22 replies to this topic

[iamlost]

Barry has an interesting summation on the 'Matt Cutts outs SEOs As PageRank Clueless' hissyfit, Blog Comments And Google.
Note: yes, it is true, Barry is 'The Other Bloke', not to be confused with 'The Bloke', whoever he might be...

I happen to have been one of those who totally ignored rel=nofollow and find the current brouhaha hysterically funny.

Let's do a historical recap (taking a very bogbasic view of PageRank):
BNF (Before NoFollow):
* Each and every link to a page (internal or external) brought some portion of the referring page's value with it.
* Each and every link out from a page (internal or external) took away some portion of the that page's value.

NF1.0 (NoFollow version 1.0 aka Blog Comment NoFollow
* Knowing the historical value of BNF and having ravaged, plundered, and pillaged guestbooks into oblivion PR seeking linkdropping spammers descended onto the riches of unmoderated blog comments.
* In an implicit admission of abject failure in algo/auto determining copy quality and being unwilling to abandon blog comments as they had guestbooks SEs invented the 'link condom' or rel=nofollow.
* And so the great 'dofollow' goldrush began.

NF2.0 (NoFollow version 2.0 aka LinkSculpting via NoFollow)
* A few smart folks realised, all things PR being equal, that with PR being dammed up behind NoFollow it created 'reservoirs' of PR and that by carefully blocking and channelling onsite links could build controllable reservoirs and PR 'irrigation'.
* A few foolish folks shared this idea publically.
* Unofficially, the SEs sort of shrugged.
* And the great LinkSculpting goldrush was on.

NFEE (NoFollow Extinction Event)
* Unofficially the SEs began to talk back their unofficial LinkSculpting shrug. Almost no one noticed, fewer cared, even fewer commented.
* Eventually, in answer to a direct question Matt Cutts implicitly said that LinkSculpting had no value. Given the fortunes/reputations made through the promoting and sale of LS this caused a great shock within that portion of the webdev community and Mr. Cutts' words were parsed for any justifications for their past LS touting.
* And finally Mr. Cutts point blank stated that each NF killed that links PR value portion dead - and worst of all, had done so for over a year. And no one had (publicly) noticed.

ANF (After NoFollow)
* A lot of people have been wailing and running about shrieking that the sky has fallen. Some, for varying reasons, have loudly demanded that Google put it back up or otherwise compensate for their bubble bursting.

What has occurred is that external link value has gone back to what it was BeforeNoFollow: without NF the link carries away it's portion of page value as it ever has, with NF the value is simply extinguished. Back to link value as usual for external links.

Internal LinkSculpting is the biggest loser. Before value simply flowed. It might eddy or sit in backwaters but it never died. But now all that fancy reserviour building and controlled irrigation is seen as actually diminishing page values. Big Woops. Amusingly (publicly) no one apparently noticed. Even Bigger Woops.

Why the silent change is uncertain - the backlash is likely (as Barry has noted) to especially impact blog comments as currently structured. Have the SEs' ability to judge copy quality improved? Had LinkSculpting broadly impacted the SEs access to content they wanted? (I see implicit acknowledgement of this in one small explicit Matt Cutts sentence)

Barry highlighted the most important short term blog comment consideration - in the balance between added content value versus outgoing link drain value the quality of the comments becomes paramount. (Of course, best practice has always implied that but most bloggers aren't exactly enamoured of best practices) The quick fix is gone, moderation remains. What percentage of bloggers will put in the time and effort - will it be seen as necessary?

Structurally it just may force a greater appreciation for site architecture. Because all that LinkSculpting could do so can a proper architecture. Of course the one is more expensive (time and effort) even from scratch, fixing existing monstrosities becomes a nasty cost/benefit exercise.

And sitting here, cozy, the tempest blows. And so many are out there getting wet. Sillies. That's what you get for chasing SE prognostications. Bubble and burst. Welcome to the down side.

Regardless of your pov, read Barry's The Other Bloke's article (and follow the links). A good read.

