| 3:50 pm on Oct 27, 2005 (gmt 0)|
SteveB: Sorry, I'm still confused. I thought supplemental pages were those pages are listed under the main result but indented.
I have those kind of indented results, current cache, current page, different from the 'main' result (not duplicate content).
| 8:45 pm on Oct 27, 2005 (gmt 0)|
|maybe there is a separate place that snippets are generated from? |
Some are from DMOZ listings, at least that is the case periodically with our site.
| 10:19 pm on Oct 27, 2005 (gmt 0)|
"I thought supplemental pages were those pages are listed under the main result but indented."
No, those have nothing to do with supplemental results.
| 10:46 pm on Oct 27, 2005 (gmt 0)|
Supplemental Results have those exact words in green lettering just to the left of the "cache" link in the results.
| 10:50 pm on Oct 27, 2005 (gmt 0)|
>> >> maybe there is a separate place that snippets are generated from? << <<
>> Some are from DMOZ listings, at least that is the case periodically with our site. <<
Yes, the ODP descriptions must be stored some place else too, but the example I was quoting was where I was searching (just a few days ago) for words that were removed from a particular web page nearly two years ago. The page that was returned as a matching result, had the snippet showing the words that I was searching for, but the cache was from only three days ago and did not contain any of the searched-for content (because it hasn't existed for nearly 2 years).
| 9:06 am on Oct 28, 2005 (gmt 0)|
After about five days I finally have used the URL removal tool on a supplemental that did stay redirected on the 15% datacenters, but reverted to a supplemental on the 85%. This is the first time a supplemental stayed 301ed on the 15% datacenters. It may still have reverted but I wanted to get all my problem pages in the same circumstance in preparation for Jagger3, so that whether they get fixed or not for the long run they will all have a similar status.
| 12:17 am on Oct 29, 2005 (gmt 0)|
Have others found setting up a Google site map is helpful in getting rid of supplemental listings on pages that have been removed from your sites?
It sure would be great to find a solution to this.
| 7:25 pm on Oct 30, 2005 (gmt 0)|
Last pre-Jagger3 Supplemental post for posterity...
Just had my first page that was only Supplemental on the 15% datacenters have its 301 kick in on the 85% datacenters, which you would expect since it was merely a "normal" page there, but it reverted to being a Supplemental on the 15% datacenters. Since 301s seemed to work on the 15% for supplementals listed on 100% of the datacenters, I hoped this one that was only on the 15% would work, but it didn't.
So, bottom line, no 301 redirect of a Supplemental has been obeyed on 100% of Google's datacenters, even though in all cases there was a current (non-supplemental) page regularly crawled on that URL. (It sounds so confusing to even type...)
| 7:36 pm on Oct 30, 2005 (gmt 0)|
Yep. I see supplemental pages everywhere, and have never managed to get Google to update them, delist them, or forget they ever existed.
I am hoping that they will clear many of them away later in the week, but I suspect that all that will happen is that certain classes of supplemental page will simply be hidden from view in the SERPs rather than fixing the underlying problem.
It is very easy to create a supplemental result. Have a page that can be accessed through 2 URLs. Get both of them indexed, and wait for them both to be cached (and note that one URL is cached more often than the other) then change the content very slightly. If you are quick enough Google will not see them as being duplicates any more, as their duplicate checking mechanism seems to compare cache copies, or one cache copy with one live copy (and often gets it wrong). Google will then continue to present the "old" copy as a supplemental result seemingly for ever more, for those search terms that were on the page back then, and you can now change the page content to whatever you want and the other URL will rank for any search terms that are now on the live page too.
If you weren't quick enough, then don't worry. If the duplicate page checker did get the other page and hide it from the SERPs with an old cache, then after just a few weeks more it will usually reappear anyway. It does this because the page as seen through the other URL (the one that still shows in the SERPs) now has different content. The cache of the old page will come back as a supplemental result, simply because the old cache doesn't have the same content as the live page located at the other URL, so it will no longer be filtered out as a duplicate. It matters not that for the cached page with the old cache, that the live content is also different: once the page is marked as supplemental, the old cache is no longer compared to what is really on the live page at the same URL.
(Yes, this also sounds so confusing to type...)
| 8:26 pm on Oct 30, 2005 (gmt 0)|
My apologies. I was completely mistaken. Thank you for the kind clarification. I truly appreciate the professionalism shown on this forum - especially towards us neophytes.
| 7:20 pm on Nov 4, 2005 (gmt 0)|
G1 and Steve please help me understand supplementals - the Google webmaster guideline description is very vague.
