Executing a site: search, it seems that the results are different in how they are ordered now as compared to how they were ordered in the past.
In the past, there seemed to be some sort of ordering based on "importance" although how "importance" was measured, if it was in fact the case, I have no idea.
Now though, what I am seeing are the results listed almost in directory order, e.g.all / pages listed first, /first-level/ pages listed second, /first-level/second-level/ listed third etc.
The thing is though, viewing the source of the site/page that is the results for shows that is actually the order and hierarchy of the links as they occur in the source so I am not sure if the site: SERPs is ordering the results that way on purpose, by directory level or just following the source order of the navigation.
Anyone else notice the change and/or have any idea as to how the ordering might be performed?
I still see it somewhat in order of importance -- at least the really "unimportant" and "very indirectly linked" pages are far below the directly linked pages. It does somewhat mix pages of similar importance, however, eg:
A (root) -> B,C,D,E,F (1st level links) B (1st level) -> B1,B2,B3,B4 (second level links)
It seems to list: A B (first level: order of the level seems "strange" though) F D C E B1 (second level) B2 B3 B4, etc.
Perhaps the results are "grouped" by rough importance (for a site:- query it could only by PR, right?) and then returned in no particular order within the group? Seeing that a site:-query is not really the most important thing on Google, I imagine this is just a rough estimate that likely just isn't optimized for a real order. Perhaps it signals some strange internal change in Google's index, perhaps it's "just so" :-).
I had the feeling they did is based on URL length, which in an off handed way leads to hierarchy as well, specially if you use folders. Though I've seen it not always be that way, for instance sites that have their home page burried within the site: listing.
> I still see it somewhat in order of importance -- at least the really > "unimportant" and "very indirectly linked" pages are far below the > directly linked pages. It does somewhat mix pages of similar > importance, however, eg:
> A (root) -> B,C,D,E,F (1st level links) > B (1st level) -> B1,B2,B3,B4 (second level links)
> It seems to list: > A > B (first level: order of the level seems "strange" though) > F > D > C > E > B1 (second level) > B2 > B3 > B4, etc.
> Perhaps the results are "grouped" by rough importance (for a site:- > query it could only by PR, right?) and then returned in no particular > order within the group? Seeing that a site:-query is not really the > most important thing on Google, I imagine this is just a rough > estimate that likely just isn't optimized for a real order. Perhaps it > signals some strange internal change in Google's index, perhaps it's > "just so" :-).
> I had the feeling they did is based on URL length, which in an off > handed way leads to hierarchy as well, specially if you use folders. > Though I've seen it not always be that way, for instance sites that > have their home page burried within the site: listing.
> On Aug 6, 10:17 am, JohnMu wrote:
> > I still see it somewhat in order of importance -- at least the really > > "unimportant" and "very indirectly linked" pages are far below the > > directly linked pages. It does somewhat mix pages of similar > > importance, however, eg:
> > A (root) -> B,C,D,E,F (1st level links) > > B (1st level) -> B1,B2,B3,B4 (second level links)
> > It seems to list: > > A > > B (first level: order of the level seems "strange" though) > > F > > D > > C > > E > > B1 (second level) > > B2 > > B3 > > B4, etc.
> > Perhaps the results are "grouped" by rough importance (for a site:- > > query it could only by PR, right?) and then returned in no particular > > order within the group? Seeing that a site:-query is not really the > > most important thing on Google, I imagine this is just a rough > > estimate that likely just isn't optimized for a real order. Perhaps it > > signals some strange internal change in Google's index, perhaps it's > > "just so" :-).
> If it helps... my web pages with next to no content wind up at the > bottom... pages with no page rank end up underneath the "site maps"
That doesn't help at all because it is yet another way things seemed to be ordered different than what I am seeing, or maybe that is the point, there is no point. :-()
My "sitemap" is listed in the first group of pages so does that mean that the rest of my pages are of less value than my sitemap? Oh shit! :-()
Seriously though, I just thought it was interesting because however the pages are ordered in the site: SERPs now, they definitely seem different than before.
I never really paid much attention to it in the past either but seeing a change, got curious.
