phpLynx: Unexpected SEO benefit?

Wiki: The second formula above supports the original statement in Page and Brin's paper that "the sum of all PageRanks is one".

ever heard about the difference between toolbar pr aka green bar and live pr?

 

Yes, I know that.

That is a theroretical limit. Now - how much difference does any of that make in the real world? How many web pages (not sites, remember, but individual pages) are there in Google's index? How many are added every day? How many are sharing that PR with a theoretical limit of 1.0? How much do you think any single outgoing link is going to make to the PR of a single page within the index? How many zeros can you type after the decimal point before you reach the maximum character limit for a post on this forum?

The practical real-world implication is as I staed above: The page containing the outgoing links does NOT lose any PR (or to be very specific there is no measurable loss in PR) as a result of passing on PR to the pages at the other end of those links. If it starts as a PR5, it is still a PR 5 with or without the outgoing links.

 

1 or 1 billion is not the issue, the idea is that is pr is passed between page and is considered by the algorithm as a limited resource. I tend to compare this to the Kirchhoff's laws.
If you are familiar with the eigen-decompositions you will figure out that the algorithm would never converge if the pr is not leaked between the pages.

 

But it most certainly IS the issue. I defy you to show me any evidence, however slight, of any measurable impact on the PR of any given page from outgoing links. The page in question would have to contain millions of outgoing links and even then it's dubious.

Again: In the real world, for all practical purposes, PR leakage simply does not exist.

This is also WAY off topic for this thread.

 

Trying to weasel your way out of it in trying to be technical is quite comical, however, what is the basis of what you are talking about. hrmm lets see...

what does 'green bar' represent? oh yea, that's right, PageRank(PR)
what does 'pr' stand for? oh yea, that's right PageRank(PR)
what does 'live pr' exemplify? oh yea, stupid me live PageRank(PR)

Gee, I guess when we are talking about 'green bar' and 'live pr, and PageRank, they are non-relative of one another and especially that thing called PageRank e.g. PR.

Anywho, minstrel, i sent you a PM, to fix some things if you'd like to give it a go. I don't mean nor want to derail your thread, so I'll let things be as they are with ham.

 

minstrel, if the statement of the the creators of the algorithm is not enough then I don't think I can convince you otherwise. 1 or 1 billion is not the issue because this is just a representation, you don't want to handle a lot of zeros after the decimal point? then multiply everything by a billion, ...
I agree this is way offtopc.
to the stupid guy above (anus), pm me if you need to be educated more on this and I promise you I will use language you understand.

 

Clearly, there's no point in attempting to educate you or explain further. You're obviously not even close to getting it.

For the benefit of everyone else, please disregard the irrelevant digression of the whole issue of so-called "PR leakage". It has nothing to do with anything, especially the topic at hand.

 

Forget Kirchoff's Law for now mhamdi, let's go to DP's law and stay with the thread without digressing from the OP's thread.

Thanks.

 

Sometimes I despair about some of the people who inhabit DigitalPoint.

To the person who "anonymously" made the comment "how much you got paid?" (sic), I've already answered that in this thread. Nothing. I paid the phpLynx creators for three licenses. They paid me nothing.

 

I agree with you, sadly it reflects a very poor image on the people who claim to be part of the directory industry', I agree they may well be, but they are the part that is bringing it into disripute. Have you noticed how the thread rating has gone down to 2 now.

How sad are these people, I suspect the little green eyed monster called jealousy has sneaked along for it's a good topic, a pretty good set of answer and of most of the crap on this forum warrented more respect. Or it could have been someone who tried to debate with you but could't back up their comments. (Wherever you go you'll always get a few 'mass-debaters'.)

I can confirm that you haven't ever recieved payment from phpynx, hope that clarifies things. Again it just shows how petty and clearly jealous some people are here to get so desperate to try and throw nonsense like that at you.

Anyway, back on topic, I'm enjoying monitering your site: along with others to look for changes. So far your pretty static with 1630 I believe, but it does depend on what Gserver you get pulled from.

