Google cleans up Next Blog >> Presses

Back in January 2007 I did a study on next blog button pushes on blogger and blogspot. Google puts a bar across blogpost hosted blogs which lets users, well, log in, post, search or randomly reach another blog. It's really an advert for the service.

I counted how many button pushes it took for me to reach 100 English language non-spam blogs. It took 241 button pushes. I also discovered a lot of spam; in particular redirects to porn sites.

I've repeated this exercise this week and can say that the random button never once took me to porn at all. On a few occasions it took me to a hard sell site that was essentially a spam landing page.

Google's cleaned up the button. So, it is better? Ah... last year it took me 241 button pushes to find 100 English language sites. This year it took 309.

The difference is in non-English blogs. They've doubled from last year. There are a few reasons for this; Wordpress and SixApart are likely to be clawing market share away from Blogger in the UK, Canada and the USA. The Next Button isn't entirely random it; it's reciprocal and some geographic areas (thus languages) may be more likely to press the button than others.

My chart for 2007 looked like this:



The chart for 2008 looks like this:

Be Social: del.icio.us, Digg, Furl, Reddit

Click Spam

I've safely landed in New York for SES. One of the first things I do after I arrive at the Hilton New York is check Gmail.

One of my honey pots picked up a spam email. I'm seeing this sort of thing more and more. The spammer can be pretty confident that if thet target clicks through to Google (don't click on URLs in spam emails!) that they'll click on the intended URL.


The goal here isn't just to drive traffic. Well; I suppose that depends on how smart the spammers are! What they're trying to do is show Google's algorithm that there is a traffic demand/search frequency for their n-grams, that people click on their URL and that they must therefore by the authority.

I imagine emails like this will really annoy Google. Less carefully staged emails might result in people clicking on paid ads and that'll upset people. This is also a type of spam that will be hard to report. Some users may struggle to tell the difference from a spam designed for search and other spam - or, perhaps, between spam and an unexpected email sent because the user legally gave someone permission to contact them.

Be Social: del.icio.us, Digg, Furl, Reddit

Matt Cutts corrects IZEA's Ted Murphy's mistakes

Right now we've blog posts from Michael Gray and Andy Beard which are picking up viral steam and complaining about Matt Cutts insisting that all links within paid content should be nofollowed.

Here's the catch. He didn't say that. It's a misquote. Scan down the original IZEA blog post and you'll find Matt himself making that clear.

I think quoting me as saying "ALL links inside of any sponsored post should carry the no-follow tag period, regardless of whether they are required, not required or even link to the advertiser paying for the post" is different than our conversation.

That's really diplomatic. 'I think the quote... is different from our conversation'. It's a nice way of saying; "You're wrong".

Besides, this whole debate is messed up anyway. I would support any search engine call to insist that all links within a pay-per-post review had nofollow. I would.

There's a big difference between an IZEA style pay-per-post review and other commercial content on the internet. The people complaining want to ignore that.

I really do see the pay-per-post reviews in the same was as TechCrunch does. They're corrupting noise on the internet. They're used, pretty much exclusively, to game SERPs. Those are two labels which couldn't be applied to other 'commercial content' on the internet - you know, like a retail website.

It makes sense that all links within the pay-per-post review to be nofollowed because the whole review is not a valid 'editorial comment'. It would stop people finding a loop hole in the system by accepting a PPP for Site A and linking to Sites B and C - with the intent of passing PageRank to Sites B and C all along.

Be Social: del.icio.us, Digg, Furl, Reddit

Think of the human. Write for humans!

One of our golden rules about SEO is that copy should always read well. When we were a smaller company we would get people from the finance department to proof read copy for us. We wanted to avoid the suggestion that the author had been too eager to stuff in keywords. These days each country as a department of Search Copywriters who are capable of writing great copy and also getting an appropriate percentage of keywords in.

Google and the other search engines do need to see keywords on a page. They're not psychic. However, one of the biggest crimes that SEO has inflicted on the web are pages which simply churn out a mantra of keywords. They look and read awful. They're written for robots and not people. They drive me mad.

