A rather more technical post than usual, but if you are used to playing around with HTML tags or fiddling with the innards of systems such as WordPress, this post has some good news that could make your website perform better in search engines…
The multiple URLs problem
It is quite common for a page on a website to be accessible via more than one web address. For example:
http://www.libdems.org.uk/news/new-government-bailout-is-blank-cheque-131421437;show
and
http://www.libdems.org.uk/news/new-government-bailout-is-blank-cheque-131421437 (i.e. without ;show at the end)
both link to the same page.
There are two reasons this might be a problem. First, a search engine may fail to realise that these are the same page and so search results get clogged up with duplicates. Second, some people may link to one version of the URL and other people to the other. Splitting links between these two versions can mean the page performs less well in search engines than if all the links were to just the one.
Search engines are pretty good at trying to deal with this sort of problem, but they aren’t perfect.
The new canonical command
Google, Microsoft and Yahoo have however just teamed up to agree a way of making it easy for people running websites to let search engines know which is the preferred URL for a page so that, even if the search engine follows an alternative version, when it gets to the page it realises what it should be called.
You do this by putting a <link> tag in the <head> section of each page, e.g.
<link rel=”canonical” href=”http://www.williamgladstone.com/shortspeeches” />
This tells search engines that even if they access the page via http://www.williamgladstone.com/shortspeeches
http://www.williamgladstone.com/shortspeeches?sessionid=3sdr5 and
http://www.williamgladstone.com/shortspeeches;show
that all three are the same page and the first URL is the one to be used.
For more information, see Microsoft’s page about the canonical link tag. (Google and Yahoo have pages too, but I think Microsoft’s is better at describing the issue for a relatively non-technical reader and does this without sacrificing necessary technical detail.)
For WordPress users, there is at least one plugin already which helps you make these settings: http://yoast.com/wordpress/canonical/. (Health warning: it’s from a well-regarded source, but I’ve not used it myself yet.)
Follow @LibDemVoice on Twitter
Like us on Facebook
Subscribe to our feed
Sign-up for our daily email digest
London trip cancelled
The misleading mystique of "internal polling" - The Week in Polls
The Firm: Arthur Daley (E's Alright)
Shropshire Council not budging EV charging point locations in Ludlow
My tweets
The price we are paying
Martin Gray
Martin
Jeff
Nonconformistradical
8 Comments
On the subject of improving websites, is there any way of solving the problem of getting out-of-date versions of pages on https://www.libdemvoice.org? I have seen other people mention it, so it’s not just me. Clicking “Refresh” doesn’t seem to cure it, though adding a random query string to the URL does.
Why do you need a plugin? It’s just a line of code using bloginfo() isn’t it?
@anon, pages are cached on our servers, by your ISP and on your computer. This should make the internet faster and stop our servers from collapsing.
If you are seeing an old page, try doing a forced refresh (Ctrl+F5) which should request to all three that you want the latest version of a page.
James – the plugin is a pretty simple one, but it makes it easy for people who aren’t comfortable with writing / changing lines of code.
“If you are seeing an old page, try doing a forced refresh (Ctrl+F5) which should request to all three that you want the latest version of a page.”
Thanks – I’ll try that.
Irony, thy name is Pack:
https://www.libdemvoice.org/canonical-link-tag-11288.html
https://www.libdemvoice.org/?p=11288
@Jennie, I don’t see the irony. The whole point of the plugin/post is to ensure that even if you have 200 incoming routes to a page Google understand which URL they should be using.
@James Graham, had a look a bloginfo() couldn’t find anything which produced the “true” url for the page. Plus the plugin is only one php file which when I took a look at, had quite a few bits I would have missed myself.
Ryan
I’m afraid Ctrl/F5 doesn’t solve the problem.