A reader wrote to me wondering:
"Do multiple positive comments on a website produce higher google ranks? It has occurred to me that someone - the same person judging from the similarity of the comments - posts positive comments around your site and I'm wondering if this is an extraordinarily enthusiastic person or what?"
The answer is no. Let's say that a page was about "sudo".
If someone added a comment saying "this article helps me understand sudo", yes, the additional use of "sudo" could help Google recognize that page as being about "sudo". But so would "This is the dumbest explanation of sudo I have ever seen". In neither case would that affect Page Rank very much, in fact probably not at all; see https://www.google.com/corporate/tech.html
Google Page Rank is mostly about who links to you, both in terms of quantity (how many other sites link to your page) and quality (is the site that links to you linked to by other pages). Originally, quantity had more value than it has now; the advent of "link farms" (pointless pages having nothing but links) caused that value to decrease somewhat.. though Google does not tell anyone how they do this, I'd bet their software now has some ability to recognize link farms.
One "trick" that apparently has worked in the past is to register a bunch of domains, and put up near junk content that links back to your main site. That's much harder for Google to identify, of course, but that's where the "quality" does help - those "junk" sites aren't likely to attract any attention from sites that already have strong Page Rank. I would also suspect that it wouldn't be too hard for an algorithm to notice incestuous sites like this.. and it may be that Page Rank already takes that into its consideration.
But all of this is just speculation. Google holds those cards very close to its chest.
Got something to add? Send me email.
More Articles by Tony Lawrence © 2009-11-07 Tony Lawrence
I just invent, then wait until man comes around to needing what I've invented (R. Buckminster Fuller)
Mon May 16 15:10:37 2005: 509 dhart
great article, very informative ;)
Mon May 16 15:38:12 2005: 510 TonyLawrence
By the way, I've added ip address info to the author info here so you can see if this overly enthusiastic person is the same or not.
Mon May 16 15:43:42 2005: 511 BigDumbDinosaur
A totally pointless comment to show that the displayed IP address isn't that of the originating machine. <Grin>
Mon May 16 15:50:30 2005: 513 TonyLawrence
Of course it isn't.. anybody could circumvent this, deliberately or otherwise.
You'll see a different ip for me when I'm out at our summer weekend place.. where I am now. If I make a comment tonight after I get home, it will be from my "normal" Comcast address. But both of these are dhcp and could change.
Mon May 16 17:36:00 2005: 514 dhart
So now everyone knows my IP address and can nmap it? This doesn't seem very desirable... :(
Mon May 16 17:45:10 2005: 515 TonyLawrence
Hmm.. good point - let's take it out.
Tue May 17 22:41:51 2005: 527 Bela
So make it a cryptographic hash of the IP address (SHA-1 or MD5).
Tue May 17 23:36:11 2005: 528 TonyLawrence
Good idea, thanks Bela.
Wed May 18 00:08:55 2005: 529 TonyLawrence
It was easier to use "crypt", so that's what I did.. looks a little odd, but serves the purpose.
Wed May 18 00:12:59 2005: 530 TonyLawrence
Ooops, encrypting an empty string doesn't help :-)
Wed May 18 01:07:59 2005: 532 Bela
`crypt` is weak, it would be very easy for someone to crack this and dump out all the IP addresses. But I suppose if the goal is simply to avoid web page scrapers, this will do. I suggested a modern cryptographic hash because it's much harder to reverse. It provides some degree of continuity between the same users' posts (as long as their in-house or ISP DHCP doesn't renew, or they switch machines...) without actually giving away live IP addresses.
In the end, it makes little difference -- you give away your IP every time you browse anywhere...
Wed May 18 10:02:36 2005: 537 TonyLawrence
If you notice, I pulled a substr of the encrypted string - I did that because I thought the display was "too much", but it accidentally answers Bela's point: you can't decrypt from this because you don't have the full string. I wasn't thinking about crypt being weak, but it probably answers the issue for anyone who is concerned about this.
So now you can (possibly) connect an anonymous comment to other anonymous comments - assuming they are coming from the same ip, of course.
Printer Friendly Version
Do comments affect page rank? Copyright © May 2005 Tony Lawrence
Have you tried Searching this site?
This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more.
Printer Friendly Version