| 8:51 pm on Aug 27, 2004 (gmt 0)|
|... they are all internal- could this cause the problem? |
|... does google consider a dynamic page with different parameter values as different pages? |
Yes, but there are problems with the PR displayed in the toolbar for dynamical pages.
| 9:15 pm on Aug 27, 2004 (gmt 0)|
Google does recognize a page as dynamic if it has variables in the URL. From my experience, it assigns most of the PR to the base URL without parameters and does not like to assign PR to the individual page+variables (reasoning behind is is that you can create thousands of "unique" pages with one page+thousands of variables thereby flooding their index with pages that require less effort to create than static pages).
A good way to see this is that you very rarely get results in the SERPS that are the result of a search engine with variables UNLESS the search engine specifically masks every unique search as an individual static page. You will find lots of results in the SERPS like this.
It is widely accepted that you should shy away from dynamic pages and focus on masking them as being static with no "=" signs, "?" marks, or "&" symbols.
Also, regarding your pagerank question, Google has not updated page rank on the toolbar in almost 3 months now, so I wouldn't rely on it for much information.
Changes are afoot, hopefully.
| 9:59 pm on Aug 27, 2004 (gmt 0)|
|From my experience, it assigns most of the PR to the base URL without parameters and does not like to assign PR to the individual page+variables |
In the PR calculation there is no difference between a page with parameters and without. It just seems that Google doesn't display the correct value in the toolbar for pages with parameters.
| 10:28 pm on Aug 27, 2004 (gmt 0)|
I will correct my statement:
Whether or not the PR is transferred to dynamic pages equally, they are definitely given less weight than static pages. Hence:
will appear in the SERPS higher and more readily than
and in most cases, the latter will not appear at all.
Whether this relates to PR in the toolbar showing up as it does is quite irrelevent, because the real concern is where it shows up in the SERPS, not its supposed pr. So to go back to the original question, NO, google does not entirely consider them seperate pages, because it does not give them equal weight in the SERPS.
| 12:20 am on Aug 28, 2004 (gmt 0)|
Some time ago it was advisable not to have pages "deep linked", you know no page more than one or two clicks away for a visitor, of course this is impossible for large sites, but most webmasters realised a link to home and other important pages was a must on all pages - I could see the benefit of a site map page for human visitors but thought hey, make the index page a site map? anyhow I'm seeing more and more pages doing well with www.mysite.com/about/my/product/id456/dis/today/blah (sometimes .htm/l - php =?)etc
I'm talking very competetive search terms, and yes most have PR0.
I know this is not exactly on topic with the dynamic stuff but I think it may be related.
I know most of the directory file names are phantoms, but gbot gets there and seems to be ranking them well, at least in the stuff I follow.
Point being PR don't matter at sub-page level, indirectly it does matter @index/home page.
I think this can proven with indented results, where a click on the indented result often gives better results for a given search query, even if PR0 and deep routed.
| 11:43 am on Aug 28, 2004 (gmt 0)|
|... they are definitely given less weight than static pages |
That's simply incorrect.
Comparing a static page www.example.com/page.html with a dynamical www.example.com/page.php?topic=xy, the only (small) difference is due to the different URL. Therefore, if the on- and off-page factors are the same and the search term doesn't appear in one of the URLs, the pages are ranked the same.
| 12:21 pm on Aug 28, 2004 (gmt 0)|
doc> I speak from experience with the many dynamic sites I have worked with as well as the results I find in Google. I have very rarely found any dynamic pages ranking higher than static ones (or ones that appear static). There is a reason mod_rewrite exists and is used extensively by anyone who is attempting to optimize their site.
I am not certain where you are getting your information from, but I am certain you are mistaken. Dynamic looking pages suffer in the SERPS compared to static looking ones. This is widely known.
| 1:17 pm on Aug 28, 2004 (gmt 0)|
|There is a reason mod_rewrite exists and is used extensively by anyone who is attempting to optimize their site. |
Of course, there are several reasons to use mod_rewrite. One important one is that some search engines have problems to spider dynamical pages with several parameters. (However, Google is very good in spidering these kind of pages.)