[bwelford]

Thanks for bringing that to folks' attention, iamlost.

More controversially I have suggested that perhaps this is all explainable if you assume that the PageRank process works in two phases. The basic PageRank calculation is done for all links without any consideration of robots.txt, nofollow tags or even bad neighborhoods. How that first phase PageRank is then used within the second phase search ranking algorithms is determined by all those filters and constraints. I put forward this Null Hypothesis because it seems to fit the Google pronouncements and is computationally do-able. Until someone of authority states that this is wrong or proves it to be so, I'm sticking with this.

[iamlost]

While it is impossible to state how the Google algo might work I have always believed that all inputs are modified after the fact. My only complaint with your Null Hypothesis is that is too simplistic.

But then I don't much care about the SEs. And they seem to like my sites just fine without my caring.

[bwelford]

Perhaps the DKITS principle must take over from the KISS principle to make sure we appeal to the intellectuals as well, iamlost.

DKITSS stands for Don't Keep It Too Simple, Simone.

[Ron Carnell]

It's a shame PR Sculpting has suddenly become synonymous with no-follow when, in reality, SE Optimizers have been successfully sculpting PR for most of a decade.

Frankly, I think the idea of limiting the expenditure of PR by limiting blog comments is self-defeating.

You can't run a successful business by putting all your money in the bank; you have to be willing to spend money to make money. Similarly, you can't run a successful web site by hoarding PR. If you don't spend it, you won't get it. However, just as you don't spend money indiscriminately in your business, you shouldn't spend PR without a real good idea of what you'll get in return. Spend it, but spend it wisely.

Limiting blog comments, in my opinion, is like designing a budget for your business that is based only on money amounts and not on specific needs. It's like saying you're going to spend a thousand dollars this month -- then spending it willy-nilly and running out of money before the rent has been paid. A good budget doesn't just take into account how much you have to spend, it also prioritizes what you need to spend it on. Spending money -- or PR -- shouldn't be on a first-come-first-served basis.

[bwelford]

Perhaps I can put up some counter arguments for discussion, Ron.

First I don't think it's a question of PageRank Hoarding. A given web page has a certain amount of PageRank which flows out through all the links. If later links are allowed into the pool then everyone gets a little less PageRank. According to Matt Cutts there is nothing you can do to stop that.

Often within the blog post I will have links to other relevant blog posts or articles. I also do not add nofollow to commenters links but do adopt a 'tough love' approach in getting rid of any comments that seem to be done only to gain a link.

I block the blog comments after 21 days since I find that the proportion of spamming comments goes up with time. I do have Google Friend Connect for each post so people can still add comments if they wish. Those comments are purely for the human visitors since as far as I know they have no weight within the Google system.

I believe any approach has its pluses and minuses, but this is the right approach for me.

[EGOL]

All of this confuses me.

I have content blocks on most of my pages that change several times per day. I was using nofollow on the links in those blocks because they are temporary.

If I keep nofollow on them some PR will evaporate. If I take it off then it might look like I am trying to manipulate.

[Ron Carnell]

Barry, I think one key to your plan is to answer this question:

Will every visitor, most visitors, some visitors, or almost no visitors bother to read any comments stored in Google Friend Connect?

I honestly don't know the answer to that question, but my first order of assumption is that GFC is a great big wastebasket where people will be allowed to browse if they really want. If we assume that it's not greatly different than simply deleting comments left after 21 days, then I think it becomes clear that your "financial budgeting" is based on a time frame and not on the value of the expenditure. Is that really what you want? If Matt Cutts responds to a post on Day 22 do you really want to throw him into the Back Forty with the spam?

In my opinion, any content (comment) that is worth publishing at all is worth publishing where it can be easily read by a visitor, without ANY additional decisions, questions, or clicks. If the cost of that is a link (which is another decision entirely), then so be it. I personally don't believe that the PR loss to "links to other relevant blog posts or articles" is big enough to accurately measure, let alone big enough to warrant the elimination of good content.