Actually maybe I should ask, to make sure, are people seeing a change from what it was previously or am I just halucigenating? :-()
> I have never noticed what you guys are talking about... or at least > I've never looked for it...
> On Aug 6, 12:09 pm, JLH wrote:
> > I had the feeling they did is based on URL length, which in an off > > handed way leads to hierarchy as well, specially if you use folders. > > Though I've seen it not always be that way, for instance sites that > > have their home page burried within the site: listing.
> > On Aug 6, 10:17 am, JohnMu wrote:
> > > I still see it somewhat in order of importance -- at least the really > > > "unimportant" and "very indirectly linked" pages are far below the > > > directly linked pages. It does somewhat mix pages of similar > > > importance, however, eg:
> > > It seems to list: > > > A > > > B (first level: order of the level seems "strange" though) > > > F > > > D > > > C > > > E > > > B1 (second level) > > > B2 > > > B3 > > > B4, etc.
> > > Perhaps the results are "grouped" by rough importance (for a site:- > > > query it could only by PR, right?) and then returned in no particular > > > order within the group? Seeing that a site:-query is not really the > > > most important thing on Google, I imagine this is just a rough > > > estimate that likely just isn't optimized for a real order. Perhaps it > > > signals some strange internal change in Google's index, perhaps it's > > > "just so" :-).
Another step by Google to stop the backward engineering of pages/sites from the data displayed in the SERPs if you ask me. It is becoming more and more apparent that Google is trying to reduce the analysis and manipulation of their results...
Anyone else feel that whatever small tolerance of SEO there was over in Mountain View is quickly evaporating?
> Executing a site: search, it seems that the results are different in > how they are ordered now as compared to how they were ordered in the > past.
> In the past, there seemed to be some sort of ordering based on > "importance" although how "importance" was measured, if it was in fact > the case, I have no idea.
> Now though, what I am seeing are the results listed almost in > directory order, e.g.all / pages listed first, /first-level/ pages > listed second, /first-level/second-level/ listed third etc.
> The thing is though, viewing the source of the site/page that is the > results for shows that is actually the order and hierarchy of the > links as they occur in the source so I am not sure if the site: SERPs > is ordering the results that way on purpose, by directory level or > just following the source order of the navigation.
> Anyone else notice the change and/or have any idea as to how the > ordering might be performed?
> Just noticed this. > Another step by Google to stop the backward engineering of pages/sites > from the data displayed in the SERPs if you ask me. It is becoming > more and more apparent that Google is trying to reduce the analysis > and manipulation of their results...
It would of course be consistent for them to list pages in order of the <priority> expressed in a sitemap.
> Anyone else feel that whatever small tolerance of SEO there was over > in Mountain View is quickly evaporating?
I've seriously thought about attempting to reverse engineer their algorithms. Something like a SETI approach - there's a lot of MIPS on web developers' machines.
> It would of course be consistent for them to list pages in order of > the <priority> expressed in a sitemap.
Fair cop - hadn't considered that aspect - has that been officially announced?
> > Anyone else feel that whatever small tolerance of SEO there was over > > in Mountain View is quickly evaporating?
> I've seriously thought about attempting to reverse engineer their > algorithms. Something like a SETI approach - there's a lot of MIPS on > web developers' machines.
Ah g'wan - you know what I mean. ** Replace 'reverse engineer' with 'better understand' **
How much info can (or used) you get from the SERPs that could be classified as useful? The SERPs are quite definitely my primary source of all research on sites.
> > It would of course be consistent for them to list pages in order of > > the <priority> expressed in a sitemap. > Fair cop - hadn't considered that aspect - has that been officially > announced?
I'm not saying they're doing it, but it does make a degree of sense.
> How much info can (or used) you get from the SERPs that could be > classified as useful? The SERPs are quite definitely my primary > source of all research on sites.
I've only just started thinking about it. Lots of variables like search word order.
So the ordering of pages in a site: SERPs are different than previously?
> Anyone else feel that whatever small tolerance of SEO there was over > in Mountain View is quickly evaporating?
I read this mentioned quite often but I've never really seen anything specific that shows Google is negative towards "SEO".
