12 May 2003

The CSS Zen Garden showcases how CSS can be used to implement radically different designs of the same underlying content. I'm a great believer in separating content/structure from presentation and look forward to the day when we can depend on reliable CSS implementations across all of the main browser flavours and versions. We're not there yet. Running the Zen Garden's tranquille design through several different browsers, some render the page with no design at all (plain HTML in a linear layout), others have missing images or navigation shifted off-screen. Delivering unstyled content to down-level browsers is, of course, the only way to go if we're to move web pages forward to take advantage of XHTML, CSS etc. But this demands careful consideration of the structure of the original HTML to make sure that it still works acceptably when the styling is missing (or only partially implemented). We will have to accommodate older browsers for some time to come and it's likely they will begin to see less and less presentation as we move from HTML-based layout (eg tables) to pure CSS. However, there's no excuse for current browser versions to break these CSS-styled pages and we should all continue pushing the browser vendors to support a minimum set of common web standards. [Thanks to Column Two.] 9:33:56 AM    comment []

Lou Rosenfeld writes about backing away from enterprise-wide metadata initiatives: In fact, all enterprise-wide metadata initiatives sound pretty damned ambitious. Strange for me: I'm a librarian by background, but I'm increasingly finding myself advising that such initiatives be delayed or avoided altogether. They're just too difficult and expensive for most enterprises to take on. Two main reasons: metadata interoperability and metadata merging. Metadata interoperability means implementing a common metadata scheme (such as Dublin Core) across various content repositories. Agreeing on a common scheme is usually hard enough - implementing one in a sufficiently consistent and interoperable way can verge on the practically impossible, particularly if there are legacy systems with limited metadata options and existing integration with other external systems. Metadata merging is the next, and even more difficult step: getting the same vocabulary used for synonmous metadata terms to provide a standardised semantics across the enterprise. The result? Step back from metadata nirvana and take a pragmatic approach: My advice is to focus on more realistic areas, like enterprise search, that can provide immediate benefits to helping users find information. (Lou's article was prompted by Tony Byrne's interesting piece on content integration in EContent.) 12:35:00 AM    comment []