Shared publicly  - 
 
Google+ is using rel="nofollow" on some of its internal links. Why?
Link: http://goo.gl/mod/6vzQ
8
Dan Petrovic's profile photoEdwin Jonk's profile photoEnrico Altavilla's profile photoLyndon NA's profile photo
30 comments
 
I don't understand "rel="nofollow". Why is this answer important? 
 
Google says we should not be PageRank sculpting and that there are better things to do with our time, yet I see them nofollowing internal links on their own social network. It's confusing.
 
do as i say... not do what i do... eh?
 
Well... in all fairness it's two different companies or entities. I'm just curious to see why they do this and what is to be gained.
 
As a method to deter SEOs from tryingf to abuse G+ from the get go?
Because it's User Generated Content, and we all know you shouldn't trust it?
Because G doesn't have the time to moderate every post on here and check every Link?
 
Give us some examples then :D

(I doubt if they are "sculpting" as that doesn't work ... it could be trying to avoid screwing with crawl priority ... or it could be dev silliness :D)
 
Your link in this comment is rel="nofollow". If I +Lyndon NA then the nofollow is dropped.
 
Yes
Why should any Value be passed simply because of a unidirectional commitment?
G may be basing "value" on omni-directional communication.
Further - look at how often others get Plussed In ... what sort of signal is that to you?
 
Why should any Value be passed simply because of a unidirectional commitment?

But there is, some things are blocked some are not and I find it lack consistency at this stage. Still looking...
 
Lacks consistency, lacks logic or an incomplete view or comprehension?

Why not try viewing it from G's side, and try to figure their end game, and see if the pattern would fit.
 
Google wants to replace hypertext and links with it's own proprietary one using Google+ and +1 "links".
 
I don't think "replace" is the right word.
It's relationships.
Links suit site to site, or site-owner to site-owner relationships,
but the general populace is left out of that loop.
So G needs to include additional signals that are less tech-focused.
So enter G+ Profiles and +'s, shares and mentions.
Two often complimentary signal sets.

If you see contradiction or a lack of relation between the two sets - it is likely a fair indication somethign may not be within normal bounds.

G needs to control it though - as it knows things are easily abused.
Thus the initial use of nofollow.
?
 
True, but you are thinking short term. I speak about the long term strategy. All that nofollow, unnatural links, penalty, authorship and real name stuff points in the same direction. Google wants to won the Web, rewrite the rules on the technology side.
 
Yes obviously but why they do not block URLs which anyone can drop in is beyond my comprehension.
 
There are various attempts at spam, some direct (I just deleted one link in comments) and some which are a bit more roundabout like this one: plus.google.com/ 114462694912953119218/posts

(purposely disjointed URL, merge to view)

Can you guess why they're doing it?
 
That's inserted via JS after inital load (do you not see it flash?) - it's not on page, nor in a frame, and I don't see a noscript alternative
 
Twitter for one when it used to nofollow URLs.
 
Yes - we spotted that at the start :D
It's interesting that G handles the URLs as "real",
yet appears to utilise JS to render the relevant part.

Techncially it would seem that you have URLs with dupe markup/content, and a bit of cloaking.
 
Well, I believe the profiles (and thus posts/comments) are included in Sitemap files - so crawling based on URL discovery is unlikely an issue.

But weighting/valueing ... maybe different.

Alternatively - as this is a G domain, G could technically guineapig the whole thing.
What if G have ditched/devalued PR here, and bases more of it on other values/signals?
 
Agree but, think it's unlikely they'll ditch PR. Why would it display then for pages currently?
 
There is a difference between there being PR, and them using it.

As for the ditching... I don't think they will ditch PR.
If only there was a subset of data that they could play and tinker with to see how far they could go with lessening the PR influence......................
 
yet appears to utilise JS to render the relevant part.

G+ does serve some plain HTML and you can render a G+ URL well if you disable javascript... (think this changed when G+ changed its design, but I am not sure)
 
When you find a link with a nofollow, it doesn't mean that its purpose is to prevent the target URL to get PageRank. It means only that the search engine should ignore it. Ignoring links has profound implications in crawling, bandwidth optimization, speed of indexing, web graph complexity and many other things that you must take under consideration when you are a search engine that has to manage a property with hundred of millions (or even billions) of URLs.

Since PageRank siloing is impossible to achieve using only nofollow relations, the reasons why nofollow relations have been used are probably unrelated to the flow of PageRank.

Also, I would like to put emphasis to the following statement: "PageRank should be on display at the Museum of Ancient Algorithms.".  :-D It is still useful for some needs and for academic discussions, but I think that we should all face its inadequacy for modern and more complex needs.
 
We still don't know all the parts in PR ... if we did, it may be more useful.
Maybe - if G are still unwilling to ditch the damned thing, they should go a step further?

1) Introduce the other metrics as visibles
2) Include them in the SERPs

That way - you can look and not only gauge a site because of it's position, it's Title and it's Description ... but see Googles Trust score in that site.
Add a comment...