David Cushman and Sean Warwick - both from the publishing industry - are calling SEO spam. Sean thinks writing with a 15% keyword density produces awful copy - and he's not wrong. David hates the way SEO 'tricks' you into clicking into off-topic content.

Of course, Sean's been given harsh advice. You don't need to write at 15% keyword density! Please. Please don't try and write at 15%.

David might be being a little harsh himself. Proper SEO (okay; let's use the word 'ethical') is about ensuring the search engine can see what your site is about. In fact, proper SEO includes basic tips like ensure web page content has an unique URL so communities and forums can link to it (as opposed to an Ajax style or single Flash URL which is used to display all the content). It tends to be the horrid combination of search spam and poor websites which result in Google searchers clicking themselves into a worthless experience.

Google's increasingly good at analysing copy. Do you want to bet that a website that's full of grammatical errors is a website giving off negative quality signals? Do you want to bet that a webpage with an unusually high density of a particular word is a webpage that's giving off negative quality signals? Pretty safe bets, huh?

Let's have a look at what Google's Adam Lasnik has offered for advice on this topic:

Our algorithms want to see something that's a happy medium cleanly between:

Extreme A: Not listing relevant terms at all on the page.

Extreme B: Focusing on increasing keyword density to the point that your English/Writing teacher would thwap you with a wooden ruler. Hard. Repeatedly.

And I'll let you in on a little algo secret: There is no single magic number. People who say "The guaranteed optimal keyword density is [x]%" would ideally meet the same fate from an angry English teacher. Or Googler or Webmaster.

And lastly, let me respectfully (and pleadingly) reiterate one key point: The fact that you *can* find sites that rank well for a particular keyword engaging in "keyword stuffing" is NOT evidence that such keyword stuffing is an effective SEO tool. I can also show you many sites that use the letter "Q" exactly three times that also rank well. And no, this is not an indication of a secret "jump the 'q'
rule."

Be Social: del.icio.us, Digg, Furl, Reddit

Selling Links Can Harm Your Site

Danny Sullivan has put together a well written post on Search Engine Land titled Official: Selling Paid Links Can Hurt Your PageRank Or Rankings On Google.

The title says it all.

Back in March I suggested that Link Sellers are taking a risk with their site. I don't think we'll ever see the black mark I suggested - as Danny points out, this makes it easier to try and successfully buy links - but Google's gone further by actually starting to penalise sites.

What strikes me about Danny's post is how it came about it. We had seen some evidence that Google was acting out against link sellers but nothing that would earn the "Official" tag. Danny writes;

Last week, I noticed the Stanford Daily had dropped from when I wrote the above in April to PR7 today. That's a huge drop that has no apparent reason to happen. Some others were also reporting PageRank drops. So I pinged Google, and they confirmed that PageRank scores are being lowered for some sites that sell links.

That relationship with key people in Google is just one of the many reasons why its worth keeping tabs on Danny and Search Engine Land. I'm hugely busy. I fly around the world - but I always try and keep up with Search Engine Land.

So, now you have two actions you can take when your competitor (or client competitors) start to buy links.

1. Report them to Google
2. Contact the link seller and point them to http://searchengineland.com/071007-173841.php

I know a lot of SEO folk still think Google is wrong to punish paid links. I know a lot of them still recommend paid links to their clients. Whether you agree Google is right to do this or not, Google is doing this. If you are taking money from people in order to promote their website please have a sit down and a think about your paid link stance.

Be Social: del.icio.us, Digg, Furl, Reddit

Paid link, again

Thank you! I was beginning to worry that I was the only one.

When I hear arguments for paid links they are always about “what’s fair” or “what Google should do”. None of them seem to admit or take on board what Google is *actually doing*

Can you imagine trying to defend link buying in a year. Can you imagine saying, “Sure, I knew Google was against [it], I didn’t think it was fair so I kept on doing it?” That’s crazy to me. It’s twice as crazy as Google is also telling us to concentrate on content. Spend the money on content.

That was my comment to John Andrew's write up of Search Engine Strategies with Matt Cutts. It was a cathartic comment to write. It's a good blog to read.

Hat tip to Lyndon of Cornwall SEO and Sphinn for the find. Watch Lyndon's post at Sphinn. He has a knack for finding the good ones.

