Recently, I published an article called Advanced SEO for Affiliate Marketing Links, which explains linking techniques that prevent the transfer of PageRank to affiliate links, while also allowing the page to load faster–in some cases, a lot faster. However, the article has been generating some concerns amongst my fellow SEO practitioners that perhaps my advanced SEO linking techniques are actually violating Google’s Webmaster Guidelines.
UPDATE: I submitted a question to the Google Help Forums, asking whether or not this technique violates the Google Webmaster Guidelines. There were quite a few answers from the “Top Contributors” and even a Google employee responded. As I expected, their responses focused on the fact that this technique manipulates Google, and they desperately tried to avoid the inescapable reality that this method improves the User experience. So the question basically came down to this: is it okay to hide content from Google if it results in a better user experience? The jury is still out, but I’ll update again when I get an answer. Meanwhile, you can follow the heated debate on the Help Forums: Can I use JavaScript to hide affiliate links from Google?
I’d like to take a moment to share my thoughts on this issue, but I’d also like to hear some outside opinions from other SEO/SEM folks–especially all you white hat extremists out there. Hopefully I can persuade Matt Cutts or someone from the search quality team to comment on this post and provide a more definitive answer. Personally, I’m very interested in Google’s official stance on this issue, since…um…I’m currently using these techniques all throughout this site, and if I get booted from Google’s index…my nine children will surely starve.
What are these advanced SEO techniques you speak of?
I should probably recap the techniques I wrote about, to make sure everyone understands what we’re talking about here. Here’s a quick overview:
- Remove affiliate links from your pages’ HTML code.
- If it’s a text link embedded in page content, then just remove the
<a>tags and leave the anchor text there (as plain text). - If it’s an image link, remove the
<a>tags AND the<img>tag. - Replace the
<a>tags with<span>tags, and assign a class name to them (e.g.<span class="affiliate">). - Write a JavaScript function that looks for
<span>elements that containclass="affiliate"and replaces them with your affiliate links. (Yes, the ones you removed in Step 1). - Put the JavaScript function in an external .js file and block Google from accessing it (using robots.txt).
- Attach the JavaScript function to the
onloadevent, which means users’ web browsers won’t call the function until after the page is finished loading. - Googlebot–and users without JavaScript enabled–can’t call the JavaScript function, so your affiliate links are never inserted into the page content.
- Only users with JavaScript-enabled browsers will see your affiliate links. Everything/everyone else sees plain text instead of affiliate text links, and an empty space instead of affiliate image links.
If you still don’t get it, check out this awesome diagram I spent 6 hours making just for you: lazy-loading affiliate marketing anchors.
If you still don’t get it, even after viewing the diagram, then there’s no hope for you. I don’t mean just with this article…I mean your entire life is hopeless and you should seriously consider killing yourself. However, if you do understand…then let’s continue.
The Big Question: Does the practice of lazy-loading affiliate marketing links violate the Google Webmaster Guidelines?
I’ll kick off this debate by quoting some of the related information from Google’s Webmaster Guidelines–specifically the Quality Guidelines portion, which focuses on the things that might get your site penalized.
To make things more entertaining, I’ll turn this into an argument between 2 imaginary characters, whom I’ve arbitrarily named Blackie and Whitey. In this debate, Whitey thinks the SEO technique in question is deceptive and shouldn’t be used. Blackie thinks the SEO technique in question is not inherently evil whatsoever and it can actually benefit users, webmasters, AND search engines. Read what these guys have to say…and then feel free to express your opinion in the comments section.
Webmasters who spend their energies upholding the spirit of the basic principles will provide a much better user experience and subsequently enjoy better ranking than those who spend their time looking for loopholes they can exploit.
Blackie: The SEO technique in question is not designed to exploit any loopholes; it’s designed to benefit users, Google, and me…the webmaster. Users benefit from faster page load times. Google benefits because it doesn’t have to waste resources trying to crawl affiliate links, and because it avoids contaminating the link graph with links that have no inherent editorial endorsement from the affiliate marketer. And as a webmaster, I benefit from higher rankings that might result from minimizing “PageRank evaporation” and decreasing page load times.
