|
. This a fascinating discussion about what is fundamental to XML, but I find that most of the discussion focuses on whether or not something is useful without really stating for what it might be useful. . So what is really trying to be accomplished here ? 1. The first objective seems to be to “trim the weeds” of an overly complex specification. This might be very useful if it were the basis for continued efforts, but that seems unlikely. . From James Clarks’s “RandomThoughts” the main advantages here are ease of support in parsers and tools, ease of use – especially for new users, and the avoiding of explicit (and implicit) use of subsets that are being developed anyway. 2. It seems clear that this not meant as a markup language. 3. MicroXML can be useful for messages. 4. MicroXML can possibly be useful as a data store for various specifications. 5. Other ? . Fundamental simplicity is achievable by eliminating attributes (with greater potential for simplifying schema), simplifying relative URI’s, and severely sub-setting the characters sets allowed (except where needed for URI’s). This simplicity would also greatly simplify schema validation. But I suspect the simplicity route is already lost to JSON and to the myriad tools that generate and edit XML for neophytes. Also, such radical surgery raises issues of what can still be called XML. . For items 3 and 4 some linking capability for specifications is very useful, and this is provided in the xmlns attribute. . The term “affordances” is interesting but I can’t seem to figure out what it really means, other than a set of implied attributes. I suspect that it is meant to imply support for actions for an application to take, but I’m not sure what would be generally useful here other than “include” (which is implicit in xmlns) or to include some verification rules or schema. Actions do imply a higher application level, but what is proposed seems to enable such a layer rather than to provide it. . I believe that, unfortunately, the real complexities of XML have sprung up from the weed garden with other specifications, tools and applications that have introduced far more untamed growth. If one really wanted to make XML based standards a lot simpler, dealing with that would of course require far more comprehensive efforts. Given the accelerating rate of proliferation of diverse, inconsistent and ever more specialized technologies and tools, this may become worthwhile as a comprehensive foundation for application development. Bill Clare P. S. Sorry that AOL doesn't handle replies to this list correctly.
In a message dated 7/5/2012 9:32:03 A.M. Eastern Daylight Time,
Len.Bullard@ses-i.com writes:
Thanks John. It doesn't explain why these ideas died with a whimper |