Are the following all true?
A page is either supplemental or not - the query does not matter.
Supplemental pages effectively won't rank for any but obscure searches.
Pages can move from supplemental to regular listings but this is rare and nobody understands how it happens.
| 7:48 pm on Nov 4, 2005 (gmt 0)|
>> A page is either supplemental or not - the query does not matter. <<
The query does matter. For a page that still exists, but the content has been changed, the page may be supplemental for one search query, often for very old page content, while also being found for the new content of the page as a normal listing. The "new" query will show a modern snippet and a modern cache. The "old" query will usually show an "old" snippet reflecting the "old" content, but the cache that it links to may be a cache from 18 months ago, or it may be a cache of the "new" content - I haven't yet worked out why one or the other may be shown. I think it just depends on whatever data was imported to Googles database at the time.
>> Supplemental pages effectively won't rank for any but obscure searches. <<
They will usually rank for content that used to be on the page, as the supplemental is usually one where the supplemental result represents an old version of the page. Some supplemental results are, however, just left over remnants of sites where the whole site is 404, or even where the domain no longer exists at all.
The other class of supplemental result is where the page was originally a duplicate page, and was previously filtered out, but which has re-appeared when the page was modified and the old cache copy for "URL.A" is no longer a duplicate of the real page at the canonical URL, "URL.B". So "URL.A" reappears in the index to bug you for all time. Many of these are on sites where "URL.A" represents a non-www URL that has been redirected to www for the last 6 months, but where Google just re-imported old data back into their database in July or August without doing a reality check on it.
>> Pages can move from supplemental to regular listings but this is rare and nobody understands how it happens. <<
It is almost impossible to update a supplemental result to a regular listing. If it is supplemental when seaching for "red widget" and the page content has been changed to "blue gadget" then the "blue gadget" search may show the page as a regular result, but the page will continue to be returned as a supplemental result when searching for "red widget". The cache for the "old" search may show the old cache from 18 months ago, or may link to a cache only days old - showing the "blue gadget" content.
Google needs to re-evaluate the data that it has. There is a lot of junk in the supplemental results; page content from yesteryear. Google should link to the current vesion of a page, and try to forget what the old version contained.
They should leave that job to archive.org who do it in a much better way.
| 8:06 pm on Nov 4, 2005 (gmt 0)|
"A page is either supplemental or not - the query does not matter."
What g1smd said. A URL can be both a Supplemental and a non-supplemental, depending on the query.
| 8:46 pm on Nov 4, 2005 (gmt 0)|
Thanks g1 - really appreciate that excellent detail!
I'm still digesting the implications for our site, which recovered about 10% of our former G traffic during this update for reasons I'm trying to understand but seem to be related to supplemental changes.
If I query "site:mysite.com oregon" I get regular results, but "site:msysite.com montana" It's all supplementals.
Also specific Oregon queries do fine but similar queries for other states fail, even though we have equally thorough content.
| 8:55 pm on Nov 4, 2005 (gmt 0)|
Can anyone give me any insight regarding pages that have gone supplemental that have never been changed and are not the result of duplicate content? We have a few regular pages from a site this has happened to, and in addition we have loads of pages from this same site that came from a discussion board.
Why did they go supplemental, can we reverse it, and what are the ramifications of the supplemental pages for the site?
Any help would be greatly appreciated; we have not been as well versed in this area as we should be.
| 9:02 pm on Nov 4, 2005 (gmt 0)|
Pages can go Supplemental because they are lightly linked, meaning they haven't been crawled in a long time. This appears to have been the original way Supplementals appeared, but now it's almost as if only a trivial amount of supplementals are this "classic" type.
There is no way to get rid of a Supplemental, currently. Your best bet is to tweak the content of any Supplemental, and get it crawled. You would have to believe that eventually Google will get rid of the Supplemental database as it exists now.
| 9:18 pm on Nov 4, 2005 (gmt 0)|
I have found a way to get supplemental pages reindexed, but it creates a duplicate entry.
However, it has also seen the supplemental result get dropped from the index in a matter of only a week or so.
I have been testing this for several months; it has worked nearly every time (for pages where the page is still online).
The original URL is a static page at www.domain.com/folder/page.html and I linked to it using www.domain.com/folder/page.html?-index-me- instead.
That new URL appeared in the index within days and ranked OK. Within a few weeks the basic URL listed as a supplemental result had dropped out.
[edited by: g1smd at 9:37 pm (utc) on Nov. 4, 2005]
| 9:18 pm on Nov 4, 2005 (gmt 0)|
Thank you, much appreciated.
