Why SEOmoz bot consider these as duplicate pages?

[MattAntonino]

**Canonical for the first link:  **

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html" />

Canonical for the second link:

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html" />

---

You're telling search engines, including the Moz Bot, that the two pages have the exact same content as /score/PatrickCollectionFlPf.html

Now I'll break this down simply.  First link is A, second link is B, canonical link is C.

A=C

B=C

Therefore A=B.

You've told bots that the mp3 tab is the same content (canonical) as the .html page.  You have told bots that the pdf tab is the same content (canonical) as the .html page.  Therefore if they are both duplicates of /score/PatrickCollectionFlPf.html, they are duplicates of each other.

[MattAntonino]

From a human point of view they are different.  But humans don't manage bots, just bot rules.  Bot rules will follow logic and thus the answer I wrote out below is accurate.

IMHO your canonical tags are wrong.  That's the problem.  You have told bots that both pages are "the same" (canonical) to /score/PatrickCollectionFlPf.html  They aren't - they have separate content.  By putting in the wrong canonical tags, you've confused search engines.  Bots follow the rules as stated.  Your rule says they are the same, so search bots treat them the same.

[fablau]

I am sorry Matt, but your statement puzzles me. I have "confused search engines"?Google states:

"A canonical page is the preferred version of a set of pages with highly similar content:"

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=139394

If SEOmoz bot tells me that those two pages are "duplicate" pages, and with the fact both pages belong to the same item, I don't see what's wrong using a canonical tag pointing to the "main" page of the same item.

[fablau]

I don't think with a canonical tag I tell search engines that those page are "identical", I just tell them that those pages can be "consolidated" as belonging to the same item. Or, as Google stated:

"A canonical page is the preferred version of a set of pages with highly similar content"

What's wrong with my canonical definition then??!!

[fablau]

Thank you for your advice, but I am not really a SEO newbie. I begun working on SEO back in 1996 and I have been mentored by Bruce Clay a big deal. I am aware of my website situation and I joined recently these forums trying to improve my SEO knowledge furthermore and to stay up-to-date.

Thank you again.

[fablau]

I am sorry, I have realized now that your are suggesting me that the SEOmoz bot has marked those two pages as duplicate "because of my canonical definition"? Is that what you meant? If so, that puzzles me even more because I don't think a canonical definition shared by two or more pages can "create" two or more duplicate pages by itself! Doesn't make sense, according to my knowledge a canonical tag helps avoiding duplicate issues, not the opposite way around.

[AlanMosley]

You are correct a canonical will take care of it, and using a canonical does not tell the search engine they are identical. It works just like a 301 except for the fact that it does not physically move the users to the canonical page.

But does the search engine take the content from all urls and give the canonical value for al the content, I an not sure it dose, I have never tested it, so I would rather do something with JavaScript or maybe use previous and next tags.

[MattAntonino]

This is circular.

"If SEOmoz bot tells me that those two pages are "duplicate" pages, and with the fact both pages belong to the same item, I don't see what's wrong using a canonical tag pointing to the "main" page of the same item."

Your original question was "I don't personally see how these pages can be considered duplicate since their content is quite different."

You need to make a choice. Either you think they ARE duplicate and you want to use canonicals the way you have, or you do NOT think they are duplicate and your canonicals are wrong.  You can't have it both ways.

---

The** rel="canonical"** attribute should be used only to specify the preferred version of many pages with identical content (although minor differences, such as sort order, are okay).

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=139394

"Should only be used on pages with identical content."

You don't believe this content is identical (thus your original question) so clearly you should not have the canonicals pointing the way they are.

[fablau]

I see your point, but you are still looking and my posted issue here the other way around. My question again then is: the fact SEOmoz bot tells me that those two pages are "identical" can't be because of my canonical definition.  Therefore must be due to:

1. SEOmoz bot sees those pages identical from a SE stand point (and then I shouldn't worry about my canonical definition because the canonical tag should "fix" that problem). But in this case SEOmoz bot should not mark those page as duplicate because of my canonical tag definition.

