| 6:27 am on Feb 17, 2004 (gmt 0)|
I'm wondering about something similar - though only two sites for different purposes on different IP's with *some* difference but I'm concerned that it isn't enough.
How about the code and the page template? Are those the same? And are there elements in the site navigation that are the same, like link text in global navigation and page and/or directory names?
| 7:52 am on Feb 17, 2004 (gmt 0)|
This I believe would be considered duplicate content. Having the same text on all these pages is very bad. If google picks it up they may ban the sites or even the ip address. Are the sites still in google and ranking well?
| 9:15 am on Feb 17, 2004 (gmt 0)|
Last week there was a post in a different forum about a guy with bunch of duplicate contents that got away. Turns out he had published an article and submitted it to bunch of newsletters and sites that got picked up. The body of the sites were all the same, but the headers and footers were different. Search engines use a certain criteria to detect duplicate pages. For example, AltaVista will compare the links of the page and if you have same exact links, then it might consider it duplicate. Some others may compare the content and if a certain percentage is the same then it is considered duplicate. So as long as you keep certain percentage of your content different, then you might get away and this might take you some trial and error to figure out. I would also recommend you to use different titles and Meta Tags. I would also highly recommend not to use the same server and IP's. That is another way duplicate pages can be detetced.
| 10:00 am on Feb 17, 2004 (gmt 0)|
Google has a patent that allows to compare two web pages for similarity by comparing only those parts of the web pages that are related to the search term.
U.S. Patent Office page [patft.uspto.gov]
The Patent [patft.uspto.gov]
allows them to compare documents by assigning a number of fingerprints to a given document.
I am not sure what they already implemented but at least in future you might get in trouble.
[edited by: Marcia at 4:19 am (utc) on Feb. 18, 2004]
[edit reason] Fixed side scroll. [/edit]
| 3:40 pm on Feb 17, 2004 (gmt 0)|
|How about the code and the page template? Are those the same? And are there elements in the site navigation that are the same, like link text in global navigation and page and/or directory names? |
They are all going to be running the same template, but the nav options are all different. the Directory structure is all the same.
|Are the sites still in google and ranking well? |
Currently we are blocking those pages from being indexed via the robots.txt file until we know for sure that it will not adversely affect the parent site, or the affiliates.
|I would also recommend you to use different titles and Meta Tags. |
They all have unique title tags, and we don't use meta tags due to their lack of SEO value.
|I would also highly recommend not to use the same server and IP's. That is another way duplicate pages can be detetced. |
Unfortunately, this is not possible. It is all run off the same backend on the same server.
It should help for you know my purpose of making these syndicated pages available to search engines. Basically, the affiliate sites are not being deep crawled and one of the reasons suggested is due to the lack of PR and backlinks from other sites. So what I want to do is make a section of a few web directory pages where every one of the affiliate sites directly and legitimately link to the other 250 sites hopefully boosting PR. What do you think of this idea, and what would be the results of implementing this idea? Good or bad?
| 5:05 pm on Feb 17, 2004 (gmt 0)|
Basically it's a very bad idea all round. I was aware of another site doing much the same with around 400 different versions of the same pages (with 2000+ different pages, giving 80,000 plus individual pages). All the duplicate pages and the PR they were passing on were completely removed and now no longer show up in the google index at all. From what I can tell google is looking very closely at duplicate content from different pages on the same server, especially if they are linked together. If you do attempt this, I expect your site (or at least the duplicate pages) would be removed from the index fairly quickly.
| 9:16 pm on Feb 17, 2004 (gmt 0)|
The other point with duplication is that if you cannot gain good results with the initial set you also stand the chance of further diluting all your results.
One objective of gianing good PR is to get your site recognised as having authority on a subject. If you dilute that down across several duplicate sites and domains the whole exercise dilutes the authority that you are attempting to gain in the first place.
In days of old we used to split our results across several domains (identical IP or not) - however the G algo soon illustrated that all we were doing was dissolving our own relevance.
We now concentrate on one domain and work it really hard with on topic content. A far better model - less work - very consistant and G loves it.
| 9:53 pm on Feb 17, 2004 (gmt 0)|
What you are saying is true. But we are not trying to raise those pages per se in the SERPs, rather use those few pages as a tools to raise our affiliate sites in general via reciprocal links.