Whitey: The SEO technique in question is clearly exploiting loopholes. The most obvious example is the fact that it’s not really decreasing the amount of time it takes to load a page. All it’s doing is delaying the loading of affiliate links/content until after the main content is loaded. If we ignore the onload event, the amount of time it takes to load ALL of the content has actually INCREASED, because the user’s web browser now has to request and process the external JavaScript file.
Blackie: First of all, webmasters can add the code to their existing .js files, so this technique doesn’t necessarily require an extra HTTP request.
Second of all, what the hell are you talking about, “if we ignore the onload event”? While we’re at it, let’s ignore the definition of a “second,” and I’ll argue that this SEO technique reduces page load times by 2 millennia.
The fact is…the onload event is extremely important and cannot be ignored. It’s the web browser’s way of saying “Ok, User, this web page is done loading, and you are now free to begin perusing the content.” It’s also the point at which the “Loading…” message/icon goes away. Sure, some affiliate banners might appear a few seconds later, but that has little or no effect on the user’s experience. All the user cares about is accessing the main content as quickly as possible, and with this advanced SEO technique…the user doesn’t have to wait for your affiliate banners to load.
Whitey: Whatever. Let’s move on to the next section and see if your SEO trick upholds “the spirit of the basic principles.”
Make pages primarily for users, not for search engines. Don’t deceive your users or present different content to search engines than you display to users, which is commonly referred to as “cloaking.”
Whitey: There you go, Blackie! Read it and weep! It clearly says “DO NOT present different content to search engines than you display to users.” Sorry, but that’s exactly what you’re doing. You’re cloaking. End of story.
Blackie: No…NOT “end of story.” Cloaking uses server-side programming to serve different content to different user-agents or IP addresses. My server returns the exact same document to users and search engines. If a user views the HTML source of my page, they will see exactly what Google sees. The HTML code is identical, down to every last character.
Whitey: Yes, but a web page is much more than just HTML code; it’s also CSS, JavaScript, images, Flash objects, videos, and all kinds of other stuff. You may be serving the same HTML document, but you’re not allowing Googlebot to fetch your JavaScript file, and it is this file that instructs web browsers to rewrite the HTML code internally to show affiliate links and ads. So ultimately, the user sees their browser’s rewritten HTML, which is different from the HTML Googlebot sees.
Blackie: First of all, I highly doubt that Google even requests external files to begin with. Even if it did, I don’t imagine Google is capable of parsing JavaScript, using it to rewrite the HTML of any documents that referenced it, and then retroactively updating those documents’ index data to reflect the newly-discovered content. So unless Google is actually trying to fetch my JavaScript–which it isn’t–then blocking it with robots.txt isn’t really doing anything anyway.
Second of all, bandwidth costs money, and I’m not obligated to serve Googlebot anything I don’t want to. It’s my JavaScript file; I have the right to control access to it.
Whitey: You’re right…but similarly, it’s Google’s index and Google has the right to ban your website from it.
Blackie: Yeah yeah yeah…blow me. Next quote.
Avoid tricks intended to improve search engine rankings. A good rule of thumb is whether you’d feel comfortable explaining what you’ve done to a website that competes with you. Another useful test is to ask, “Does this help my users? Would I do this if search engines didn’t exist?”
Whitey: The SEO technique in question is definitely a trick intended to improve search engine rankings.
Blackie: True, the technique is intended to improve search engine rankings, but it’s not a “trick.” That’s like saying AJAX is a trick. Or Flash is a trick. There are always going to be emerging web technologies and languages that Google can’t crawl or doesn’t understand. That doesn’t mean using them is intended to trick search engines.
Plus, let’s consider the two suggested ways of testing whether or not this technique should be used. “Does this help my users?” YES. “Would I do this if search engines didn’t exist?” YES.
