homepage Welcome to WebmasterWorld Guest from 54.166.123.2
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member
Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

This 182 message thread spans 7 pages: < < 182 ( 1 2 3 4 [5] 6 7 > >     
Pagerank queries - the <rk> parameter
What do such figures as Rank_1:1:6 Rank_1:1:5 mean?
selomelo




msg:757461
 10:10 pm on Feb 11, 2006 (gmt 0)

We already know that toolbar PR is in fact an historical figure. When you query google for current PR, you get some mysterious figures such as:

Rank_1:1:6 Rank_1:1:5 Rank_1:1:4 Rank_1:1:4 Rank_1:1:5 Rank_1:1:6 Rank_1:1:5 Rank_1:1:0 Rank_1:1:1 Rank_1:1:5 Rank_1:1:2 Rank_1:1:3 Rank_1:1:0 Rank_1:1:0 Rank_1:1:3

These figures are for a site that has a current toolbar of PR4.

I searched the web for an explanation, but failed to get a good one.

Is there any idea as to what all these magical numbers mean?

 

Eazygoin




msg:757581
 12:14 am on Feb 22, 2006 (gmt 0)

Jim, I just read this on a website. I can't post the link, but can sticky it if you like:
"The actual value in the "Real PageRank?" column is the RK value on the XML page where U matches the URL (the first match with or without the www prefix). If there is no match, then "Unknown" is displayed."

Sorry, that didn't make much sense, but it related to two PR scores- the current one showing on the toolber, and a Real RageRank, which hasn't been displayed on the toolbar yet. It states that the RK value is the latter

Jim Westergren




msg:757582
 12:25 am on Feb 22, 2006 (gmt 0)

Thanks Eazygoin.

I knew that they took it from the RK but didn't know they took the first match in the feed.

I think the only question of real interest now is if RK really is Live PageRank or not?

The guy also told me that he has observered that RK is changed often and he has seen an increase in value after backlinks being placed - the same day the new pages with new backlinks have been cached in Google.

I think I will do some tests on this.

Jim Westergren




msg:757583
 1:10 am on Feb 22, 2006 (gmt 0)

Just realized while comparing RK DCs for different sites.

You are now able to see when a DC is updating it's index or has found your backlinks by tracking backlinks by dates on a site and seeing the RK update :) cool.

Now listen to this:

Matt Cutts latest post (Feb 17) shows as RK 3 on the DCs that found it and nothing on the rest. A post from feb 15 shows as RK 6 on half DCs and RK 3 on the other half.

And now this:

On those DCs that the RK of the blog post was 6 compared to those it was 3 he ranked higher in the SERPs!

Example 1:

Difference in RK: 6 compared to 3
Name of post: "WSJ on SEO contests"
Date bloged: feb 15
Query used: WSJ SEO contest
Difference in rank: 3 compared to 5.

Example 2:

Difference in RK: 7 compared to 6
Name of post: "Road trip: Ask Jeeves in Campbell"
Date bloged: feb 13
Query used: ask jeeves Albuquerque
Difference in rank: 3 compared to 5.

Why older blog posts by Cutts has higher RK? GoogleBot has not yet found the the many RSS feeds that are linking to his posts (most probably).

So it perhaps is the Live PR?!

Jim Westergren




msg:757584
 2:15 am on Feb 22, 2006 (gmt 0)

Just verified: yes, the RK is always the same for a URL no matter which query you use.

And I just read an article on how the Google Toolbar works. After information is sent to a Google server, data in the form of an XML document with data about the URL is sent back.

There are values and info on a lot of things and guess what the field of the value of the PR is called?

RK

The same name as the SERP XML, but in the SERP XML the RK is not the toolbar PR, just has the same name. Code name for PR?

Jim Westergren




msg:757585
 3:22 am on Feb 22, 2006 (gmt 0)

Just to add ...

From another document written by Google:

---
U - The URL of a single search result - [Absolute] URL (needs HTML-escaping to view)
T - The title of a single search result - HTML
RK - Google's rating of how good a single search result is - An integer in the range 0-10, inclusive
---

"Google's rating of how good a single search result is" and with "An integer in the range 0-10, inclusive"

PageRank in other words.