Be Social: del.icio.us, Digg, Furl, Reddit

Paid Links and AdSense are not the same

One of the infuriating defenses those supporting paid links use is to compare paid links to AdSense. The next step is to criticize Google from influencing how you monetise your site.

Just to be clear; AdSense is a combination of JavaScript and iframes. The presence of AdSense on your site does not effect the PageRank or the search positions of the sites displayed on the advert.

I don’t think I’ve seen people object to what Google considers spam before.

People do not object to Google calling white text on white background spam. Why would they? No one pays webmasters to put white text on white backgrounds so perhaps that’s the issue. People don’t even object to avoiding light greys on white or dark greys on black.

The money is clearly an issue. The fact that you can get money from selling links means that the paid link issue is entirely separate from hidden text.

That said, I wonder if the people who are happy to sell links would be happy to take money to hide text on their site or sell hidden links on their site. I doubt it – they would be risking their own site and that is not a risk they would be willing to take. However, I suspect we wouldn’t hear the defense “Google can’t penalise me for monetizing my site!”.

People made money from linkfarms. They were advert supported. I’ve not seen anyone object to Google cracking down on linkfarms. Why not? I suspect this is because linkfarms came and went long before many of today’s popular SEOers were in the industry. Heck. Search engine optimisation wasn’t even an industry back then. Back then we did not a “blogoscope” large enough to let people voice their collective dissatisfaction or even gain kudos for being outspoken about their dissatisfaction.

Paid links did work. Paid links were a toy that people had. Paid links were a toy that made money. Google’s now taken that toy away and that’s why people are upset.

I can see why people are upset. I can sympathise as well. I can sympathise to a point. Trying to compare paid links to AdSense confuses the issue.

Paid links without nofollow are simply there to try and influence search engine rankings. They are.

Paid links with nofollow must be there for the traffic. Paid links with nofollow do not influence search rankings.

AdSense is there for the traffic. AdSense does not influence search rankings.

Do you know what? I won’t be liked for this post. It’s probably a good time to point out that my I Support Nofollow campaign as been running since January. The campaign has only one signature so far: mine.

Be Social: del.icio.us, Digg, Furl, Reddit

I support Matt Cutts

Ah-ah. Lots of SEO people normally like to reinforce how much they like Matt and how they got on well with him. I've only personally spoken to him a few times at SES this year - and I'm not going to claim anything more than that. What struck me was how pleased he was to talk to other search engine programmers but also webmasters.

I've not been afraid to be unpopular with this blog and say that I support nofollow. I even pushed the boat further out by explaining why link sellers are taking a risk. I imagine that if this blog had a higher readership then I would have had more angry comments.

It should come as no surprise that I wholeheartedly support Matt's post to encourage people to report paid links. Use the spam form (authenticated or not) and include "paidlink" in the text body.

A lot of people are hostile to this. Why?

• They've sold campaigns to clients and explicitly said they would buy links - now they're exposed
• They've convinced themselves that some paid links are appropriate - I think this is the most common one
• They make most of their money buying or selling links
• They need to buy links in order to get their SEO to work

The thing to note is this - Google's guidelines have said that ethical agencies report spam. They have said this for a while.

Be Social: del.icio.us, Digg, Furl, Reddit

Rant: What is cloaking?

My oldest blog was once titled “The Rants”. I like to rant. Here’s a rant. This is a personal bugbear rant. We all have bugbears, sometimes they are justified and sometimes they’re not. Let me know how I do.

I argue that cloaking is a very specific activity.

Firstly, cloaking is server side and not client side. In other words you do not cloak with JavaScript or CSS. Oh yeah, Google’s public release of Website Optimiser (it’s actually spelt with an S in the UK) is responsible for this rant. Website Optimiser uses JavaScript (rather nicely) and so it is not cloaking software.

Secondly, cloaking is a deliberate attempt to trick the search engines. Cloaking is malicious. It is sneaky. There are some techniques where you use server side detection to trigger a redirect or perhaps strip a season.

