| 4:25 am on Aug 24, 2009 (gmt 0)|
Thanks for pointing this out. It's the first I heard of it, and historically Google has not paid any attention to page fragment identifiers.
Wikipedia has a lot of this type of source code on their pages, partly because of their internal linking. It seems clear to me that Google search results will not (and even technically cannot) offer an internal page fragment link unless such an anchor actually exists on the page.
Even then, I'm thinking this type of mini sitelink would still rarely be a page fragment unless there were also external links pointing to it - and probably in a significant number.
[edited by: tedster at 8:43 pm (utc) on Aug. 24, 2009]
| 10:16 am on Aug 24, 2009 (gmt 0)|
I also see less-link intensive (compared to wikipedia) sites displaying these, although they do appear to be restricted to sites with at least some level of authority. I guess its mostly a matter of getting your intrapage navigation sorted out and praying. Much like 'big' sitelinks are a matter of getting your intrasite navigation sorted out and praying.
| 11:44 am on Aug 24, 2009 (gmt 0)|
I believe this was first spotted in March at [webmasterworld.com...]
| 12:02 pm on Aug 24, 2009 (gmt 0)|
rusty, those were intrasite navigation. I'm talking about anchored navigation within a single page.
| 9:37 pm on Aug 24, 2009 (gmt 0)|
Sorry, that does indeed seem new.
| 10:35 pm on Aug 24, 2009 (gmt 0)|
I've added linked anchors throughout one of my sites. Let's see what happens.
| 4:42 pm on Aug 25, 2009 (gmt 0)|
Here's some information from the Google Operating System blog [googlesystem.blogspot.com]:
|Google started to show some special sitelinks for Wikipedia articles... The sitelinks for Wikipedia results are special because Google doesn't send you to a different page, but to a section of the article. This is especially useful for long articles that include a lot of information. |
This particular source bills itself as "unofficial news and tips about Google", so take that into account. But it looks like this new feature is only for long Wikipedia article pages right now. If anyone spots page fragement sitelinks for any place other than Wikipedia, that would be interesting news.
| 4:49 pm on Aug 25, 2009 (gmt 0)|
It doesn't do this for wikipedia alone. I've seen them for another site. I can point you to a query in the Dutch language space for confirmation.
| 5:05 pm on Aug 25, 2009 (gmt 0)|
Thanks johnnie - yes, there definitely are other sites getting this treatment for longer articles with page fragment identifiers.
| 5:13 pm on Aug 25, 2009 (gmt 0)|
To take the discussion back here - I noticed that the page in question is using explicit <a> anchors, even using the depreceted 'name' attribute. I'm wondering if google will also 'see' linked-to anchors that are declared by means of the 'id' attribute on a block-level element (e.g. <h1 id="introduction">).
| 5:54 pm on Aug 25, 2009 (gmt 0)|
My best guess is that Google can handle both the name and id attributes for page fragment identifiers. If I were hoping to get this kind of inline sitelink treatment for a site I worked with, I would use both.
| 6:41 pm on Aug 25, 2009 (gmt 0)|
I wonder if they'll enable sites to limit which in-page anchors may get this treatment via Webmaster Tools, similar to what they do with the more common Sitelinks.
My guess is that they won't, since sitelinks happen site-wide and in-page anchor links could happen in every page within a site -- Webmaster Tools doesn't allow for all that much granular page tuning.
| 7:40 pm on Aug 25, 2009 (gmt 0)|
I did a test on my site and it worked and it was fast. Just do a few anchors on the page and then do external links to the anchors. Mine showed up in two days.
| 8:00 pm on Aug 25, 2009 (gmt 0)|
"external links to the anchors" seems to be the key.
I use fragments extensively in one section of a site, with the anchors pointed to from a table of contents on the page itself and other pages on the same site. Haven't seen this appear on G yet, I'm sure my users would find this very useful.
| 11:29 pm on Aug 25, 2009 (gmt 0)|
This could be useful to users if the pages are legitimate authority sites, and the fragment identifier is to actual content.
