|What is standard to insert in the <head> these days?|
Apologies for the general question, but I see so many things in the <head> of a document these days, that I was wondering what you guys put in the <head> and why?
For example, I've seen that the new Wordpress default theme adds this:
<link rel="profile" href="http://gmpg.org/xfn/11" />
And I was wondering why? Then there's the BBC news website which has various things like:
<meta http-equiv="X-UA-Compatible" content="IE=8" />
<meta name="OriginalPublicationDate" content="2010/09/05 13:49:09"/>
<meta name="UKFS_URL" content="/news/"/>
<meta name="Headline" content="INDEX "/>
<meta name="IFS_URL" content="/news/"/>
<meta name="Section" content="Home"/>
<meta name="contentFlavor" content="INDEX"/><meta name="CPS_ID" content="10263779" />
<meta name="CPS_SITE_NAME" content="BBC News" />
<meta name="CPS_SECTION_PATH" content="Front page" />
<meta name="CPS_ASSET_TYPE" content="IDX" />
<meta name="CPS_PLATFORM" content="HighWeb" />
<meta name="CPS_AUDIENCE" content="International" />
<meta name="bbcsearch_noindex" content="atom"/>
<link rel="canonical" href="http://www.bbc.co.uk/news/" />
<link href="http://feeds.bbci.co.uk/news/rss.xml" rel="alternate" type="application/rss+xml" title="BBC News - Home" />
A few of those I understand like the RSS feed and canonical link, but are all those CPS_SITE_NAME etc. meta tags relevant to the Dublin Core? Are they important to include these days?
It would great to hear what you guys think is standard / important to include these days in the <head> of your pages (and if you can be bothered, why as well).
out of all those, i only do the IE compatible and the canonical ones. havent got a clue what the other ones do. maybe they are tailored specifically for the BBC site.
other than that i've just got the standard ones -- charset, robots, expires, description and keywords.
i stick in a <base href> on each page as well
Metacrap [well.com] is still all the rage in big companies ;)
You should maybe find a charset (although it's better to set this as a HTTP header), maybe one for robots, then the usual ones for CSS, JS and alternate versions (RSS feeds). Only
<title></title> is mandatory. The rest is cruft.
Well not *all* cruft . . .
<title> - most important meta info, unique to each page
<meta http-equiv="content-type" content="text/html; charset=UTF-8"> (or whatever you're using)
<meta name="Description" content=""> S.E.'s will use this if it's present, is relevant to the page instead of the first content on the page. Use it wisely.
<link rel="stylesheet" type="text/css" href=""> Try to use as few as possible, reducing server requests.
And only if you use favicons on a given site,
<link rel="icon" href="/favicon.ico" type="image/x-icon">
<link rel="shortcut icon" href="/favicon.ico" type="image/x-icon">
If you don't, it can often reveal your site is hosted on a Plesk managed server, or Bluehost, or some of the other popular ones by the icon in the address bar.
Use **one**, put it at document root, and use / in every document to reference it, unless you want different favicons for each section.
Great stuff! Thanks for all these replies, much appreciated.
this thread may have a few more ideas for you to digest:
Can I remove all my site's meta tags?:
|You should maybe find a charset (although it's better to set this as a HTTP header) |
as the above referenced thread mentions, the charset definition in a meta tag will be required for validation.
even more important, it's useful to have the meta charset definition so that if the document is requested from a proxy server or otherwise obtained from a cache, the charset is defined without requiring the since-forgotten HTTP response headers.