This year at SES London I flapped my arms around when David Naylor tried suggesting to someone in our Site Clinic audience that she try cloaking. In this instance David was not suggesting a black hat technique. He really was trying to advise the woman on one way to handle having multiple sites for different geographic locations, have them exactly the same but still not worry about duplicate content. I also remember disagreeing with Ammon Johns during a NMA roundtable in that there can be such a thing as “ethical cloaking”. I think Ammon was also citing geographic detection as a valid example. Once again I argued that that is geographic detection and not cloaking.

There is a debate as to whether Website Optimiser might be interpreted as hidden text. Could you use it to “test” an <h1> tag which normally says “Andrew Girdwood” with one that said “ARHG!” and never stop the test? Well. Yes. You could. I wonder if Google will receive a flood of “I was multi-variant testing!” defences after sites are punished for spam.

Be Social: del.icio.us, Digg, Furl, Reddit

Why link sellers are taking a risk with their site

I’ve already tried to make myself unpopular with the I support nofollow campaign (which is currently supported by only one blogger; me) and my reasons why I support nofollow.

My blog-sense is at it again. I just know that by pointing out selling links could be risky I suspect I'll attract even more frowns.

Right now link selling seems to be risk free. Google, via Matt Cutts, have made it clear that link sellers loose their ability to pass PageRank on. They loose their ability to vote for the sites that they link to. The link buyers cannot see that and so keep buying links from the caught link sellers.

But what if Google tweaked things so that link buyers could see which sites had been caught selling links? What if the PageRank bar in the toolbar turned from green to black for link sellers? The link seller would find themselves with a black mark. Quite literally.

It's hard to gauge whether Google would do something like that. Right now, as long as Google is confident that link selling is not corrupting the search results too much then the search engine will be happy that its defences are strong enough. After all, Google doesn't mind if people pour money into useless bought links.

Google certainly seems keen to avoid removing quality sites which just happen to sell links from its own search results. Many high profile sites sell links. The invisible penality which prevents these sites from passing PageRank on does not prevent them from gaining PageRank or even gaining high search positions.

A public black mark approach wouldn't disrupt that desire either. Quality sites which earned a little cash through link selling would continue to rank well. The only change would be that those who knew where to look would know the site had been caught. One risk could be on the legal front. There may be legal issues in calling someone a link seller if they're not. False positives are not a problem in the current system because no one knows they are there.

Would a public black mark encourage less people to sell links? I think it would. Some people simply would not want to risk entering Google's bad books. If the black mark became a significant stigma or shame then even more webmasters would shy a way from selling links.

Ironically, I suspect a public black mark for link selling would encourage more webmasters to use 'nofollow'. In order to avoid the risks of false positives and being wrongly labelled as a link seller then more webmasters and bloggers would be far more liberal with their application of 'nofollow'.

There may seem to be no real risk in selling links today. Tomorrow black marks might start to name and shame sites around the web.

Be Social: del.icio.us, Digg, Furl, Reddit

Why do I support nofollow?

Since a meeting cancelled itself on me this morning I thought I'd fire off a cheeky blog post and comment on my support for nofollow.

It's easy for me to whisk through reasons why I support nofollow. Let's begin as a webmaster:

• I can link to sites I want to talk about without boosting their profile in search engines.
• I can code my forums to automatically add nofollow to any outbound links. I don't need to worry about users linking to bad neighbourhoods.
• I can code my comment section so that old comments automatically gain nofollow. I don't need to check back through years of old comments to see if any domains have changed owners/content and become bad neighbourhoods.
• I can make it clear on my site that I use nofollow and this will only ever deter spammers, not encourage them.
• I can add funky and fun widgets, badges, or memes to my site and use nofollow to make sure I'm still "Link Safe".
• I can use nofollow as much or as little as I like and it does not directly impact my site.
• I can use nofollow to safely text link advertise on sites (sponsorships, partnerships, etc) without being mistaken for a link buyer.
• I can use nofollow and safely link to partners without being mistaken for a link seller.

