write a series of web pages using my keywords and phrases, that are interesting to the viewer, and that sell his product to the viewer. I then set it aside for not less than two days, reread the pages, polish up the writing and make sure that the relevant tags and titles are in place, and submit it to a couple of directories. Depending on the project I may then spend some time getting links from relevant pages.

Once the SERPS begin to come in I review them against the higher ranking pages and tweak the pages based on what is done to achieve the rankings in the higher ranked pages.

So far I have had no need to cloak, use invisible text or multiple titles, and it seems to work.

[Mel]

Deleted as a multiple submission

[Alan Perkins]

I think the closest metaphor would be the regulations of an industry body. There is no such thing in our industry, but if there was, don't you think the industry body would like the opportunity to examine the work of its members? Cloaking prevents that.

If laws is the analogy you want to use, then I am not making laws. I'm just trying to interpret them.

My definitions of spam etc. are an attempt to interpret and distill the various rules of various search engines and place them into a context to allow people to understand why a particular technique is labelled as spam.

I know that search engines don't list all of the "rules" on their Webmaster help pages. I think they've made a mistake listing any "rules" at all, because there are an infinite number of rules. The principles are much simpler. That's what I'm trying to put across.

You didn't really answer my question: how do you know whether something you are considering doing is spam or not?

[Spider Man]

People IP cloak to deliver content to search engines that will never be seen by the users of that search engine. By definition, that content is spam. It doesn't matter whether or not that content is "relevant" to the search term. Who's to say whether it's relevant or not anyway...that's what this whole thread is about. It's only relevant to users if it can be seen by users. Otherwise, it's spam.


Alan, you should know my thought processes well enough by now to know that i won't just take "By definition, that content is spam" as much of an answer. The person who can best decide whether it's relevant is the search engine themselves. Using my keyword example, if they have had a large part in the deciding of the keywords and it has been formatted as they like it - how can this be spam? Yet it uses the IP address to feed unique content the the search engine - so it is IP cloaking. To be perfectly honest, the main reason for me playing devil's advocate here is that I've read quite a few threads elsewhere that have a similar theme. The two arguments basically go:

1) Feeding any cloaked content to the search engines is spam, because we say so.
2) Feeding cloaked content to the search engines is not spam, because we say so.

However verbose these arguments are put, the thinking person will have greatly difficulty accepting either viewpoint. Now I'm sure, Alan, that you're not just another viewpoint 1 pusher and that there's a reason that the answers to my examples have always involved 1) declaring using of cloaking technologies in a particular way as not cloaking or 2) using an unsubstantiatable definition such as "all cloaking is spam" 3) ignoring it. But I have to say I'm as unconvinced as any other thread/viewpoint I've read and still feel comfortable believing in the very obvious grey area.

This is my problem with spamming and cloaking. It is ultimately self-defeating, not only for its practitioners, but for everybody. Witness the move to the ultimate off-the-page method - paid placement. It's not really surprising when you think about it - as a search engine, why put effort into developing an easily tricked algorithm, when you can make cash out of a difficult-to-trick algorithm?


Doh. Alan, I presume that's "IP cloaking"? And if so why is it "spamming and cloaking" and not just "spamming", it being the "definition" that all ip cloaking is spamming. Engines have several zillion different criteria they can rank your page on, the fact that they've only just woken up to which ones are less susceptible to webmaster manipulation is what's surprising. One method of spamming isn't going to suddenly kill a whole type of search engine, just ultimately the ability of that method of spamming. Search engines are in the business of developing accurate and DIFFICULT to trick algorithms, that's what they do, that's what maximises relevancy and brings in the visitors and advertising $. Is PPC really difficult to trick? What's all this fraud and ppc stuff i've been reading about then?

To play devil's advocate myself, how unethical is opt-in-by-default? As a Web site owner, you have chosen to publish a Web site and make it publicly available. Do you have a right to expect that robots will not access your Web site? Where does it say that?

Look, devil's advocate is my job. But as you asked.