To say it again, as long as the page is spidered, there is no difference (apart from differences in the weight coming from the URL).
|Dynamic looking pages suffer in the SERPS compared to static looking ones. This is widely known. |
Sorry, but I can't agree with that. Has someone proved this?
I also don't believe that this is 'widely known'. Even if, this wouldn't proof this statement.
For example, it seems 'widely known' that the logarithmic base of the toolbar is between 6 and 8. However, this 'knowledge' is incorrect - this value is far away from the real one.
| 2:37 pm on Aug 28, 2004 (gmt 0)|
Another strange thing is that this particular page we have with the pagerank of 0 is always found by google with any variables and is even listed in very good positions. Aren't pagerank and SERP related?
| 4:28 pm on Aug 28, 2004 (gmt 0)|
|Aren't pagerank and SERP related? |
PR and SERPs are related, but PR is just one factor. Also the internal PR (used for the ranking algorithm) might be different from the PR displayed in the toolbar. And - as already mentioned - there seem to be a general problem with the PR displayed for dynamic pages with one or more variables.
As long as the pages rank well I wouldn't care too much about the PR displayed in the toolbar.
| 8:47 pm on Aug 28, 2004 (gmt 0)|
Right now (if you read many threads other than this one) you will see that the SERPS are acting strange with many 0 PR sites showing up very high. Also, you have to remember that the toolbar PR meter has not been updated in almost 3 months.
doc> Nothing about Googlebot has been proven. "Widely known" indicates the statistical probability based on a community of experience, rather than just your own, which is obviously more legitimate.
|(However, Google is very good in spidering these kind of pages.) |
Do not be fooled into thinking that information you read based on one person's opinion is valid, however... do your research and find out what the majority feels and thinks.
The majority of SEO professionals attempt to mask their websites to look like static pages. That is a fact. Take it as you will.
| 9:02 am on Aug 29, 2004 (gmt 0)|
|The majority of SEO professionals attempt to mask their websites to look like static pages. That is a fact. Take it as you will. |
You should better read my statements - I never said something against it: "Of course, there are several reasons to use mod_rewrite... ". The point is that ranking doesn't depend on the kind (static/dynamic) of page. There are (numerous) other reasons the use mod_rewrite.
Also spidering and indexing was not the problem of the original question.
Google was always very clever in spidering pages. About a half year ago there was a further improvement. Now one can find URLs with up to 5 (!) parameters in the SERPs - I call this "very good".
| 8:28 pm on Aug 29, 2004 (gmt 0)|
|Now one can find URLs with up to 5 (!) parameters in the SERPs |
There is absolutely no factuality in this statement, whatsoever. URLs with up to 5 paremeters existed in Google a year ago as well.
I am not certain what the point of all your posts are, since all you seem to claim is that Google is able to spider dynamic pages.
No one said it couldn't. What is being said is Google penalizes dynamic pages for being dynamic in the SERPS, which is widely known and accepted, which is why professionals avoid making it visibly known that their pages are dynamic.
End of story.
| 8:48 pm on Aug 29, 2004 (gmt 0)|
In the past there where microsoft.com pages with many more as 5 parameters.
| 8:51 pm on Aug 29, 2004 (gmt 0)|
"Google penalizes dynamic pages for being dynamic in the SERPS"
Complete and utter nonsense!
| 7:55 am on Aug 30, 2004 (gmt 0)|
|URLs with up to 5 paremeters existed in Google a year ago as well. |
That's what I said: "Google is very good in spidering these kind of pages".
|What is being said is Google penalizes dynamic pages for being dynamic in the SERPS ... |
| 10:25 am on Aug 30, 2004 (gmt 0)|
> does google consider a dynamic page with different parameter values as different pages?