To see the document, Google: "Google's rating of how good a single search result is"

EDIT:
BTW: Just realized that the document probably is confidential "The contents of this document are confidential and proprietary to Google." and is hosted by some guy that mirrors controversial and important documents "that is in danger of censorship", hehe.

dmje




msg:757586
 6:43 am on Feb 22, 2006 (gmt 0)

Dont know if I can post this URL here or not, but some one was asking about the firefox extension for the live pr
[livepr.raketforskning.com...]

If the url does not show, can go to the firefox website and do a seach in extensions for page rank and it will show the url

Dayo_UK




msg:757587
 9:33 am on Feb 22, 2006 (gmt 0)

Jim Westergren

And I just read an article on how the Google Toolbar works. After information is sent to a Google server, data in the form of an XML document with data about the URL is sent back.

There are values and info on a lot of things and guess what the field of the value of the PR is called?

RK

Can you sticky me a copy of a link to that document.

Intresting finds :)

I am not 100% sure that the <rk> values are in use until after a data export etc though.

EG the lastest Page Rank update happened using values of about February 2nd - 4th, at this time there were a number of posts regarding a serp shift - so were the latest <rk> values then used in the serps and we need another export for the next <rk> values to be used - especially the Big Daddy DCs.

On this date my homepage changed from a <rk> 0 to an <rk> 4 (5 in big daddy) - but it still cant rank for anything - it might just take a little while for these values to kick in/effect the serps though.

This site is also hopefully coming back from a canonical error - so it might be a bad example/held back for other reasons.

So I guess what i am saying is that <rk> values are continually calculated - they are then exported to the serps which results in a change in serps and then a couple of weeks later people start noticing the new PR....

Hanu




msg:757588
 3:56 pm on Feb 22, 2006 (gmt 0)

Dayo,

On this date my homepage changed from a <rk> 0 to an <rk> 4 (5 in big daddy) - but it still cant rank for anything - it might just take a little while for these values to kick in/effect the serps though.

Fingers crossed you're not in the sandbox ...

Jim,

I spoke to the guy and asked him on what bases he can back up RK as being the live PR. He believes that the current PR value that is now being exported is old and that it is the RK that is the actual PR that Google uses.

I looked at the source of that "Rocket Sience" LivePR extension and I hate to say that I think it's vodoo. It uses an allinurl: query to get results and then tries to find an exact math of the URL in question among those results. This is bound to fail. Firstly, there is no guarantee that

http://www.example.com/page

is contained in the serps for the query "allinurl:www.example.com/page". The allinurl: results as well as the link: serps have been artificially crippled by Google for several years. Also, the tool doesn't look at the supplemental pages. If the URL in question is part of those, the tool just randomly picks one of the non-supplemental. Obviously, that pick returns the wrong PR.

Secondly, the extension doesn't escape the URL properly.

Thirdly, why would Google return live PR (aka live RK aka internal PR) for allinurl: queries when it returns toolbar PR for all others including info:, site: and so on?

selomelo




msg:757589
 10:19 pm on Feb 22, 2006 (gmt 0)

big daddy DCs almost always has higher <rk> values then the rest.

As far as I can see, this needs some qualification. It is true for "old" sites. But quite the opposite holds true for "new" sites, that is to say, RK values for new sites are always lower in BD DCs, as I pointed out in the other thread on pr update.

Jim Westergren




msg:757590
 10:42 pm on Feb 22, 2006 (gmt 0)

As far as I can see, this needs some qualification. It is true for "old" sites. But quite the opposite holds true for "new" sites, that is to say, RK values for new sites are always lower in BD DCs, as I pointed out in the other thread on pr update.

That is a damm interesting observation. Age more important in the new BD infrastructure?

I will check more things tomorrow.

DerekH




msg:757591
 11:04 pm on Feb 22, 2006 (gmt 0)

As far as I can see, this needs some qualification. It is true for "old" sites. But quite the opposite holds true for "new" sites, that is to say, RK values for new sites are always lower in BD DCs, as I pointed out in the other thread on pr update.

What always amazes me about this particular forum is that when you leave it alone for a few days, a whole new nomeclature seems to spring up. And then when you go back and check that nomenclature out, you find that it's nothing to do with Google and everything to do with hypothesis.
I quite regret being away for a few days, but it's quite odd to come back, check out Google and then come here to see that suddenly RK<normal on BD DCs; it's a whole new language!
So what are we talking about?
DerekH

Jim Westergren




msg:757592
 1:02 am on Feb 23, 2006 (gmt 0)

I found it!

A definition problem!

Let's look at the definition of <rk> from Google.