Let's continue my support of nofollow as a search engine optimiser:

• I can educate my clients about bad neighbourhoods, the dangers of linking to them and present the nofollow as a practical and effective technical solution alongside best practise link training.
• I can find safe ways to protect my client's rankings when they may contracted or obliged to promote/link to a partner/advertiser when those partners/advertisers are engaged in unethical search engine optimisation. This solution tends to apply when sites must display a logo or a partner link as those contracts do not tend to mention neither nofollow nor search engines. Deals struck explicitly for the "link juice" tend not to be solved this way; instead I have to advise the client as to the risks in the deal and advise them to re-negotiate and stop link selling as soon as possible.
• I'm against link buying and the option of "nofollow" makes it easier for search engines to address the paid link problem.
• Nofollow is an elegant technical solution which is easily applied to large Content Management Systems.
• Nofollow on cooperate blogs helps encourage comments from users and customers with a valid need to comment and not just from people looking for some well placed link juice.
• Nofollow helps enforce the importance of a "link audit / external link audit". In some ways nofollow creates additional job opportunities for the SEM industry.
• Nofollow is precise. Nofollow is applied to a small unit on a page, a single anchor element, rather than the entire page.
• Nofollow links still show up in your analysis of which sites link to you (Yahoo Site Explorer, Google's Webmaster Console) and so you can still analyse your "social media cyberscape" whereas JavaScript (etc) solutions would prevent this.
• The use of nofollow is a Quality Signal.

Why should you support nofollow:

• The more people and sites who embrace it the more effective it becomes.
• Help avoid the "blogosphere phantom economy" where highly ranked blogs receive a disproportionate number of comments/interaction as other bloggers scramble for "link juice".
• Be ethical - work with the search engines rather than against them.
• Be proactive - help combat search spam and comment spam.

I launched the "I support nofollow" campaign on January the 27th, 2007. To date the campaign as attracted one named supporter - me. Ha-ha. On well.

Be Social: del.icio.us, Digg, Furl, Reddit

MyBlogLog Battles - MyBlogSlog?

Erk. Two heavy hitters are in the process of duking it out. Over at his blog Jeremy Zawodny takes a pop at Andy Beal for 'spamming' MyBlogLog.


I don't blame Zawodny for making the comment. I can see why Andy is furious though.

Why has Jeremy made this claim? He's a loyal Yahoo and a savvy guy. He's seen another savvy marketer use MyBlogLog's features to broadcast a commercial message. This is likely to be something that Eric Marcoullier and his team may not have intended their system to be used for.

Is it spam? It is an image that doesn't look like Marking Pilgrim. This is true. Is the chance to win a Zune representative of Marking Pilgrim? Gosh. I guess that's for Andy Beal to decide. He's certainly not the only savvy marketer to change their MyBlogLog picture.


SEO by the Sea isn't spamming. Bill Slawski is easily smart enough to know that that beautiful picture is going to attract more attention on MyBlogLog than all the other SEO blogs out there. Bill's still playing by the rules.

As a tangent I have to admit how SEO By the Sea has blasted on to my radar. I found it via Search Engine Land, commented at the time that I didn't read it - now I certainly do. Bill Slawski writes to an audience of intelligent optimisers. Too many SEO blogs write to an audience of rookie webmasters learning about SEM.

What we have here, with Beal and Zawodny, is a clash of netiquette. The manners expected from MyBlogLog users aren't yet clear. We don't yet have the ground rules.

You might say that you don't want to look at a Zune competition graphic on MyBlogLog but - c'mon - I don't want to look at the ugly mugs throughout most of MyBlogLog! My ugly mug included! Haha. On the other hand, imagine if every other community image on MBL was a commercial lure?

The debate (despite Andy's anger and Jeremy's annoyance) is good. We're having the conversation. We're thrashing out the ground rules in a new area. That's always needed.

I have to say that I'm impressed with MyBlogLog. This blog sat in "parked" for over a year. I've been active in the social media scene with this blog for just a month, am making progress and much of this is due to MyBlogLog. MyBlogLog drives a significant chunk of ARHG's traffic right now. Further to this you can see just how quickly and how well Eric and co respond to the blogosphere. Despite Yahoo making them a bit richer I can see that MBL senior management are still here for us. They're still hands on.

Eric will need to be hands on for a while longer. This won't be the only MyBlogSlog we see this year.