 

I have just modified the robots.txt file for the directories to disallow the cache folders and a few additional admin folders. If I'm correct about the indexed search results pages, this should in due course drop them out of Google.

For other phpLynx users, here's that robots.txt file currently:

User-agent: *
Disallow: /admin/
Disallow: /cache/
Disallow: /cgi-bin/
Disallow: /configs/
Disallow: /images/
Disallow: /includes/
Disallow: /javascript/
Disallow: /newsletter/
Disallow: /paypal_log/
Disallow: /support/
Disallow: /templates/
Disallow: /templates_c/
Disallow: logout.php
Disallow: save_search.php
Disallow: suggest-link.php
Disallow: suggest-category.php

Code (markup):

 

This newbie is now totally confused. I have never done a directory but want to do a very specialized niche and more than just the directory such as event announcements, artists statements etc. etc. Have reg'd the name in com,net,org and us and no trying to decide on a script. Each seems to have strong and weak points.

An important thing for my project is a dark theme which I can tweak. Seems like themes are not so important in the directory world. Theme/template information seems to be sketchy at best.

So much for off topic. I looked at the source of several sites. After reading every word of this thread I find it strange that not a soul mentioned clean and possibly compliant (is late and I didn't check any) html might be part of the reason for what appears to be superior indexing of PHPLynx.

Just a thought.

 

Templates and html have been mentioned Harkster, and it does have something (not a lot) to do with good templating as well.

What good html writing does is allow the bots to go straight to the 'content' without wading through a load of obfuscated html. The real capability is in the php though.

 

There were a couple of typos in that version. Here's the corrected version:

User-agent: *
Disallow: /admin/
Disallow: /cache/
Disallow: /cgi-bin/
Disallow: /configs/
Disallow: /images/
Disallow: /includes/
Disallow: /javascript/
Disallow: /newsletter/
Disallow: /paypal_log/
Disallow: /support/
Disallow: /templates/
Disallow: /templates_c/
Disallow: add_link.php
Disallow: login.php
Disallow: logout.php
Disallow: save_search.php
Disallow: suggest_category.php

Code (markup):

 

Yes. As you can see, since I last posted, I disallowed certain pages in robots.txt but the number of indexed pages is holding well despite that and despite going through and pruning approximately: 60-75 links that had either gone dead or had not maintained the quality I want for the directory:

Results 1 - 10 of about 1,340 from directory.psychlinks.ca

 

I had your results down at 1360 at the beginning of this thread so you've dropped a few which I see as a positive and to still hold up is ever better.

 

did you notice that half of those pages are "refer this link" and "report this link" pages? Is there a real benefit in having them indexed?
site:directory.psychlinks.ca/report/ 385 pages
site:directory.psychlinks.ca/refer/ 374 pages

 

No. That's something else to disallow in robots.txt.

But it again bears out my original observation - Google is indexing everything it can find since i stitched to phpLynx. There is something about this script that is ultra spider friendly. I certainly wasn't worrying about trying to disallow pages with phpLD or eSyndicat - I couldn't even get them all indexed.

Added: Version 2 of robots.txt...

User-agent: *
Disallow: /admin/
Disallow: /cache/
Disallow: /cgi-bin/
Disallow: /configs/
Disallow: /images/
Disallow: /includes/
Disallow: /javascript/
Disallow: /newsletter/
Disallow: /paypal_log/
Disallow: /support/
Disallow: /templates/
Disallow: /templates_c/
Disallow: add_link.php
Disallow: login.php
Disallow: logout.php
Disallow: save_search.php
Disallow: suggest_category.php
Disallow: rate_link.php
Disallow: refer_link.php
Disallow: report_link.php

Code (markup):

 

Nice to see someone so analytical on the results there. Are they good? Are they bad? The Jury is out on that one for me, the fact that they get indexed is good, the fact they can be blocked if you don't want them even better.

Thanks for the observations though mhamdi, maybe I'll get someone over to analyze the results for the script you have a vested interest in, we can compare notes then.

@Minstrel, sorry I never saw your post first, as I think it just about say's it all really, trouble with you psychs, you have this way of putting into a sentence what minions like me put into a thousand words.