Do the supplemental pages have any negative effect on the site? Is there a down side to them, (other than I assume the pages themselves now longer rank in the regular index)
| 9:51 pm on Nov 4, 2005 (gmt 0)|
g1 - you linked to the desired new page using a clickable <a href =, right? Do you think this could be done with a 301 redirect?
edited for clarity
| 10:00 pm on Nov 4, 2005 (gmt 0)|
I did it with a link from another site: the link was placed on a PR7 page too.
It was on a disposable domain; just messing about to see if there was any way to dislodge supplemental results in any way at all -- and there are very few.
You need a real link for this. Google indexes the "new" URL. Try it for a couple of pages and see what you get.
There are no guarantees; use at your own risk.
| 11:51 pm on Nov 4, 2005 (gmt 0)|
Ha - these days I think GOOGLE should post that at the home page!
Using Search can be hazardous to your health. Use Google only in well ventilated suicide-proof well stocked saloon environments
| 6:35 pm on Nov 7, 2005 (gmt 0)|
Ok, say I have three pages:
1) mysite.ext/mypage.html (crawled daily)
2) mysite.ext/Mypage.html (supplemental) (old cache date)
3) mysite.ext/MyPage.html (supplemental) (old cache date)
They are all the same content, and I suspect that page 1, the correct page, is being penalized for duplicate content. If I put in a 301 redirect from page 2 and page 3, to page 1, then place links somewhere on a regularly crawled page to page 2 and page 3 (so that they get re-crawled?), would that have no effect on the supplementals indexed?
| 6:46 pm on Nov 7, 2005 (gmt 0)|
After a little more thought..I guess my more fundamental concern is eliminating the duplicate content penalty. I don't care about the supplementals being there, necessarily.
So, if I just put different content (or no content) on the supplemental pages, and then link to them from a daily crawled page, would they then get re-indexed and eliminate the possibility of dup content penalty on page 1?
| 7:23 pm on Nov 7, 2005 (gmt 0)|
Once the page is supplemental, the content on it now will live (at this point) forever in Google's memory.
Put new content on the pages you want to get rid of, let them get crawled for a period of time (week, months, decades...) then 301 to somewhere. This won't accomplish anything right away, but someday Google should get its act together and start obeying the 301s.
| 8:27 pm on Nov 7, 2005 (gmt 0)|
Related discussion at: [webmasterworld.com...]
See post #400.
| 12:00 pm on Nov 10, 2005 (gmt 0)|
I have alsoc the problem that i have for almost a year hundreds
of thousands Supplementals in the Google index and get these not
Because of this I have used the Google Removal Tool and I deleted all
these old URLs with "noindex, noarchive".
All these URLs came again after 180days.
I do see the possibility to delete this URLs not with metatags,but with or 404 410.
Is this method better or do the URLs also come back definitely after 180days?
Do you think that deleting with the Google removal tool has any effects?
Do deleted ("hided") URLs actually play no more role for calculations like duplicate
content during the 180 days?
Thank you for your help!
| 12:42 pm on Nov 10, 2005 (gmt 0)|
Google Remove does not remove pages, as you have found out. It just hides them for 90 or 180 days.
For a page that is 404, or even for a domain that no longer exists, you can "remove" the pages from the index, but, as you say, even if the pages no longer exist at the end of the 90 or 180 days then they are automatically added back in. That seems to me to be wrong. If the pages still no longer exist, then they should be permanently removed.
Even where Google might discard the old cache, they still keep a record of the old title and the old snippets for ever. For pages that are supplemental results there is no way to get Google to forget about those pages.
| 2:10 pm on Nov 10, 2005 (gmt 0)|
Yes, but my question was:
Do deleted ("hided") URLs actually play no more role for calculations like duplicate
content during the 90/180 days?
| 2:17 pm on Nov 10, 2005 (gmt 0)|
Only Google knows the answer to that... and they aren't telling.
| 6:57 pm on Nov 11, 2005 (gmt 0)|
Yes, but perhaps has somebody already concluded observations regarding this?
Question: Does anybody have UELS removed with the removal tool and after this the duplicate content filter has gone?
| 11:06 pm on Nov 12, 2005 (gmt 0)|
I've done a lot of URL removing in an effort to solve ranking problems and it has not seemed to have any effect.
However I suspect that it could matter in many cases and I wonder if part of Google's current problem of killing good pages stems from conflicts between the gigantic and unmanageable supplemental index and the regular index.
| This 77 message thread spans 3 pages: < < 77 ( 1  3 ) > > |