(The I support nofollow campaign reaches day 5 with 1 'heroic' supporter.)

Be Social: del.icio.us, Digg, Furl, Reddit

Social Media - taking causalities

I don't need to point out that there's a lot of interest in social media and linkbaiting right now. That's a good thing. However, we're going to see many social media sites close down this year.

There are people who think social media is new. It's not. Over at work our RSS feed has been in place since 2003. It's common to find a blog in its fifth year. Despite all the interest and enthusiasm some of the original social media sites are struggling and some even dying.

The social media field is getting tough. Spam is on the increase and some of the older sites are shutting down. Maple is closing down. Maple has been a bookmark manager since 2005. It was one of the first to integrate with Firefox extensions. Taggle is closing down too. SocialBookmarking.org is a state of disrepair.

It's worth noting that Taggle.de and Maple.nu were multilingual. Social Media is strong at the core right now - but only in English. It's the Asian languages which are vying for second place. Mainland European languages are punching below their weight.

Spam is a problem too. Digg's battles with its algorithm and users are infamous. Spend a minute reviewing the drivel that comes into del.icio.us and you could loose heart. Over at Linkatopia, Robert Giordano is looking for moderators to help Design215 fight the spammers. Linkatopia say they're one of the first to fight back against the spammers. They are. It's a tricky battle though - will they get any moderators? I certainly don't have time to help. What about false positives? Imagine if you social bookmark network started to delete your bookmarks?

Adam Loving at Lookmarks notes the problem too. Lookmarks had a good year. The site made people's lists. As a result Lookmarks is on spammers' lists too. He's left to weed out the trash himself. He's put on a brave face but it must be a soul destroying task.

The social bookmarking sites face a problem that the traditional hierarchical directory sites do not have. Directories like Best of the Web don't need people to work, they need websites and the incentive of PageRank to bring in customers. Social bookmarking sites don't have customers, they have users and will struggle without them. One website can be submitted to many different directories but it is unlikely that one user will submit to very many social bookmarking sites.

Simply put - more and more social bookmarking sites will close. Small directories can continue on forever. Small social bookmarking sites won't.


(The I support nofollow campaign reaches day 2 with a mighty 1 supporter.)

Be Social: del.icio.us, Digg, Furl, Reddit

I support nofollow!

I'm just going to come right out and say this: I support nofollow.

I support the optional enhancement to the web. If you don't like it - don't use it. If someone links to you, but nofollows the link - you'll just have to accept that they didn't want to 'vote' for you.

This isn't a popular view among the SEO community right now. Graywolf's diatribe against nofollow [follow allowed - my choice] is proving popular. I'm prepared to be unpopular on this one. :) Do read Graywolf's post if you have time and the comments on it. Matt Cutts' and Jeremy Zawodny's comments, in particular, are worth your time.

I support nofollow

• Andrew Girdwood

Support nofollow too? Drop me a comment and I'll add your name to the supporters list.

(This blog was kicked out of the long grass less than a month a go. Readership is tiny. A lack of fellow nofollow supporters is as indicative of the low readership as much as it is of lack of nofollow support.)

Be Social: del.icio.us, Digg, Furl, Reddit

Wikipedia and nofollow - the outcry, why?

I'm not surprised at the number of SEOers who are deeply annoyed at Wikipedia adding "nofollow" to its links. Why would it bother anyone? Lots of SEOers are stating that they're not wiki spammers and they're annoyed. They're not wiki spammers so why are they annoyed?

1. The principle of thing
2. They might be affected later
3. They dislike the "nofollow" concept and oppose its use

I'm reminded of the anti-speed camera popular opinion here in the UK. "I never speed" is a common claim, "But why should the police get money from speed cameras? It's just a way to earn money." Drivers hate speed cameras - even those who don't speed. My pet theory - and it applies to wikipedia - is that even safe drivers subconsciously acknolwedge that sometimes they might speed. They might get caught. That would even more "unfair" as they're generally a safe driver. They can't reason with or explain that to a speed camera and therefore they hate speed cameras.

SEOers who don't spam wikipedia may sometimes benefit from a link in wikipedia. Now they can't. They can't debate or reason with the code. Therefore these SEOers dislike the "nofollow" policy.