[bwelford]

@Ron I believe the possible comments that might appear after 21 days are so rare that they are hardly worth bothering about. The only ones I've seen until recently when I instituted this policy were clear arguments for instituting such a rule. I particularly want to make sure that links within my post get as much link-juice as I can reasonably give them.

On this topic, but at a slight tangent, Aaron Wall has an excellent post today, Expert SEO Testing: Usually Worthless:
http://www.seobook.com/worthless-hype

[SEOigloo]

I mistrusted (and subsequently abstained from) the whole no-follow parade. I read about it with interest and thought people were clever to figure this out, but it flew in the face of Google's basic premise of doing things for people, not search engines. Since the average human has no concept of and no use for behind-the-scenes no-following, it didn't strike me as a human-centric approach to the web.

It's been weird to see all of this blow up and blow over. I feel out of it, but happy to be out of it.

[SEOigloo]

Barry, that was a good article from Aaron. His blog is always a breath of fresh air, isn't it?

I must say, I agree with him regarding this whole blog comment moderation idea. It's artificial and extreme. Obviously, delete bologna comments. They don't add to the conversation whether they contain links or not. But I'd never delete a genuine comment with links in it that the reader though relevant to the conversation. Why do that? For some mote of hoarded PR? What a lot of trouble and hassle...and rudeness. Let the conversation flow. Make connections with people. If this means being ill-used by someone who is craftier than I am and is disguising their fancy intentions behind thoughtful blog comments, I guess I'll let them carry out their aims. Ostensibly, I suppose this could mean a slightly less elevated ranking for me and, by dint of this, less business. But what about the business I might lose by cutting people off and refusing to let them participate? How can I even measure that?

[Ron Carnell]

I believe the possible comments that might appear after 21 days are so rare that they are hardly worth bothering about.

I understand, Barry. I'm guessing your articles are topical? People either aren't reading the articles after 21 days or simply don't have anything useful to add once the expiration date has passed? Ergo, you get mostly spam.

I don't, however, think that's a rule than can be generally applied or advice that can be offered without a lot of caveats.

FTR, I have 79,774 published comments on my main site. While I don't track it, I suspect there was at least an equal number of comments I refused to publish because of foul language, less than constructive feedback to the author, or some other obvious disqualifier. I read and personally approve every comment that comes in the door, which may sound onerous to some but is still a lot easier than personally creating new content all the time. I depend on my visitors (just as we do in this forum, I think) and give them the time they deserve.

And, yea, I have stuff that was posted clear back in 1998 that still garners good, publishable comments every month. When people stop leaving comments? The content will be removed.

I particularly want to make sure that links within my post get as much link-juice as I can reasonably give them.

I do, too, Barry, but I won't do it by trying to limit legitimate out-going links. Instead of trying to increase link-juice for one internal link, I prefer to create more than the one link. Four links with half the juice is still twice the value of one link with full juice.

The secret to increasing PR where you want it is ALWAYS to create more content. Not try to hide the content you've got behind a non-spiderable veil.

Matt Cutts quickly extended nofollow's purpose to include use on paid text link ads as well. But given that Google AdWords sells links (and often to scammers) some people may have seen trade issues with forcing the new proprietary nofollow tag onto the web. (emphasis added)

Aaron pretty much lost me with that comparison. Prophets, after all, aren't supposed to make mistakes lest they lose their credibility. AdWords sells advertising, not links, and if Aaron doesn't know the difference I have to take anything else he says in that article with a grain of salt, too. Especially when he starts telling me how important domain names are to SEO.

We all have off-days, I suppose. Aaron usually offers a lot of useful information, but I don't think this was one of those times.

I read about it with interest and thought people were clever to figure this out, but it flew in the face of Google's basic premise of doing things for people, not search engines.

Exactly so, Miriam. And I think the ensuing backlash we're seeing is ill conceived for exactly the same reasons. Are we limiting blog comments for our visitors? Or for the search engines?

The later never seems to consistently work for long.

