XML Marklets 0.9
XML Marklets are bookmarklets that turn a web browser into an XML editor, providing a free and easy way to put addtional markup tags on any webpage.
You can use the additional markup tags to point out structures to an application that can process them. For example, suppose you want to extract all the names of people on a webpage. No software can easily recognize which words are human names, but by using suitable markup tools, you can select the names yourself and the marklets will mark which ones you've chosen. Then the page can be fed to software which extracts all the words that have special markup. There is a mock application based on this example.
XML Marklets can be generated for any valid XML DTD. This means that you can use marklets to indicate features common to any standard type of document.
Choose a type of document below and push the display button to see the marklets for that type (these use preexisting DTDs written by others).
Note: This is a work in progress - not a finished product. XML Marklets have been released early in order to get feedback from the XML community and to challenge the notion of what an XML application could be. The current versions (0.9) have some limitations:
- They only work on Internet Explorer 4 (and above) for Windows.
- There is no support for entities or empty elements.
- The marklets do not work on framed webpages.
- Tags are in uppercase only (due to a quirk of IE).
- The marklets may alter presentational HTML (another IE quirk).
- There is no error-handling within the marklets.
- A blank space is inserted before each opening tag.
- Attributes are not quoted (contrary to the XML spec.)
- There is no support for indicating a DOCTYPE.
- The marklets may not generate correctly for some DTDs.
Most, if not all, of these limitations will be overcome in the final version (1.0).
I've written a simple application to demonstrate the type of processing that is possible with XML Marklets - TRC Marklets.
Steve Kangas
XML Marklets: Online since May 24, 1999