From their official "Google XML Reference":

"Provides a general rating of the relevance of the search result"

Where does this come from? Seem total wacko, and yes it is.

It comes from the "confidiential" document I wrote about above called: "Google's Search Results Protocols"

And there it says:

Definition of RK:
"Google's rating of how good a single search result is"

But check this:

In that same document it defines what is a "single search result".

And it says:

"R - A single search result - Contains a U; an optional T; an RK; any number of F's; an optional S; and a HAS"

That is the XML!

Every SERP listing in the XML starts with an <R>.

The old definition of R as per that same docuement is:

"A single search result"

The new definition is:

"Provides encapsulation for the details of an individual search result"

So the guy that wrote the new version of this document now called "Google XML Reference", earlier called "Google's Search Results Protocols" translated RK:

From:

"Google's rating of how good a single search result is"

To:

"Provides a general rating of the relevance of the search result"

Which is total wrong, the person didn't see there was a special definition for "single search result".

And this has caused headaches for SEOs ever since ...

A "single search result" is meant to be a listing in the SERP.

Which means that RK is:

"Google's rating of how good a listing in the SERP is"

Which is: PageRank!

Jim Westergren




msg:757593
 1:10 am on Feb 23, 2006 (gmt 0)

RK<normal on BD DCs; it's a whole new language!

RK is the Live internal PR Google is using. BD DCs is BigDaddy Datacenters. We were discussing the differences of RK from BigDaddy and non-Bigdaddy datacenters which is very interesting as it can give more light on what BigDaddy really is.

Nikke




msg:757594
 7:16 am on Feb 23, 2006 (gmt 0)

Hey Dayo,

You can't have missed Bigdaddy, Google's new index infrastructure (as Matt Cutts puts it). That's BD for you and since there is now three Data Centers showing BD results, those three are BD DCs...

Now. As for <RK>, you'll just have to read the thread from post 121 and onwards.

300m




msg:757595
 2:17 pm on Feb 23, 2006 (gmt 0)

I wanted to share something I have found out. Last week, I noticed that withink the xml there was a link from the site i checked that was listed in R N="2" L="2", but it was a non existant link. After doing a lot of research on it, I found that the previous owner from 1999 had created that particular subpage. I decided to create the page and see what would happen and what I found was that all of the BD datacenters removed that dead link from the xml; that was there last week and replaced it with another link that does not exist on the site. Interestingly enough the RK value is the same for R N="2" L="2". Also last week All DC's had me at the normal PR6, but this week I am split between a PR5 and a PR6.

I do not know if this means anything, but I wanted to share what I found.

Oliver Henniges




msg:757596
 6:05 pm on Feb 23, 2006 (gmt 0)

Whats puzzling me the last two days is the results you get inserting toolbarqueries.google.com instead of a specific Datacenter-IP. I also checked pinging to toolbarqueries.google.com some times: firstofall I got different result-IPs every few hours, secondly I got some redirected answers from toolbarqueries.1.google.com, but I lack the experience to interpret what this means.

Finally I most of the time received an RK-value of 5 querying the toolbarqueries, whereas TBPR constantly showed PR4 for my site. So if toolbarqueries.google.com is not the resource, where the heck does the toolbar get its value from? Has this been discussed in another thread before?

Ellio




msg:757597
 6:22 pm on Feb 23, 2006 (gmt 0)

This tool is much more interesting and informative:

[livepr.raketforskning.com...]

Sorry for the query string but it really is relevant to the thread as it seperates TBPR from Live PR.

Dayo_UK




msg:757598
 6:24 pm on Feb 23, 2006 (gmt 0)

The PR8 are the Big Daddy DCs.

Over 50% now by the looks of it.

Dayo_UK




msg:757599
 7:02 pm on Feb 23, 2006 (gmt 0)

>>>>4.We should also assume that the <RK> returned by non-BD DCs is the toolbar PR.

Huh - no I am not making that assumption.

Toolbar PR may have been <rk> values two weeks ago - but <rk> values on the non-bd dcs have moved on since the lastest/current toolbar data export.

>>>>Do the <RK> values depend on the type of query or not?

This question has already been answered to my satisfaction.

>>>>Is the <RK> returned by the BD DC's a live value, i.e. the ominous internal PR, or is just the toolbar PR?

Why seperate BD - the question should be is the <rk> value the ominous internal PR - there are two sets of <rk> values out there non-bd and bd - as Big Daddy has a different crawling structure then this makes sense.

