I used to work with a Russian mechanical engineer named Yakov many years ago at MTS. On the my very first day of work Yakov came off the manufacturing floor where they were assembling for the first time a machine he had designed. He rummaged around and pulled a ball-peen hammer and hacksaw out of his filing cabinet. As he stomped back out to the manufacturing floor I head him mutter under his breath:

"It not fit. It not fit? I show him it fit. I make it fit!"

Norm Walsh:

I think I could just about (have, even) accept any one of the items on that list above. Fixing the Unicode problem in XML 1.0 by erratum is stretching the definition of erratum to the breaking point, but by itself is probably an acceptable compromise. Adding pseudo-QName identifiers to the world is confusing and ugly, but by itself probably not the worst thing that could be done. And allowing XML 1.0 documents to undeclare namespace prefixes, by itself, seems sensible in retrospect.

But all three? Really?

Perhaps, dare I say it, it is time to consider XML 2.0 instead.

I make it fit!

Seriously? I'm not sure what's worse, the tinkering-with-ball-peen-hammers going on with XML 1.0 or the specter of XML 2.0. Are you people trying to make JSON look better and better to me every day?

Well, obviously XML solves some problems that JSON doesn't, otherwise attempts to translate ATOM to a JSON format (for example) would have gone much easier. I certainly didn't expect how hard a problem that was going to be.

Posted by roberthahn on 2008-02-20

Of course XML solves different problems than JSON does. Each is tool, each can be applied to a problem with differing amounts of ease, discomfort, etc. For instance, I'd never want to 'markup' my HTML files, my resume, etc, in JSON. To me, that's the simplest line to draw - if you're dealing primarily with text, then 'markup' languages, like XML, are a great fit. On the other hand, if I need to serialize a multi-dimensional array of integers to a text string, JSON is a more natural fit.

BTW, I still don't get the point of JSON-ifying XML. Besides job security.

Posted by Patrick Mueller on 2008-02-20

I don't get it either, but sometimes you have to try stuff like this to discover where and what the problems are.

Posted by roberthahn on 2008-02-20

if you're dealing primarily with text, then 'markup' languages, like XML, are a great fit.

Ooh, you had me nodding right up until the "like XML" assumption.

Posted by Mark on 2008-02-20


Maybe you're right. XML is still the right answer for a lot of the problems I care about. And I don't care a lot about the problems for which JSON is the right answer, I guess.

Like I said, we're in an odd place. The status quo is broken. Tinkering with ball peen hammers is broken. Trying to get community support for XML 2.0 is probably broken.

What flavor of broken do you want today?

Posted by Norm on 2008-02-23