I'm indifferent. Wikipedia still has a place in any online marketing campaign. If you work with big names and big brands you need to make sure that their wiki page does not have any inaccuracies in it.

I'm also a cynic (my default nature). Is there a connection between Wikiseek and the nofollow rules. This quick little diagram attempts to see if there is a connecting line.


Wales said that Google's algorithm couldn't out perform human decisions. I wonder if wikipedia's human decision has helped or hindered Google?

Be Social: del.icio.us, Digg, Furl, Reddit

Safari for Windows

If Mac released Safari for Windows and it became popular then how many web sites would suddenly need a re-design? How many would need to tweak their CSS and XHTML? More than a few, I would say. Mass re-designs would shake up SERPs.

ZDNet are running a poll on the issue and although it is early days yet (early days favour the Windows masses due to mathematical distribution curves) the idea of Safari for Windows clearly has a lot of support. Mind you, Mac have just announced the iPhone and everyone loves Mac just now. It's the fashion. Mock Mac? No. Not me. You must have been thinking of someone else.

If we do have a three-way field in the browser race; Internet Explorer, Firefox and Safari (or a four-way if I'm kind enough to remember Opera (and at least one Organic Search Manager at work is a fan)) then we're likely to see more web sites employ user-agent detection scripts.

User-agent detection is always risky. Google (Yahoo and Live Search) need to be sure that the web site they are seeing is the same web site that everyone else is seeing. If user-agent detection is place than there is doubt. Sometimes some features simply do not work on other browsers and an alternative presentation is necessary. At that point you are showing a different version of the web site to different user-agents.

In fact, ZDNet's Mary Jo Foley has dug up some obscure suggestion that Mozilla is expecting Mac to bring Safari to Windows. Could they? Yes. Would they? That's harder to answer.

This is not a problem which the search engines would struggle to serve. Googlebot already visits sites pretending to be Internet Explorer in order to try and catch cloakers out. That's one of the reasons why user-agent cloaking is scoffed at.

If the web was viewed by a competitive range of browsers and user-agent detection became commonplace then the search engines would have to respond by spoofing their own user-agent more often.

All this comes at a time when the SEO community are discussing whether Google is/whether Google could parse CSS. Google is. Google can. It strikes me as madness to suggest that Google could not cope with CSS. Of course Google can cope with mere CSS! Of course! It's just text. It's simply the case that (and we're back to the distribution curves) that CSS checking has not been worth Google's effort. For all the benefits that CSS spidering would bring the cost would be too great. It's a numbers game.

CSS continues to grow in popularity, so hidden or conditionally visible text becomes more and more common, and it is therefore not a surprise to find that Google is taking stock of things. It could well be the case that Google concludes this look at CSS and decides it is still not worth the effort. This would not be unheard of. In 2004 a test Googlebot was seen scooping up .js files. We didn't see this bot back in a hurry.

Google can cope with JavaScript too. For heavens sake, spammers can defeat captchas, of course Google can cope with JavaScript and CSS. Finding a way through captchas is worth the spammer's while. Picking through JavaScript may not be worth Google's while.

... unless, that is, alternative browsers like Safari become more popular and JavaScript based user-agent detection becomes the norm.

Be Social: del.icio.us, Digg, Furl, Reddit

Answers.com sells links and still enjoys Google's favour

I don't do link buying. The search engine optimisation agency I work for does not do it either and takes the "white hat" approach of directory submissions, articles, press releases, social media and encouraging the client to have the right sort of quality content instead. It's important to me, to us, to do it right.

Back in September 2005 Matt Cutts wrote:

I wouldn’t be surprised if search engines begin to take stronger action against link buying in the near future.

The quote is from the Text Links and PageRank post.

Answers.com enjoys a special relationship with Google's GUI. If you search for any word or phrase then, in the top right, there is a link to Answers.com page about it. There are somethings to note here.

• For the most part Answers.com's context is duplicate and taken from elsewhere
• Google crawls Answers.com

I've never needed to scroll to the bottom of an Answers.com page before. However, I was checking why an Answers.com URL appeared as backlink to a finance site today. The answer? Answers.com seems to sell them.