2 SEOmoz bot sees those pages identical from a UI stand point, which I don't agree on (as a human I see those pages NOT identical). If canonical tags were made for humans, I wouldn't use them if this was the problem (UI duplicate issue). But since canonical tags are made for robots, I shouldn't worry about my canonical definitions if this is the case, specifically if SEOmoz bot marked those pages as duplicate from a UI stand point.

Does this make sense?

[fablau]

I see your point and I agree that maybe a Javascript solution could better help, but the use of rel=prev/next, in my opinion, wouldn't be appropriate. That's more pertinent for multiple page lists/indexes.

[Dr-Pete]

The duplicate content interface can occasionally be confusing in our campaign manager. I think you're reading this wrong, as I look at your account (to be fair to the other people trying to help, they don't have the ability to do that and are doing their best to assist). You have some duplicates due to a navigational issue, I think. For example:

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html?tab=mp3

http://www.virtualsheetmusic.com/score/PatrickCollectionVcPf.html?tab=mp3

These appear to be nearly identical, except for breadcrumb links. I think that's what we're picking up on. They each canonical to their core HTML page (without parameters), but those two pages are different, so the duplicates appear to be true duplicates.

I think your tabs are generally ok, and Google doesn't seem to be indexing the "tab=mp3" vs. "tab=pdf", etc. versions. I'm not sure canonical is completely consistent with Google's intentions (they aren't true duplicates), but it's probably a safe bet.

I won't give any numbers, but your duplicate content error count relative to your total indexed page could is incredibly low, so I think this may just be a fluke of a product that got double-listed or a category that has two paths.

[fablau]

Yes! Got it! You are absolutely right, I read the report in the wrong order! Here is how the reports listed the duplicate pages:

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html?tab=mp3

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html?tab=pdf

http://www.virtualsheetmusic.com/score/PatrickCollectionVcPf.html?tab=mp3

http://www.virtualsheetmusic.com/score/PatrickCollectionVcPf.html?tab=pdf

So, I thought the first couple above was a duplicate, and the second couple the second duplicate, instead here are the right coupled duplicate pages:

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html?tab=mp3

http://www.virtualsheetmusic.com/score/PatrickCollectionVcPf.html?tab=mp3

and the second couple:

http://www.virtualsheetmusic.com/score/PatrickCollectionFlPf.html?tab=pdf

http://www.virtualsheetmusic.com/score/PatrickCollectionVcPf.html?tab=pdf

So, I agree that the SEOmoz duplicate report should be improved graphically to avoid such a kind of confusion.

And that kind of duplicate issue is actually something that I might need to fix on my part... but with the fact that both duplicate pages belong to two different items and have two different canonical definitions may possibly solve the problem by itself... or not? I guess this is one of those rare cases where SEs can actually get confused!

What would you suggest to do with this kind of cross-similar product pages? Those are legitimate pages belonging to two different items that have the same kind of content (i.e. same included music pieces) but written for different instruments! And here is, in fact, another thread where I am discussing about how to handle these kind of similar products found often in the music industry, where the same piece of music can be written for several different instruments causing nearly-duplicate pages:

http://www.seomoz.org/q/canonical-tag-how-to-deal-with-product-variations-in-the-music-industry

Any further thoughts are very welcome.

Thank you again Dr. Meyers!

[Dr-Pete]

We get this confusion often enough that we'll be changing it up a bit in the near future.

If this was a common problem, I'd probably recommend a different structure, with a parent page that splits into arrangements (cello/piano, flute/piano, etc.) and then rel=canonical to the parent product. Practically, though, this looks like a very isolated case on your site affecting maybe a dozen pages out of thousands. I probably wouldn't lose sleep over it, as I doubt it's having much impact either way. I think it's just something to be aware of for down the road, as the site grows.

[fablau]

Thank you so much! I really appreciated your reply which clarified everything for me.

I will follow your advice!

All the best,