Not so long ago this was an easy question with SEs, most of them took them as separate URLs and penalized the dupe content.
But today I'm not sure about this anymore.
A while ago I had a shop with session IDs. Yahoo indexed the URLs with session IDs (not very clever). Google just discarded the IDs and indexed the site.
I would like to see some prove of the dynamic or not discussion.
| 10:52 pm on Aug 30, 2004 (gmt 0)|
I have a site that has thousands of pages generated dynamically. No problem getting them indexed and no problem getting Google to return them for relevent queries. Furthermore, I changed from a static site to a dynamic site and the pages did not suffer.
Anyone reading this thread who is worried about the effect of dynamic URLs in SEs - don't be. If it makes sense to manage your content that way then go ahead.
| 2:43 am on Aug 31, 2004 (gmt 0)|
Just my 2c on the "dynamic vs static" pages debate.
We have several sites with 5000-10000 pages, mostly dynamic. At one stage we were evaluating whether to recode content management to generate static pages instead of index.cfm?id=123 type pages.
We did some tests and found no difference - Google spidered the pages regardless, and they seemed to rank equally.
We also found that when we created a new product template page, and left the previous product template page, Google spidered both the old and the new for about 3000 products, doubling the amount of pages in the index. Now on some search terms the site gets two entries, one for the old product page and one for the new, despite the fact that the old product pages have virtually no links to them.
One thing to note is that these are simple dynamic URL's - index.cfm?ID=3434 - no additional parameters.
From previous experience you may need higher PR (5+) to get Google to index URL's with 3+ parameters...we have some sites with more complicated URL's that don't seem to get indexed, but I'm not certain of this.
Another factor is that even sites with excellent PR (8+) will not be indexed by Google on URL's with UUID's like this:
Can provide several examples of this...
| 3:34 am on Aug 31, 2004 (gmt 0)|
I have heaps of sites all running dynamic with up to 3 parameters on the querystring.
No problems listing and ranking dynamic pages whatsoever.
The only time you will come across issues is if you use a parameter that looks like a php session id. Spiders avoid these as they get trapped crawling infinite pages.
Each dynamic page is the same as a static page to the search engines.
The reason i have used mod_rewrite and the reason that many people i know use them is a user issue not a search engine issue.
It is a common belief that people will link to a static url more readily than a dynamic url. Over the years I believe i have seen this trend too. Thus mod rewrite is used to promote natural linking.
| 4:09 am on Aug 31, 2004 (gmt 0)|
Interesting finds. A majority of my developments did not get spidered deeply at all until I converted them to look static. Mind you this involves reducing the length of the URL, removing client side session ids and a variety of other things.
All in all, making it look as static as possible did aid in the depth of it being spidered and showing up in the SERPS.
Perhaps my opinion is out of date, but I still choose to make all my sites appear completely static.
| 4:23 am on Aug 31, 2004 (gmt 0)|
"removing client side session ids"
as mentioned, this will cause problems if the session id is appending to the url.
Google sees every different parameter combination as a different URL. Thus if your session is allocated to the client in the form of cookies, which spiders cannot handle, then each page the spider visits will have a new session id appended to the query string. Thus making the site infinite in size.
Google has therefor prevented the spiders from crawling pages with sessions IDs, also if you name a parameter "ID" or have a value that is a guid (random long string) then google might ignore your site.
Make sense? I dont :p
| 5:04 am on Aug 31, 2004 (gmt 0)|
Just to clarify - we have quite a few dynamic sites as well, but the two sites I mentioned were the ones we tested on as case studies. On the larger ones quite long URL's with 4-5 parameters are spidered, but we did notice on smaller sites with <PR4 some longer URL's did not get spidered.
Point re use of session ID's is correct, pages with session id's are not spidered - we use coldfusion not PHP and avoid using session ID's in URL's altogether.
Also agree re static pages attracting more "natural" links. People are definitely less likely to link to a dynamic URL.