Yuck. Just yuck.

Dictionary.com must have wept the day that they lost the right-hand spot from Google's GUI. Answers.com must have clapped with joy. The prototype Google GUI, which everyone has seen, with the PageRank-like bars on the left of the page looses this link though. Answers.com may still loose out. Further more, Google has the define search command which it could link in instead or, perhaps one day, use GoogleBase.

In the meantime I hope Google continues to crack down on link buying. A great way to do so would be to scare the heck out of link sellers. If you could harm your site's Google's presence for selling links... would you? No.

Be Social: del.icio.us, Digg, Furl, Reddit

Blog Comments 2.0

Google have provided a brief but handy write up of a year in Google blogging and the blog's parting words which caught my interest.

And before long, perhaps you can begin leaving comments directly. We're working on that.

Oooh! As someone who works with international companies on their SEM campaigns I often find myself in discussions on the pros and cons of comments. More and more companies receptive to the notion of blogging. It's still new but now no longer a fad. Increasingly consumers are expecting to be able to communicate, or at least hear, from companies in a friendly "real language" and not in the less friendly "PR language" or second hand from old media. More companies seem to recognise this and wish to provide a blog. Oh, sure, this is not entirely altruistic as these companies also have an eye on the PR wisdom of a friendly blog. There's also the role of the blog in the SEO strategy. When was the last time you saw an official PR release on Digg's home page?

But there's a catch. Comments. If you let users comment then... gasp... they might say something negative. They may even say something illegal and here in the UK that then becomes the problem of the company providing (or even hosting) the blog. You can moderate comments but that requires effort. What's the ROI benefit from blog comment moderation? How much would you pay someone to moderate comments on a corporate blog?

I often point to Google blog as the easy solution. Simply put, Google has a friendly blog but it does not allow comments. I don't think the blog suffers at all from this. I go there for my official Google news. I feel that I'm closer to Google because I have access to the blog.

I'm sure Google considered letting people comment on the blog but that really would open the flood gates. Can you imagine how many inane comments they would get? The blog spam alone would be intense (Google's blog would make one hell of a honey pot!). The "to reply or not to reply" debate would be worthy of a Shakespearean drama.

Google are considering allowing comments in 2007? Well! Google are just considering it. It would be quite understandable if they conclude; "Nah! That's a crazy idea!"

... but what if Google are working on clever automated and scalable solution? Hmm. Nice. Give Google a problem and they'll naturally gravitate towards a clever automated and scalable solution.

Back in April I waffled on about captchas and cash and came up with a half-baked deposit idea. Simply put, you pay a deposit whenever you comment. If your comment is important enough to you then it should be no problem to deposit 1p until the blogger approved the comment. In a year you might end up loosing, oh, maybe 25p? A blog spammer could deposit thousands of pounds and would stop being a blog spammer. What's the ROI for moderating comments? The blog would have a revenue stream of its own (and, I admit, a new public relations hurdle to overcome).

That's one solution. I very much doubt that this is anything like Google's solution for blog comments in 2007.

Be Social: del.icio.us, Digg, Furl, Reddit

Ebookers

Ebookers are apparently weighed down by a large spam penality in Google. They don't even rank for their brand name. They've been there for a week. Further evidence that big spenders simply can't buy their way back into organic resulsts. Cendant, ebooker's owners, spend a fortune++ at Google.

There is discussion as to what got ebookers caught. Lots of people have found and pointed the finger at the bought links. True. I think that was a yellow card.

The red card, however, are the hidden links on the supporting sub-domains. Check out the source code of http://hotel.ebookers.com/common/maps/MapSearch.asp?rid=3 and you'll see what I mean.

A screen capture says it all.

Be Social: del.icio.us, Digg, Furl, Reddit

Google advise: Hide text

I'm being obtuse... but...

Google asks that you hide text. Google Analytics uses display:none to hide tracking tag values within a <textarea> tag set.

I'm being obtuse for many reasons but one clear response to this is that search engine spiders should never access the confirmation page that this code is designed to sit on.

Be Social: del.icio.us, Digg, Furl, Reddit