I am not sure when/why you jumped to the conclusion that the <rk> values on the non-bd are the current toolbar PR - or am I misreading you?

So we have Toolbar PR. (Which could be a snapshot of <rk> values at a certain time)
Possible Live PR - eg <rk> values on normal DCs.
Possible Live PR - eg <rk> values on Big Daddy DCs.

To me this thread is perfectly understandable in the way it has developed.

Oliver Henniges




msg:757600
 8:30 pm on Feb 23, 2006 (gmt 0)

Apart from that firefox extension there is a simple test according to which DC is queried by the toolbar. I just opened my router's status protocoll and one single internet-website. With the toolbar installed on IE the router-protocoll showed four http-requests: One for the router-page, one for the website and two for the toolbar-queries on 64.233.183.104.

I tried to make some more tests but my son has just docked in and his ICQ is currently flooding the protocolls;)

I'd support Hanu's interpretation but would like to add another question:

On what basis does google distribute requests from the toolbar and/or queries to toolbarqueries.google.com to the different Datacenters? Does google decide whether a request comes from the toolbar itself (in contrast to a browser), and if yes: how?

Oliver Henniges




msg:757601
 8:43 pm on Feb 23, 2006 (gmt 0)

> or show N/A if that is wrong URL.
what does that mean precisely? If that is not the url requested? In that case I can't support that. I receive a perfect rk-value for my site in the XML-file, while the rocketpage shows N/A.

Hanu




msg:757602
 9:15 pm on Feb 23, 2006 (gmt 0)

Oliver, AFAIK Google uses DNS-based load balancing, aka. round robin DNS. When you query at the SOA ("zone info") for toolbarqueries.google.com using nslookup or a similar tool you will notice that the time-to-live value is set to 1 minute:


C:\>nslookup
Default Server: <snip>
Address: 12.34.56.78

> set type=SOA
> toolbarqueries.google.com
Server: <snip>
Address: 12.34.56.78

Non-authoritative answer:
toolbarqueries.google.com canonical name = toolbarqueries.l.google.com

l.google.com
primary name server = a.l.google.com
responsible mail addr = dns-admin.google.com
serial = 1267475
refresh = 900 (15 mins)
retry = 900 (15 mins)
expire = 1800 (30 mins)
default TTL = 60 (1 min)

> set type=A
> toolbarqueries.google.com
Server: <snip>
Address: 12.34.56.78

Non-authoritative answer:
Name: toolbarqueries.l.google.com
Addresses: 216.239.53.99, 216.239.53.104
Aliases: toolbarqueries.google.com

>

This means that clients may only cache the resulting IP address for 60 seconds and that they have to resolve the name again after that. Every time the client resolves toolbarqueries.google.com, it could get a different IP, i.e. a different datacenter (DC). That way Google can control which DCs are live as well as distribute the load among their DCs. The fact that they use a special name for toolbar queries might indicate that they use a different set of live DCs to serve these requests. What I don't know is whether the Big Daddy DCs are live for either normal or toolbar queries. I just haven't kept track of that.

Oliver Henniges




msg:757603
 10:29 pm on Feb 23, 2006 (gmt 0)

Could it be that the toolbar identifies itself in a special way (HTTP user agent?) so that googles server knows that the request doesn't come from browser and distributes the request to very particular DCs?

I found that sometimes when querying for these rk-values at toolbarqueries.google.com in the past days the xml-file showed my "future page rank", whereas the toolbar constantly sticked to the old value.

Ellio




msg:757604
 11:03 pm on Feb 23, 2006 (gmt 0)

>>>> or show N/A if that is wrong URL.
what does that mean precisely? If that is not the url requested? In that case I can't support that. I receive a perfect rk-value for my site in the XML-file, while the rocketpage shows N/A. <<<<

The only time we have seen the "NA" result on the rocket page is when that DC does not list the page being searched.

ie: the page is not in that DC index for some reason.

We quried this with Rocket page and then found our page had gone missing from the index!

King of all Sales




msg:757605
 11:57 pm on Feb 23, 2006 (gmt 0)

Our homepage shows a tbpr of 5 but n/a in the rocket tool. It is listed correctly in the index when searching site:www.widget.com. All second tier pages show pr4 in the rocket tool.

Jim Westergren




msg:757606
 4:04 am on Feb 24, 2006 (gmt 0)

I just realized the abbreviation for RK.