Whitey: But would you feel comfortable explaining what you’ve done to a website that competes with you? I doubt it.
Blackie: Um…I spent the last three days trying to explain this, in vivid detail, to every SEO/SEM “competitor” who will listen. I even made a diagram, so idiots like you would understand it.
Whitey: Yeah yeah yeah…suck it. Next quote.
Use of JavaScript is an entirely legitimate web practice. However, use of JavaScript with the intent to deceive search engines is not. For instance, placing different text in JavaScript than in a noscript tag violates our webmaster guidelines because it displays different content for users (who see the JavaScript-based text) than for search engines (which see the noscript-based text).
Whitey: Gee…that sounds kinda like what I was saying.
Blackie: The SEO technique in question doesn’t deceive anyone. I’m explicitly telling search engines they can’t request my JavaScript file, and therefore they can’t see the content it generates. I’m not displaying different content to search engines; I’m simply telling them this content is for humans only…no bots allowed. I mean come on, Whitey, surely you of all people can understand the value of segregation and discrimination.
Whitey: Of course. Wait…what?!
Blackie: Next quote, please.
Avoid hidden text or hidden links.
Whitey: Obviously, your technique is a way to hide links from search engines.
Blackie: I also disallow search engines from trying to crawl my contact form. Does that mean I’m hiding my email address from Google? And I also disallow search engines from crawling my infinite calendar pages. Does that mean I’m hiding my dentist appointment next week?
I’m using JavaScript to create links that only users can see; there’s nothing in the Quality Guidelines that says you must show Google everything that users see. In many cases, that’s not even possible. The hidden links section of the Guidelines even defines hidden links as “links that are intended to be crawled by Googlebot, but are unreadable to humans.” That’s not what I’m doing. I’m doing the exact opposite of that.
Whitey: I just realized something…you’re right. This SEO technique is awesome. Where can I subscribe to your blog?
Blackie: You can click the RSS icon below. It’s one of the 18 social media buttons I lazy-loaded onto this page while you were reading the article content. I didn’t want to make you wait.
Whitey: Well thank you. I appreciate that. Your lazy-loading trickery has indeed improved my user experience here today. I was a fool for doubting you, Blackie. Thankfully, the fog has been lifted, and for the first time…I can clearly see the–
Blackie: –Just click the icons.
{ comment Leave a comment }
I think that in spirit this link-hiding technique is not ‘evil’ – the idea behind it is good. However when you take the guidelines literally – showing different content to spiders and humans, which is what this technique does in the end – it could be seen as a cloaking technique and thus contrary to Google’s guidelines.
Big grey area here. Hopefully we’ll get some more or less official word on it soon.
hahaha very very interesting and nice article!
I’d like to know, what Matt Cutts says about that :D
“links that are intended to be crawled by Googlebot, but are unreadable to humans.”
Your not violating this guideline, that’s for sure.
yeah, Barry get’s it right… is somehow IS a JS cloaking technique, because your showing something different to users and search engines.
Blocking entire pages from indexing is not the same!
But then, link cloaking for PageRank sculpting would also violate the guidelines and a lot of sites actually do that!
Awesome. Just awesome. I have no idea when I last laughed so hard reading an article about online marketing.
On a more serious note – I cannot for the life of me understand how this falls even close to black hat.
You can load ads using java. Google doesn’t download and crawl your ads, nor does it crawl the page the ads link to. An aff. link basically IS an ad. If you put it in external.js or 302 the link or rel=nofollow it doesn’t matter. The net result is the same. Google ignores the link – it does not pass value.
The argument is then what the difference is between having the link visible to google vs. not.
If google can “see” the link, but doesn’t follow it – will it affect how your page ranks and how much link value you pass to other links?
What about ‘onload’. How does googel handle that when determining pagespeed. Toolbardata is used for pagespeed. This includes script – even if googlebot doesn’t care.
This post has raised enough questions for me to merit an entire post, but in short I like the idea and I think it is an elegant solution.
You’re a funny mofo seomofo, I like the cut of your jib.
In the argument between Blackie and Whitie (kudos for bravery there mate) you ignored the ever present 3rd play in this production: Greyie
His take would always be “F8ck it”
Black or white you are putting a lot of trust in the javascript will load and it won’t screw up other elements or scripts on the page.
Why not push it that much further and do it on mouseover…
@David
Because mouseovers aren’t always triggered by googlebots. :)
KILL WHITIE
I think not only is it brilliant and perfectly acceptable to Google, I’m surprised it’s not something that Google requires of all sites. They’re the ones that try to get people to nofollow affiliate links and ads because they can’t always figure out what they are.
This is a perfect, elegant solution. Of course they wouldn’t care. They would welcome it with open arms.
I Disagree. One of the tests Google uses to determine relevance is “how does this site add value?” In fact, the example in the leaked quality rater guidelines talks about an affiliate site with nothing but (let’s say amazon) affiliate links. The example says that such a site doesn’t add any value over going to the actual site (amazon) thus, it’s not a “relevant” result.
By hiding affiliate links from Google, you can affect Google’s thoughts on how useful your site is. That, by definition, is cloaking and against the quality guidelines. If I were working as a Google quality rater, I would flag this technique.
I’m not sure if google uses affiliate links as a ranking factor, but they could. I do know that their human quality raters do though – so I would make a strong case that they are algorithmically relevant.
I don’t see the user benefit of hiding affiliate links from Google. How does that help your users?
Don’t tell me page load, because it’s negligible. It actually makes the site slower overall and uses more of the user’s computer power to add the links. It also makes the links load slower. Depending on my computer I may already read the first paragraph before your .js file loads the links for me. That’s not a benefit.
If you’re really concerned with not having Google crawl them, just use rel=nofollow.
@Ryan I was assuming that Mofo was adding valuable content and not just posting a page full of affiliate links. But Google already has filters in place not to show sites (affiliate or not) highly that provide no added value.
Adding the javascript links doesn’t help users, but it doesn’t hurt them either.
Look, Google’s the one that wants affiliate links and ads not to count. Mofo has just found a way to help them not count them. It’s what Google suggests people do. In fact, it’s in their guidelines as something you supposedly HAVE TO DO.
Personally, I don’t think it’s up to us to hide or otherwise nofollow ads and affiliate links so that Google can do their job. But if you believe in following their guidelines regarding that aspect, then this is as good of a solution as any.
Great post with great “arguments”, and I swear I’ve had this conversation before in some hotel bar more than once. For the record, I’m on Blackies side.
Seeing Jill say “This is a perfect, elegant solution – Of course they wouldn’t care” did surprise me, because as Ryan says, “you can affect Google’s thoughts on how useful your site is”.
However, as Jill says – “Google’s the one that wants affiliate links and ads not to count. Mofo has just found a way to help them not count them.” .
Case closed? Hardly. No matter what’s “right or wrong”, no matter how the Webmaster guidelines are interpreted, Google can (and do) change those guidelines on the fly to suit their own agenda – which is increasingly focused on selling ads and making money.
Now I’m off to Tweet the bigger news that Jill Whalan is taking Blackies side… ;)
It’s an interesting concept, but the argument to suggest that this isn’t “presenting different content to search engines and users” is weak. Throwing in the “blow me” when a valid argument is made just turns this into the rantings of a 12-year-old. It’s cute, but it’s a failed argument.
The fact of the matter is… blocking whole pages from search engines is one thing… blocking portions of a page from search engines is another. By blocking the JavaScript file, but allowing the HTML page which USES the JavaScript file is blatantly serving different content to people than to search engines. It doesn’t matter whether or not you believe Google-bot cares about JavaScript or whether it processes JavaScript, etc… that isn’t for you to decide. Google-bot MAY process JavaScript… or if it doesn’t today, it may in the future. Nonetheless, it isn’t for you to decide. The only thing you need to be concerned with is whether or not you are specifically treating Google-bot differently than users. If so, then you are clearly violating that aspect.
So you’re doing it to speed up the page for users? Great. Then allow it to “speed up the page” for Google-bot. It really isn’t that simple. You can spin it all you like, you’re just spinning it.
Blow me.
The real question here is: Who benefits from this? Certainly not users as pages will load slower using the technique described to build affiliate links in the page.
There is no way that using JS events to add links to a page is faster than having those links in the HTML with rel=”nofollow”. Additional JS includes and an onload DOM manipulation to add links cannot make a page faster. This was the example used in your original post, and the rationale your extolling here.
Let’s call a spade a spade – this exercise is more about hiding affiliate links from Google. It’s up to webmasters if they wish to do so, just like it’s up to Google to decide if they like it or not. People should make informed decisions based on risk attitudes, and their own interpretation of the guidelines.
Honestly very surprised that Jill Whelan is saying that this is “perfectly acceptable to Google”. I’m willing to wager that this will never be black/white, and will always come back to intent. I’d make sure you understand the risks (even if they are minimal) before using this technique.
(For the record I quite like this technique, and have used something very similar for anchor text distribution on quite a few websites. I just think you should sell it for what it is rather than dressing it up as something to benefit users.)
Richard,
First of all, there is a big difference between the time it takes to load the content the user wants to see, versus the time it takes to load everything on the page.
The technique slows down the latter, but that’s of no concern to the user. The user only wants to see the desired content, and THAT is what’s being loaded faster.
How would most people answer this question:
“You can either see your desired content now…or…you can wait until after these advertisements finish downloading. Which do you prefer?”
Obviously the user wants to see the content that corresponds to the search query that sent them to your page.
When you click on a YouTube video, do you want to see the video immediately or do you want to watch a commercial first?
It doesn’t matter what my intent is. If I didn’t know what PageRank is or I didn’t know search engines exist, this technique would still have the same result.
Is this about third party ads or about affiliate links? You can flip flop all you like – it just makes you look foolish.
If this is about 3rd party stuff that’s fine and it makes sense to defer that. If it’s about hiding affiliate links in the “desired content” that’s also fine, but don’t bother trying to paint it as something that benefits the UX.
Richard,
Yay! You actually made an attempt to address the issue!
Why is it so difficult for you and your Bionic friends to just answer a question without trying to make it into a fucking soap opera?
This isn’t some kind of elaborate conspiracy that needs uncovering–it never was. I spelled everything out from the very beginning. The very first post I wrote contains examples of plain text links and banners. If you think the banners are acceptable because they improve the user experience, but the text links are not acceptable, then…um…I don’t know…maybe you could just fucking say that?
I feel sorry for anyone who actually goes to the Google Help Forum for help. It’s a goddamn circus over there, and none of you know what the hell you’re talking about. Just a pack of jaded losers, trying to compensate for your lack of knowledge by attacking the poster’s character.
I guess when you don’t have a technical leg to stand on, every question gets an emotional answer.
Where in this post – thoughtfully titled “Hey, Matt Cutts, I’m using JavaScript to hide links from Google, cool?” – do you mention using this technique for banners? This has always been about aff links. I’m sure everyone who reads this, and the original post, will identify that fairly quickly.
Or…maybe they’ll read it and identify you as the fairly retarded Help Forum troll that you are? That is…if only I could scrounge up an example of me mentioning banners somewhere…
The following quotes are all from the original post.
In the 1st paragraph:
Under the section called “Page Speed”:
Under the section called “Don’t Slow Down My Page Speed”:
From the .js code itself:
From “Step 2″:
From “HTML Markup Examples”:
[There's an example called "Image Link"]
Great idea, excelent article.
“I don’t imagine Google is capable of parsing JavaScript, using it to rewrite the HTML of any documents that referenced it, and then retroactively updating those documents’ index data to reflect the newly-discovered content” – and how about that for a next step, Google engineers? ;)
@jillwhalen: “Adding the javascript links doesn’t help users, but it doesn’t hurt them either” – how about linking malware or really spam-pages, then this technique would hurt users.
My thoughts? Sure, is a great way to replace the “nofollow PageRank sculpting”, but it has its flaws for user safety and it is possible to say that it breaks SE’s quality guidelines, although it’s also possible to stand for it.
You might have developed a big trouble to SE’s and it might be very helpful for website rankings while it’s fresh and maybe SE’s do not have a quick or already resolved solution.
Thanks a lot for sharing!
Btw, your comment form has a “website” field, although no comments have links to the comment-author website. Do you believe it is/was helpful to seomofo’s rankings to remove comment author links?
Some authors have active links, but it’s because I white-listed them: Comment Policy
And yes, I think it’s helpful to a site’s rankings to tightly control outbound links.
I can’t remember exactly how many times I’ve seen this whole crappy Google Help Forums issue come up recently, but there have been quite a few. I hate to pile on, but I’ve never had an issue “resolved” by a non-googler over there.
In my experience, non-googlers seem to either regurgitate the information they just found in a Google search (um yeah, I Googled it too, wouldn’t be asking if the answer was out there), or try to re-word my questions so they can answer it . . . like it some sort of competition for who can get the most questions answered. And the thing is, I genuinely try to appreciate their time spent answering, but it gets annoying trying to re-explain the same issues over and over again. I’ve learned to just work around my Google issues (Gissues?) and avoid wasting my time in the “help” forums.
Summary: Darren said: “Blackie: Yeah yeah yeah…blow me. Next quote. ”
hahaha lmao :D
Actually as a web user in South Africa where our connection speeds (even the good ones) are crappy at best I’m often stuck waiting for ads to download so I can see that actual content on a website.
Great example is Blackie’s next to last comment. Social media plug-ins are one of the slowest loading elements on a page. People often DON’T force them to load last, or at least once all the actual content of the page has loaded. That’s a crappy… wait no, that’s a shitty user experience.
Personally I’d endorse this practice for that reason alone. Like everything it could be abused. But where that line is drawn is entirely up to Google (or any other search engine) to decide.
I reckon anyone trying to take the moral high ground on this one are only doing so because they didn’t think of it first ;)
Am I not right in thinking most of the affilaite networks these days go through a shitload of 302/301 redirects before final landing that they are next to useless anyhow
Surely something like this that helps sort the wheat from the chaff BEFORE Google indexes the page should be encouraged (although it obviously wont as the potential for abuse here is obviously significant).
Great post Darren
great post blackey.. i loved the concept and i think this technique isn’t bad to use. we use robot.txt files and other ways to prevent search engines to view a web page so why can’t we do this practice. i will give it a try. thanks.
I think this has overall been a very informative discussion. I’ve followed the original post, the forum post discussion, and read through all the comments on this post as well.
The fact of the matter is that it is an elegant solution if utilized in that manner. Many techniques that are currently in place can be misappropriated.
After reviewing the script and the discussed implementation, regardless of the reasoning for blocking bots from the javascript file, this is simply a lazy loading implementation.
Many folks seem to be focusing on intent, when regardless of intent, the solution does follow the guidelines based on general interpretation of the person using the technique. If you review the top 10 US alexa sites (all authority sites); they implement similar lazy loading tactics to improve user experience.
My guess is Jill Whalen’s comment referred to the dual use of this implementation which overall seems to be a currently practiced technique (Gmail is using it); user experience improvement w/ some SEO bennies to boot!
User experience has always been the core of webmasters guidelines, not SEO. SEO is the secondary afterthought of applying techniques that improve the users overall ability to achieve their goal on your site.
I agree that much of this was taken out of context in its ability to be used in a white hat method.
I think this is a great way to improve the user experience. How many times have you waited (and waited and waited) on a page for google ads doubleclick BS ???? ALL THE TIME and I hate it!
Google has a double standard on ads. Do as I say not as I do.
Google serves up ads that they make money on, but if I do it then I’m not serving up “relevant content”. Bastards……….
Well, the point is that Google AdSense does not influence rankings, which is what Google is against: using links to manipulate rankings.
If you use something similar to the AdSense program, stating out that it actually is a advertiser program, bought links, etc., there’s no problem at all.
According to US laws, you must alert people when there’s money supporting an ad. When you sell links and doesn’t set up a disclaimer, that’s wrong. Google follows these guidelines to define its paid links policy.
Which I believe makes perfect sense.
You should DFB the shit out of this page to see what comes out of it!
Well perhaps you’re not just some nerd thug afterall… Excellent replication of the style in What the Tortoise Said to Achilles, by Lewis Carroll – which I can’t say I read, but the particular vibe from the writing is unique and forceful – also effectively replicated by Douglast Hofstader’s chapter “a metaphorical fugue on minds and machines in the spirit of Lewis Carroll”, in the book “Godel, Escher, Bach An Eternal Golden Braid”.
Good Job, I’m glad there are people out there making more noise then I am, I’ll just tag along at a safe distance and if you get pulled off the highway I’ll slow down! JK about the nerd thug thing by the way, most people I know would take it as a compliment.
Okay, I’m going to throw a couple of monkey wrenches into the mix on this issue.
First of all, I don’t see any discussion here about how Google offered the “nofollow” option then pulled the rug out from under the people who did use it without telling anyone. See http://www.mattcutts.com/blog/pagerank-sculpting/ where Matt wrote in June 2009, “More than a year ago, Google changed how the PageRank flows so that the five links without nofollow would flow one point of PageRank each.” Google’s justification for announcing the change? From the same article: “At first, we figured that site owners or people running tests would notice, but they didn’t. In retrospect, we’ve changed other, larger aspects of how we look at links and people didn’t notice that either, so perhaps that shouldn’t have been such a surprise.”
I’m sorry, but if that isn’t intent to deceive on Google’s part, then surely putting linking information in code that Google refuses to spider and interpret doesn’t qualify as deception either.
Now, here’s another consideration. I run a website that encourages discussion about political issues by people from all political camps, which I broadly label as: conservative, liberal, moderate, libertarian, and statist. I also require that authors who make factual claims back them up with references and links to their sources. Even though I welcome all views, it doesn’t mean that I want to endorse all those conflicting views by passing PageRank to their sources. For instance, I’m anti-communist, so if communist writer writes on the site, I certainly don’t want to pass link juice to marx.com or whatever. I mean, in many cases, I probably personally wouldn’t even agree with their sources regarding their “factual evidence”. But from the perspective of running an honest, open debate, the linking is necessary. It’s a perfect application for nofollow, and that’s what I did until recently, when I discovered the above mentioned Cutts post and learned that link juice is being passed to all those sites I would never endorse despite Google’s earlier promise that with nofollow, such a thing wouldn’t happen! Deception on Google’s part!
Finally, the classic example that Google gave from the beginning about using nofollow was that it should be used for links that are rented or sold for advertising purposes, so that no PageRank gets passed along as well. But when Google started passing PageRank anyway, despite the nofollow, they basically violated their own precept and passed PageRank to link buyers!!!
Hope this post doesn’t piss too many people off, but even if it does it’s only fair. After all, Google has pissed me off with their deceptions and their double-standards.
Google has been parsing javascript for quite a while already.
http://googlewebmastercentral.blogspot.com/2008/04/crawling-through-html-forms.html
These days, Google has the processing power to index posts within 5 seconds of being published. I’ve seen it with my own eye.
Is there a good reason to think Google won’t retrieve the external javascript file (robots.txt rules ignored?), execute it, and then process the changes that happened after the onload event?