Yes. I have chosen to make it publicly available (notice the term "publicly - to or be known by the [I]people as a whole or the people in general). Mr devil's advocate, I never asked any computer program to come along and provide a little section of my content to operator's web sites visitors in order for them to make a profit from it. It was my intention in placing the web pages on line that actual people should read them and that it should be useful to real life human beings. This is the general principle of the www and one that Google clearly agrees with through their declining of any use of automated tools on their servers. In fact, Mr Devils Advocate, many of my pages were placed on line before these robots existed. There is no formal contract but our implied agreement is that I will serve whatever content i wish to real humans and they will use it within the confines of human legislation. Now I appreciate that the robots may wish to read the content, but if it was not my intention to provide that content to robots then there is no implied agreement. I therefore have a very solid right to expect that robots do not access my pages without my permission. That's a strong case, but should I also put a statement on my site similar to Google's then I have a cast-iron statement that they are not allowed. I see little reason to argue the semantics of what is and is not a robot, as it's general definition is very well understood. The case for the prosecution rests :D

I couldn't resist answering your devils advocate bit!!! But Alan, all I was really saying with this is that the search engines themselves are on sketchy ethical grounds and they are providing very little benefit to many webmasters. Can I really stand in judgement of those webmasters if they in turn push the ethical boundaries? Should i really just side with the Search Engine rogue because it happens to help me?

I'm stopping now coz this post is really long and my fingers are tired :D

[Mel]

Originally posted by Alan
I'm interested in how you do SEO. How do you know whether something you are considering doing is spam or not?

I Have answered your question as to how I do SEO.

How do I know if something I do is spam or not? I DON'T know (no one does) - only the search engines know, the useage varies among search engines, and they are not giving us definitive answers in all areas.

IMO SEO is a dynamic thing, constantly changing as new technologies and new attempts by the search engines for better relevance make new demands on the practitioners of this arcane art. Experience, common sense, and discussions among fellow practioners are how we arrive at conclusions as to what works and what does not.

This is the bottom line - if it works, we should include it in our daily use as we need to and want to; If it doesn't work I for one will not bother with it. The search engines will change thier rules as and when they see fit, and we will react to those changes as necessary.

What was a few years ago best technology SEO is now considered spam by the search engines, and I fully expect to be able to make a similar statement a few years from now regarding todays "ethical" SEO.

[MakeMeTop]

>....delivering unique content to search engines. That unique content is spam.

Having just returned from Berlin - I think the whole cloaking thing has now been turned on its head having listened to 3 search engines detailing their 'trusted feed' programs. In this you submit a few lines of code (purely for the search engine) which is basically an outline of what you want the page indexed for (regardless of actual HTML content on that page) and provided it is deemed relevant to the SE and searchers it will be indexed accordingly on the data given to the search engine.

Isn't this cloaking? Or am I being dumb :)

[Mel]

What was a few years ago best technology SEO is now considered spam by the search engines, and I fully expect to be able to make a similar statement a few years from now regarding todays "ethical" SEO.


Having just returned from Berlin - I think the whole cloaking thing has now been turned on its head having listened to 3 search engines detailing their 'trusted feed' programs. In this you submit a few lines of code (purely for the search engine) which is basically an outline of what you want the page indexed for (regardless of actual HTML content on that page) and provided it is deemed relevant to the SE and searchers it will be indexed accordingly on the data given to the search engine

This is just what I meant, and I didn't have to wait a few years to see the beginnings of the changes.

[Alan Perkins]

Only time for a quick reply at present, but here's something to think about:

IP Cloaking (which is what I consider to be "cloaking") is identifying a search engine spider by IP name or address and, without permission from the search engine, delivering unique content to its spider. When talking about HTML documents, "content" is to be found in the <BODY> of that document. However, I find it difficult to believe that somebody would go to the effort of IP cloaking, just to deliver unique <HEAD> content. I also think, in future, it will be professionally and maybe legally important to allow peer review of work.

Trusted feed data is like an expanded form of the meta-data found in the <HEAD> section of a document. That data is not open to "Content Spam", but is open to "Meta Spam".