It is RanK. Why?

The RK values shows up on "&features=Rank", also "Rank".

PR is not a Google official abbreviation, it is something SEOs made up. They have PageRank and use the "Rank" part of the word, simple.

Checking the example above of Cutts blog posts shows that those DCs that has higher RK has a higher "Rank" as well.

Jim Westergren




msg:757607
 4:19 am on Feb 24, 2006 (gmt 0)

I think we should all come to an agreement that LivePR, my tool and the genuine Google Toolbar for Firefox are simply different ways to query the same interface.

The RK values and the TBPR are total different.

Do the <RK> values depend on the type of query or not? For example, could it ever happen that the RK for

You always get the same RK value for an URL, no matter which query you use - just like the TBPR.

Is the <RK> returned by the BD DC's a live value, i.e. the ominous internal PR, or is just the toolbar PR?

As seen in my examples the RK is updating as soon as the DC is caching pages. And it is 100% not the toolbar PR.

If it is not the internal PR, it is something that is very very close to being it.

there are two sets of <rk> values out there non-bd and bd - as Big Daddy has a different crawling structure then this makes sense.

Yep, and it is very interesting.

Our homepage shows a tbpr of 5 but n/a in the rocket tool. It is listed correctly in the index when searching site:www.widget.com. All second tier pages show pr4 in the rocket tool.

If you get N/A, use Hanus tool and search for your URL and see the RK in the XML feed for your URL.

FromRocky




msg:757608
 7:11 am on Feb 24, 2006 (gmt 0)

Our homepage shows a tbpr of 5 but n/a in the rocket tool. It is listed correctly in the index when searching site:www.widget.com.

This is also applied to one of my sites. All DC's in the Rockettool show N/A. It's a three month old site.

jameswatt




msg:757609
 8:43 am on Feb 24, 2006 (gmt 0)

hi

few days back i check my recent optimized site <snip> PR.
The PR was 0 but on the date 21 feb when i checked the Pr of that site i found PR 4 in google tool bar.
so i was quite happy.But today the same story PR shows 0 for my site.....
So i would like to know what is exactly going on...

can any body tell me pls....

[edited by: engine at 10:51 am (utc) on Mar. 6, 2006]
[edit reason] No URLs, thanks [/edit]

Jim Westergren




msg:757610
 10:27 am on Feb 24, 2006 (gmt 0)

Yep, and encoraging from my point of view too :) - However, I dont see a significant difference between the ranks between the 2 - yes we have MC example - but Matt has also said that Big Daddy is not about ranking changes at this stage - just infastructure. Of course new pages may be slightly different.

Yes, true but what I more wanted to demonstrate in my MC examples of new pages were the fact that those DCs with a low RK value are probably those DCs that has not been updated lately = low number of backlinks found = low RK = lower rank.

Also an issue that confused me is this recent PR update seems to have hit all DCs - including the Big Daddy ones - and the serp shift on the 2-4th Feb, hit all DCs - slightly surprised that the PR update that was calculated on the non-bd dcs has effected the BD results.

This is strange for me. Toolbar PR updates should not make a difference in SERPs. But perhaps some export of new data ... like RK values .. hmmm

So perhaps an exported universal PR figure overrides or works in conjunction/is a multiple of the local <rk> figure - this may not apply to new pages.

Interesting theory.

At this moment I am just not 100% sure that the <rk> values are effecting the serp position - this might be the current calculation of the <rk>/pr but is an export required for these figures to take effect in the serps? If you have a TBPR of 0 but an <rk> of 5 can you rank (except when 0 means new)?

Hmm, more testing and checking needs to be done. I will do that later. A TBPR of 0 and <rk> of 5 means for me that GoogleBot has found backlinks since the last TBPR update, otherwise it should not be 5.

My question also is:

Let's assume RK is same as internal PR. Why no decimals on the numbers? Low or high PR 8 is a huge difference.

Jim Westergren




msg:757611
 11:19 am on Feb 24, 2006 (gmt 0)

Ok I see.

I think we need more examples in order to determine anything actually. Like if we have about 5 examples with the same thing we can make a conclusion but I don't think before as there is A LOT of factors in the rankings ...

I am not really tracking SERPs so I can't give any. And it seems like it's only you, me and Hanu following this thread so I am not sure how we will do about this, hehe.

This 182 message thread spans 7 pages: < < 182 ( 1 2 3 4 [5] 6 7 > >
Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved