At 02:02 PM 7/13/2003, Blaine Cook wrote: >I've recently been doing some work on establishing a search engine for an >organisation I work with, as well as doing a lot of CSS-based layout work. >The intersection of these two activities lent me an idea that I can't find >mentioned anywhere, so I figure here is as good as any a place to start. > >The idea, in summary, is to add a new target medium to CSS. Specifically, >a "robot" type. Unless I'm really misinformed, most robots do not support any sort of CSS to start with. Consider Google's own advice: "Use a text browser such as Lynx to examine your site, because most search engine spiders see your site much as Lynx would." >This would allow web designers to create pages that enable search engine >indexers to focus on the content, enabling the users of search engines to >find better results. One of the biggest problems that I have encountered >implementing a search engine is that navigational aids and periphery >content is over-represented in the results. If you are doing CSS-based layout, use it. Put the main content first. Put the secondary content next. Put the navigation at the end of the code. Read it in Lynx; does it make sense? >A number of CSS-based layouts that use "display: none;", or even hide text >by using the same foreground and background colours, are becoming much >more common. For example, the use of :hover to create popup menus (1) >requires hiding text from users, as does using background-images to >replace text (2). The problem with this development is that for search >indexers, it is impossible (or very difficult) to differentiate between >abusive users hiding links and text to increase their rankings, and these >legitimate uses. Google explicitly recommends against using these >techniques (3). I would suggest that Google probably should start rethinking such absolute stances to some extent. Techniques are advancing with the times. Bill Mason Accessible Internet w3c@accessibleinter.net http://www.accessibleinter.net/Received on Sunday, 13 July 2003 17:46:27 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 11 September 2008 23:28:21 GMT