This is all covered in more detail in the White Paper on Spam Classification (http://www.ebrandmanagement.com/whitepapers/spam-classification/).

What do you mean by "spam"? Two key questions that have arisen out of the debates around here:

1) Why do some search engines apparently consider all "invisible text" to be spam?
2) If you are optimising a Web page, how do you decide whether your optimisation techniques are spam or not?

[MakeMeTop]

>...without permission from the search engine...

Ah, there is the rub (I think). Let us say that I ask a search engine to do IP delivery of a flat HTML alternative to a flash site and the search engine informs me that (as an inclusion partner) it is acceptable - but the pages will be checked at random to ensure that relevance is maintained and (in the SEs sole determination) may be removed if they do not meet their standards - how is this spamming?

I've asked this very question and got a remarkably similar reply.

I agree that in a non-commercial relationship with an SE this is unlikely to happen - but where there is a commercial relationship between an SEO and a search engine it certainly may happen.

Or is that the difference? Cloaking without paying the SE is spam - but who pays the piper etc., etc. (to a certain extent).

Anyway, I'll post more about the various happennings at Berlin tomorrow in the various fora. It was the best one-to-one meeting I've had to date with leading SE representatives (shame no Google or Yahoo) and we had long and very interesting discussions about spam. Much horror and red faces at Ink when I used the example put forward in http://www.ihelpyouservices.com/forums/showthread.php?s=&threadid=433&pagenumber=3 in front of the assembled people :)

[ihelpyou]

hey Barry, good to hear you are having some good meetings! A big problem with "paid" inclusion is that the page is submitted and then the "bait and switch" is done. Would be hard for Ink or PT to police that. I don't know what an answer would be other than to keep checking pages, etc, OR, make sure they listen to people when they report a page as "Spam". They should physically go to the reported page, and if spam, remove the page. Period. This "bait and switch" thing has went of for years at Yahoo as well.

[Kal]

Originally posted by MakeMeTop
Let us say that I ask a search engine to do IP delivery of a flat HTML alternative to a flash site and the search engine informs me that (as an inclusion partner) it is acceptable - but the pages will be checked at random to ensure that relevance is maintained and (in the SEs sole determination) may be removed if they do not meet their standards - how is this spamming?


I'm with Barry on this one. I personally HATE Flash, but I can understand the need for it's use in some cases where eye candy is called for (thinking web design co's and graphic artists here). To me, this "approved IP delivery" should not be classed as spam, because it is accepted by search engines in advance. It also foils the "bait and switch" idea because of random content checks.

By the way - what a great thread!

[Alan Perkins]

OK, a lot to catch up on here...

Alan said
The concept of "removed" content is, mmm, interesting... [retires to think some more]
I've considered it more. I don't think it's acceptable. As I said, the Web is a public place and if you don't want people to know your e-mail address, don't publish it. However, I think it highlights the requirement for more tags/attributes (are you listening, W3C/search engines?) It would be useful to have <noindex> ... </noindex> and <nofollow> ... <nofollow> tags to allow parts of the content to be excluded, whilst still showing the full content to allow the SE to perform its ranking calculations. And to avoid having to wrap anchors with a noindex and/or nofollow, it would also be useful to have NOINDEX, NOFOLLOW attributes on all tags that included HREFs.

he said
why is it "spamming and cloaking" and not just "spamming", it being the "definition" that all ip cloaking is spamming
For emphasis. However: IP Cloaking is always spamming. Spamming isn't always IP Cloaking.

he said
search engines themselves are on sketchy ethical grounds and they are providing very little benefit to many webmasters. Can I really stand in judgement of those webmasters if they in turn push the ethical boundaries?You don't have to stand in judgement, but search enigines themselves will. The way search engines see it, those Webmasters have three choices: play fair, play unfair or don't play. If they choose "play unfair", they may suffer consequences such as demotion or banning.

Barry said
>...without permission from the search engine...