Definitely against Blackhat techniques but against "SEO" in general?
> How much info can (or used) you get from the SERPs that could be > classified as useful?
My favorite tool of late is super long tail searches. Pick entire sentences to see if the page it was copied from actually comes up and where and then start whittling down the amount of words and see what happens. I don't necessarily mean use that to figure out how Google works but more useful, finding out how to make a given page do better overall.
The way the site: SERPs was organized previously never really seemed that useful to me, from an SEO point of view other than the dividing line between Main and Supplemental but if that is truly going away, as every indication that I can see says it is already, then even that ordering is of little use.
I suppose if a given site had numerous pages that were very similar, like product pages or something of the like, ordering by "importance" might be useful for tweaking this and that and observing the results but I don't see that being useful except for a small percentage of sites, even considering retail sites with multiple products.
> Phil Payne > It would of course be consistent for them to list pages in order of > the <priority> expressed in a sitemap.
That's an interesting idea that I hadn't thought of. I'll have to check it out. I almost doubt it though since not everyone submits a sitemap and it would seem something like this would need to work independently of any user supplied information. I'll definitely be looking into it though.
My main question though was are things different now as they seem to me or am I in need of my medication?
If they are definitely different, that does not mean that I am NOT in need of my medication but if they are at least different, I will know I am not ready for the padded room yet.
I don't agree nor understand why Google would apply its own constraint and synthesis when reading a website that is built using a registered publishing software program. i.e dreamweaver, aspdnsf, monster ecommerce and so on. Why should Google put in place obstacles to those websites that are commercialy engineered only to thwart the effort of the software engineer that designed the site in the first place? Google and all other SE's should know when the index page for a given site states that it is authored with a given program that instesad of applying its own set of logic should instead default to the respective webites navigation program/tree structure.
By applying its own criteria it slows the process of indexing new pages for months on end. That a site can take months to index should be cause enough to take action. And still, year after year, they do nothing to improve the performance time it takes to comprehensively index all the pags in a site. Instead leaving the site owner in limbo for the months on end waiting and waiting for their site to finally become the public entity that it was registered as to begin with. Only then for the same process to start all over again should the respective site owner chose to change platforms. In turn Google has no obligation nor respondsibilty for its actions and can never be held acounable for the manner in which it chooses to list, filter, or censor information it list within its page listings for any given web site. It is not fair, but bias, so much as to say it is not even democratic.
> So the ordering of pages in a site: SERPs are different than > previously?
> > Anyone else feel that whatever small tolerance of SEO there was over > > in Mountain View is quickly evaporating?
> I read this mentioned quite often but I've never really seen anything > specific that shows Google is negative towards "SEO".
> Definitely against Blackhat techniques but against "SEO" in general?
> > How much info can (or used) you get from the SERPs that could be > > classified as useful?
> My favorite tool of late is super long tail searches. Pick entire > sentences to see if the page it was copied from actually comes up and > where and then start whittling down the amount of words and see what > happens. I don't necessarily mean use that to figure out how Google > works but more useful, finding out how to make a given page do better > overall.
> The way the site: SERPs was organized previously never really seemed > that useful to me, from an SEO point of view other than the dividing > line between Main and Supplemental but if that is truly going away, as > every indication that I can see says it is already, then even that > ordering is of little use.
> I suppose if a given site had numerous pages that were very similar, > like product pages or something of the like, ordering by "importance" > might be useful for tweaking this and that and observing the results > but I don't see that being useful except for a small percentage of > sites, even considering retail sites with multiple products.
> > Phil Payne > It would of course be consistent for them to list pages in order of > > the <priority> expressed in a sitemap.
> That's an interesting idea that I hadn't thought of. I'll have to > check it out. I almost doubt it though since not everyone submits a > sitemap and it would seem something like this would need to work > independently of any user supplied information. I'll definitely be > looking into it though.
> My main question though was are things different now as they seem to > me or am I in need of my medication?
> If they are definitely different, that does not mean that I am NOT in > need of my medication but if they are at least different, I will know > I am not ready for the padded room yet.