Certainly, the W3C is loaded with them and they are very useful. The main hassle is having to bring up the source to find the #?-?-? used. For a site like W3C I would like to see the fragment identifiers noted in the text, so that I can quickly add it to the page URL when making a reference.
| 3:34 am on Aug 26, 2009 (gmt 0)|
I wonder if this is a step toward indexing AJAX content that uses the # to indicate a changed state for the url.
| 4:18 pm on Aug 26, 2009 (gmt 0)|
-- toward indexing AJAX --
That was my first thought, this would be totaly cool. I've been using both Anchors and Ajax for quet a while now.
But wait, now G wants to get into my Kitchen Sink too? ;)
| 6:56 pm on Sep 10, 2009 (gmt 0)|
< moved from another location >
I have seen today a single mini sitelink (or better called, "bookmark link") which is shown in SERPS at the beginning of the "description" of the page (i.e. it is not a separate line below page title like ordinary mini site links). This single "bookmark link" is preceeded with words "Jump to" followed by link, followed by colon, e.g. "Jump to _____:" where _____ link exists as a bookmark on the page.
I have observed this for two different sites, one on Google.com and one on Google.hu. On hungarian google, instead of "Jump to" the wording in hungarian was something like "Go to ______ section:"
The rest of page description is replaced with extract from bookmarked section.
From what I can see from the two pages that got this "Jump to" link:
- The page has bookmark(s)
- There is a link to a bookmark within the same page (in both cases where I observed it, the link to bookmark was on the top of the page)
- under the bookmark was a collection of other links (e.g. similar to directory)
[edited by: tedster at 9:02 pm (utc) on Sep. 10, 2009]
| 12:56 pm on Sep 11, 2009 (gmt 0)|
Fragment Identifiers - Axioms of Web Architecture
2007-03-02 - [webmasterworld.com...]
Well I'll be [puts on best Gomer Pyle accent]. Amazing how Google is now looking at various accessibility elements in its algorithm. Or maybe I should say, amazing how Google is now displaying the results of it's intelligence level.
Protocol states that UAs dereference everything after the hash symbol. I guess that is no longer the case. :)
Uniform Resource Identifier (URI): Generic Syntax
|Fragment identifiers have a special role in information retrieval systems as the primary form of client-side indirect referencing, allowing an author to specifically identify aspects of an existing resource that are only indirectly provided by the resource owner. |
|As such, the fragment identifier is not used in the scheme-specific processing of a URI; instead, the fragment identifier is separated from the rest of the URI prior to a dereference, and thus the identifying information within the fragment itself is dereferenced solely by the user agent, regardless of the URI scheme. |
| 8:44 pm on Sep 11, 2009 (gmt 0)|
Thanks for the observation, aakk9999. This is an even newer development - a page fragment link inline, right at the beginning of the snippet instead of in a row of mini-sitelinks. It looks like this:
|Clickable Title Text Is Here |
Jump to: [Page fragment link] Main article: [General snippet continues here...]
| 6:53 pm on Sep 18, 2009 (gmt 0)|
|If anyone spots page fragment sitelinks for any place other than Wikipedia, that would be interesting news. |
Perhaps old news by now, but worth noting, and the site is a standard enough source that we can mention it specifically....
Internet Movie Database has also regularly been showing page fragment sitelinks on searches for movie actors, directors, writers, producers, etc. People with multi-faceted careers show more identifiers, and there seems to be a certain threshold of activity needed before the identifiers kick in. imdb and Wikipedia are often 1 and 2 in such searches.
| 7:09 pm on Sep 18, 2009 (gmt 0)|
Google - fragment identifiers - and the #2 result (for me) is W3C:
Introduction to HTML 4
Jump to Fragment identifiers: This kind of URI ends with "#" followed by an anchor identifier (called the fragment identifier ). ...
www.w3.org/TR/html4/intro/intro.html - Cached - Similar
The primary link is to the page, but the Jump to "Fragment identifiers" is linked directly to the anchor.
<h3><a name="h-2.1.2">2.1.2</a> <a name="fragment-uri">Fragment identifiers</a></h3>
The <h3> text is not linked, but wrapped with anchors.