Ah, there is the rub (I think).It is partly the rub, yes. IP cloaking is the only technique where, irrespective of content actually delivered, that content must be spam. As soon as you have (written) permission, it ceases to be IP Cloaking and becomes IP Delivery. Then, of course, you might or might not deliver spam...

Barry said
Let us say that I ask a search engine to do IP delivery of a flat HTML alternative to a flash site and the search engine informs me that (as an inclusion partner) it is acceptable - but the pages will be checked at random to ensure that relevance is maintained and (in the SEs sole determination) may be removed if they do not meet their standards - how is this spamming?
The technique is not spamming if you have written permission. However, you may still deliver spam. Just because the random check doesn't show it up, that does not mean you are not spamming.

Barry said
Or is that the difference? Cloaking without paying the SE is spam - but who pays the piper etc., etc. (to a certain extent). I don't think paid inclusion necessarily means cloaking is OK. It still depends on the SE's algorithm - if, with paid inclusion, a SE ceases to index the <BODY> content, then cloaking becomes redundant. I've been trying to talk generally but that's proving obstructive, so let me get specific for this post.

When cloaking, you would purposely deliver unique <BODY> content to a search engine spider, that would never be seen by that search engine's users, in order to rank higher for certain keywords. Any search engine that bases its algorithms partly or fully on what is in the <BODY> will have a problem with this (I can go into detail on why if you like...)

Kal said
I personally HATE Flash, but I can understand the need for it's use in some cases where eye candy is called for (thinking web design co's and graphic artists here).There is no need to maintain a database of SE IP names and addresses just to deliver Flash. Simply look at the Accept-Types in the client request header and, if it supports Flash, deliver Flash. If the client doesn't support Flash, deliver something else. Show a search engine what you will show its users who don't have Flash. Don't create special content just for a search engine to see...otherwise you are spamming.

[The Ney]

after all being said, i still dont agree with calling the IP cloaking technology SPAM, only because it is being used for spamming. It is similar to blaming the pencil for spelling mistakes.

And Doug, i dont cloak. It is too complicated and too time-demandindg as compared to the actual benefit it brings.

However, Chris's idea with SSI is quite atractive... hmmmm :thebomb:

[Alan Perkins]

Hi The Ney

It's easy to confuse terms. A long time ago in this thread, I think I said "IP Cloaking is a carrier guaranteed to be delivering spam". i.e. IP cloaking technology is not spam, but what it delivers is.

Why don't you have a go at these questions:

1) Why do some search engines apparently consider all "invisible text" to be spam?
2) If you are optimising a Web page, how do you decide whether your optimisation techniques are spam or not?

[The Ney]

Originally posted by Alan Perkins

Why don't you have a go at these questions:

1) Why do some search engines apparently consider all "invisible text" to be spam?
2) If you are optimising a Web page, how do you decide whether your optimisation techniques are spam or not?


1. the question here is not why do some search engines consider something a spam. Search engines will cave in to the trends set by the public (including us, the SEO ppl). The question is whether the technology can be used solely for spamming. If the answer is yes, then the technology itself is spammy.
Now for the invisible text, using the invisible text has no other purpose than to serve the spiders with data that the users "cannot" see, IN ORDER TO TWEAK THE RANKINGS. Please note that this doesnt define the invisible text as a spammy technology (if it can be called technology at all), what is spammy is the loosing combination of the invisible text and BAD INTENTIONS.

2. If i am doing something that decieves the surfers/search engines and its sole purpose is to improve my rankings, then it is spamming.

ranking is supposed to be performed according to relevance. Every search engine measures the relevance in more succesful or less succesful way. If i am performing decieving steps in order to improve my rankings, without improving my relevance, that is spamming. Cloaking for the sake of hiding the meta tags (and not for the sake of serving the spiders one CONTENT, and surfers something else) is not spamming (IMHO). Even if search engines think otherwise.

[Alan Perkins]

Good, The Ney, a lot of common ground there.