[bwelford]

Ron, I think we are seeing very much eye-to-eye here. My comment re making sure outlinks got the juice they deserved was not for my own links but for links to blog posts by others that I try wherever possible to include when they are relevant.

It is unfortunate that the Google process assigns equal PageRank contribution to all outlinks. Wouldn't it be great if as author one could assign a degree of importance to each outlink? They now have the canonical tag to flag which link points to the most important web page in cases of possible duplicate content. In a similar manner they could have an importance tag. It would specify High, Medium or Low, equivalent to say weightings of 10, 3 and 1. These weightings would ensure that the PageRank was passed through in these proportions to each type of link. This would be somewhat arbitrary but certainly better than the machine allocation that gives them all the same.

[Ron Carnell]

Barry, did you read what Danny said at Matt's blog?

Danny Sullivan: To make matters worse, none of this matter, right? Because we’re being simplistic in all this. The reality is that Google is deciding how to divvy up the PageRank spend.

Google might see 10 links on a page that has $10 of PageRank to spend. It might notice that 5 of those links are navigational elements that occur a lot throughout the site and decide they should only get 50 cents each. It might decide 5 of those links are in editorial copy and so are worthy of getting more. Maybe 3 of them get $2 each and 2 others get $1.50 each, because of where they appear in the copy, if they’re bolded or any of a number of other factors you don’t disclose.

PageRank isn’t spread equally among links and hasn’t been for years and years, right? As I said at the beginning, this is what I long understood to be the case from things Google hinted at before 2007. And at SMX Advanced, you confirmed it to be the case.

One might well presume, as Danny is apparently doing, that your editorial links are already getting the PR boost you want?

[bwelford]

Good point, Ron. I did see that and had not seen the connection.

This is of course all higher mathematics and mathematics requires that you define everything very precisely. I'm thinking of all this partly from what is easily done in computation terms. The easiest thing is a two phase approach so that every link starts off with a basic PageRank score. Then in the second phase when they use that basic PageRank score, they can use multipliers or factors to get a derived PageRank score which is one of the 'more than 100 factors' used in the algorithm.

So if an outlink is coming from a footer, then it gets a discounting factor of say 45%. An outlink that comes from the body content of the blog post might get a multiplying factor of say 165%.

So what actually comes into play in the algorithm is this derived PageRank factor, which may be like what Danny Sullivan describes.

If it is all rolled into one whole ball of wax with repetitive calculations and these discount and multiplier factors, computationally it would be a nightmare.

In consequence I'm sticking to my Null Hypothesis (two phase approach) until someone tells me with authority that there is a better Hypothesis to explain what is going on.

[glyn]

"Originally Google created rel=nofollow in what was claimed as an attempt to minimize the effects of blog comment spam on their search results. But the tag never decreased blog spam, it only decreased the ability of bloggers to influence search rankings by leaving back-scratching comments on each other's blogs."

- NoFollow was created in order to push the workload of limiting an essentially blind crawler into following links that were not the same, and therefore reduce spider workload so other sites could be crawled and the index could be kept expanding with relevant content. It was wrapped up in the guide of making search results better and was directly implemented because it was frighteningly easy to get good results. Which doesn't wash well if you're trying to push a paid advertising platform and build a marketing database of all the local businesses on the planet.

I think SEO's would do well to separate getting in to bed with the search engines with the actual reality. There is a definate distinction in the characters you get in this field, you have the "i believe the gospel according to st.Goog" and those that question it. It was nice to see a few familiar faces come up in Aaron's post.

[nuts]

I'm still a bit mystified by nofollow.

The main principle is *supposed* to be "build sites for people, not search engines".

[bwelford]

Here is what Google says about nofollow.

[A.N.Onym]

So, given that nofollow is nonsense and, quoting Aaron Wall,

When you are competing for core keywords in big, competitive markets the SEO game comes down to industrial strength link building, public relations, social networking, branding, advertising, and other aspects of classical marketing.

why are you still debating this?

[bwelford]

I thought by now we were all agreeing.