hey Rogerio, I'm glad for you I hope this may lead you to great success!

The robots file will avoid google to show further information on the disallowed pages but it doesn't prevent indexation.

They're still indexed (that's why you're seeing them) but with no meta desc nor text taken from the page because google wasn't allowed to retrieve more information.

If you want them to start showing info, you'll jsut need to remove that rule from the robots.txt and soon you'll start seeing those pages information showing, but if you want them out of the index you can use GWT to remove them from the index after you've included in each page the noindex meta tag which is the only command which will prevent indexation.

Yes the last one is the way I'm actually doing this. In that way if someone republish the image you'll be sure that your links will be maintained if those links are useful for the fruition of your work.

What I didn't understand is that if the links are useful to use the infographic or if you want to be sure that people republishing your infographic will be giving you the links you want.

In that last case you want be able to achieve that easily because people may save the image and republish it however they want. But normally if someone finds your infographic useful they will put a link to the site where they found it (it's quite ethical) and that's what you want because links hidden in the infographic won't have this "editorial" value which google is looking for, said in other words is better to have a link editorially done than have plenty of them embedded in the infographic. At least IMO.

Well Google can index the iframe but not as Rogerio would like to.

Google can see the src link pointing to the infographic page sowhen following that google will index the content contained in the iframe but don't count with such kind of links as being valuable for you.

What I don't understand is that you want to use the iframe instead of the image because you got links in it. Why not brake the image and build it as a table and link each portion of it to the url you want? Use flash and embed the links? Or map the image to indicate zones where the links should be? I think there are more frendlier ways to put links without recurring to the iframe to have links in your infographic.

Or simply you can put a tinier version of that infographic and if someone wants to see full image they need to reach your site where they'll find all the links. Hope this may help!

Sorry Fabrizio I got mad with my old answer

that canonical doesn't make sense with a noindex, with noindex follow.you're completely fine.

Summing up I think that you have many parameters so you should try to write them down and define the role of each one.

Then add them in GWT and choose there which are the ones which doesn't add any value and which you want to "block" (instead of putting a noindex).

The valuable ones (the one which adds value and changes content) should contain the self canonical and paginated next/prev. If you can get rid of unesful parameters it could be better so to have cleaner and shorter urls.

Just be sure that you're mainly using the most important parameters so you're consistent with your strategy.

Hope this will clear your doubts, it was a nice chat!

Hey Fabrizio, I think that what Google states in their guidelines is that you have two choices:

• if you have a view all page, you should noindex and follow all your other pages so google will deliver only that page
• if you don't have a view all page or if you prefer to show paginated series (i.e. to make pages lighter and quicker to deliver to users) you may consider to use rel next/prev.

In this second case it may happen that you also add filters or session ids in the urls of those pages, in that case you should consider adding a self referentail canonical tag to avoid duplicates. But this is only if you cover this case, if you're looking to canonicalize correctly your paginated series you may not use the self canonical tag, because if not properly implemented this may get you a bit of extra work.

In this page for example

I found this:

http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=3&lpg=0">

Which  I don't think is what you want to do.

Also if you set the page to view as a table: your url changes to http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=3&viewlistflag=1

and while the canonical should remain the same (well done but I think you should get rid of the lpg parameter in the canonical), the rel next prev should change accordingly IMO.

So instead of being:

prev: http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=2&lpg=20
next: http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=4&lpg=60

you should offer the next and prev page of the filtered url:

next: http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=4&lpg=60&viewlistflag=1
prev: http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=2&lpg=20&viewlistflag=1

Or in this case (since the content is almost the same you may consider the list page as the canonical of the table one putting there a noindex.

Summing up, IMO: in this page http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=3

you'll have:

prev: http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=4&lpg=60
next: http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=2&lpg=20
(optional) a self canonical to http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=3

In this page (and in other filtered pages if you have apply the same idea):
http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=3&viewlistflag=1

You'll have:

noindex,follow and canonical to the list page:
http://www.virtualsheetmusic.com/downloads/Indici/Guitar.html?cp=3

maybe dr peter can correct me if I'm wrong but I think this should be more consistent like this. Sorry for the huge answer