Like you, I don't believe all invisible text is spam. It needs to be examined on a case-by-case basis. In particular, if the invisible text is coherent (i.e. would make sense to a human reading it if it weren't invisible) then it is not likely to be spam.

"If i am doing something that deceives the surfers/search engines and its sole purpose is to improve my rankings, then it is spamming. " - sounds good. :)

So, let's look at what cloaking is. I have defined cloaking as identification of a search engine spider by IP name or address and the delivery of unique content to that spider. By content, theoretically I mean any content, in practise I mean <BODY> content. The use of IP cloaking simply to hide your meta tags does not stand up. IP cloaking is difficult to do properly. In return, the <TITLE> and <META> description tags are displayed on most SERPs so they are not "hidden". That leaves the <META> keywords tag.

The <META> keywords tag is hardly used by any SE. Even if it was important, if the keywords it contains don't also appear in the <BODY> text (which, supposedly, isn't cloaked), they won't particularly help your ranking. So, as he implied, your <META> keywords tag can simply be interpolated from your <BODY> content. In other words, nothing in the <HEAD> section is worth hiding with IP cloaking. If you want to avoid net congestion (very admirable), use agent-based delivery, not IP cloaking.

To be worthwhile, IP cloaking has to hide <BODY> content, the <BODY> content it is hiding will never be seen by users of the search engines and therefore, by definition, it is spam.

[Advisor]

hide <BODY> content, the <BODY> content it is hiding will never be seen by users of the search engines and therefore, by definition, it is spam. What if it can be seen in the Google cache, such as is the case with the use of noframes text. Do you consider that spam?

Jill

[ihelpyou]

No, cause text in a noframes tag has another use. It is there for visitors who have a browser that does not support frames. Hence, not spam.

If though, a noframes tag is put in place without a frame, then it would be spam IMO.

[Spider Man]

I've considered it more. I don't think it's acceptable. As I said, the Web is a public place and if you don't want people to know your e-mail address, don't publish it. However, I think it highlights the requirement for more tags/attributes (are you listening, W3C/search engines?) It would be useful to have <noindex> ... </noindex> and <nofollow> ... <nofollow> tags to allow parts of the content to be excluded, whilst still showing the full content to allow the SE to perform its ranking calculation

I do want PEOPLE to know, i don't want programs to know. I'm glad to hear that my email address now counts in my web pages ranking calculation!!! Tell me, does mine give me a boost or a penalty? :)

You don't have to stand in judgement, but search enigines themselves will. The way search engines see it, those Webmasters have three choices: play fair, play unfair or don't play. If they choose "play unfair", they may suffer consequences such as demotion or banning.

Alan, I AM the search engines - all be it a group of very small ones. And with regards to mine I don't care which of the three options people take. Call me harsh but if a page looks right in the results then great but if it looks wrong then I ban the whole domain. How a webmaster gets that rank really has nothing to do with it. It's whether the end result fits that matters. Despite the public face, I really don't see that the big search engines are any different underneath.

I'm gonna bow out of this thread because I was really only here to play devil's advocate and because it seems to be going round in circles. But I can't resist a final comment...in that strange real world place there are those that declare the internet as the root of all evil. Because there are some bad things about it they declare the technology (the internet) as bad. Talk to those people and they'll declare there's nothing good about it, in fact..they've probably even done a whole lot of research focusing on the bad points. Point out the good bits to them like email, and suddenly email is defined as not being part of the internet - the internet is of course just the www! Thankfully, us "normal" people didn't let them suffocate a technology that proves to be very useful, we protected the technology not for the bad bits but for the good bits. This thread reminds me very much of them.

[Alan Perkins]

Jill, I'm not sure I understand the question - but I'll try to answer anyway. :) The issue is whether the content is designed to be seen by humans. If it is designed to be seen by humans, and is seen by humans, no problem. If it is not designed to be seen by humans, and is seen by humans, that is a problem. Cloaked content is not designed to be seen by humans, so what purpose does it serve?

sorry you want to bow out ... but I need to reply, anyway:

said
I'm glad to hear that my email address now counts in my web pages ranking calculation

Well, of course it does, if what you are looking for is pages that contain your e-mail address :)

said
Alan, I AM the search engines ... Call me harsh but if a page looks right in the results then great but if it looks wrong then I ban the whole domain


You are harsh. I'm very surprised by your comments. According to what you are saying, all forms of spam should be allowed, as long as the results are "relevant" - in whose opinion, and on what basis?

What makes the results delivered by Webmasters who have cracked your algorithm and deliver content tuned to that algorithm the most "relevant" results? Why not just stop improving your algorithm, publish it, then monitor the SERPs for "relevancy" - the PPC model springs to mind.

If you are going to base an algorithm on the contents of the <BODY> section, it has to be the <BODY> content that people will actually see - or what's the point? Most search engines don't have spam as a problem until they are popular enough to warrant spamming.

said
We protected the technology not for the bad bits but for the good bits

That's just what I'm trying to do by not labelling things like IP delivery and agent-based delivery as cloaking. I have been very careful to define cloaking as a technology that is only useful to deceive search engines. I have been very careful to separate it from IP Delivery, Agent-Based Delivery, and other forms of "personalised delivery", all of which are useful technologies IMO. I want search engines to understand that there is a place for these other technologies, but not for cloaking, which is there only to deceive them.

[Alan Perkins]

I'd just like to say what a great thread this was. Thanks for starting it DanK. There are a few loose ends and conclusions:

The original question, "Only Cloaking is Spam?". The answer is "No". There are other ways to deliver spam. So, spamming does not always mean cloaking.

However, cloaking always means spamming (i.e. cloaking is a subset of spamming). Why? Because otherwise it's not cloaking! IP Delivery, Agent-Based Delivery, etc. are not cloaking. Cloaking is a word that is used a lot but defined nowhere. I believe it's important to separate the meanings of cloaking, IP delivery, agent-based delivery, personalised delivery, etc. to show what is (or should be) acceptable and what isn't.

These are not my rules. They are a logical interpretation of search engines' rules.

Some search engines's spam detection algorithms may label things like "agent-based delivery" as spam. This means if you use such technology fairly, you stand a good chance of being unfairly penalised. Be warned.

Other search engines do send out rogue spiders from time to time. Handle them with any or all means at your disposal. But if you handle them by spamming, you may be penalised.

Some of my efforts are aimed at convincing search engines what should be acceptable, some are aimed at convincing Webmasters. But in the end, it's up to both to make their own choices. I believe more informed choices are better. I don't think many Webmasters fully understand the reason that search engines label things as spam, and I don't think many search engines fully understand Webmasters either. Mutual respect and understanding is required, IMO.

The "contract" is an unwritten one between Webmaster and search engine. The search engine can break the contract at any time without reason. Give it a reason and it is more likely to break the contract. If search engines are an important source of traffic for you, and you want them to remain like that, be very careful about spamming. And know what they mean by spam. It isn't just what they say on their Webmaster help pages. Spam is any form of deception.

The concept of removing content is unacceptable for the following reason: the coherence of the remaining content is reduced, and the keyword density of the remaining content is increased. Search engines that base their algorithms at least partly on the <BODY> content expect the content they are basing their algorithms on to be seen by at least some of their users (preferably all). This is the key difference between cloaking, IP delivery, agent-based delivery, etc. When cloaking, you have no intention of delivering the <BODY> content that you deliver to the search engine to any of the search engine's users. When using another "personalised" technology, at least some of the search engine's users should see what the search engine saw.

[ihelpyou]

I feel that this was a great thread also. Many thanks to Dank, Alan, Chris, and the others for a very good thread.

I have to say I am siding with Alan on this issue pretty much completely. The thoughts raised and debated were clearly stated by all parties and my thoughts line up with Alan. There is clearly no other reasons to use "cloaking" other than to hide the content from search engine users. Now, with IP delivery, agent-based and personalized, you can have another purpose.

[Alan Perkins]

The debate continued here (http://www.ihelpyouservices.com/forums/t1133/s.html)