On problems with the prolog

Hello,

Back in March, there was some back and forth about the prolog. I think two different issues got tangled together in that exchange:

1. The syntax of the prolog
2. The semantics of the version identifier

I think we’ve resolved the questions about the version number. Perhaps that will simplify discussions of the syntax.

There seemed to be two objections to the syntax:

1. It’s too hard to parse.
2. It looks too much like a rule.

I haven’t seen any persuasive arguments about the difficulty of parsing the prolog[1]. It parses just fine with the rules we already have in the grammar.

Yes, I suppose you could do this:

  ixml { the complete text of Ulysses goes here } version "1.0" .

Okay. No one is going to, but nevermind that. If you use the Project Gutenberg text, you’ll have to parse a ~1.6Mb comment. That comment is going to be a problem anywhere you find it. I don’t see why it’s relevant that it appears in the prolog.

I agree that it looks a bit like a rule. But I’m not sure why that’s bad.

Yes, we could do something to make it syntactically distinct from a rule. I think putting parenthesis around it is a popular option:

  (ixml version "1.0" .)

The problem is, I think we should extend the prolog to allow arbitrary machine and human readable metadata:

  ixml version "1.1" .
  author  givenname: "Norman", surname: "Tovey-Walsh" .
  license id: "MIT", name: "The MIT License",
          uri: "https://opensource.org/license/mit" .
  date    "2024-03-12" .
  release "1.0" .

And I don’t think this is an improvement at all:

 (ixml version "1.1" .
  author  givenname: "Norman", surname: "Tovey-Walsh" .
  license id: "MIT", name: "The MIT License",
          uri: "https://opensource.org/license/mit" .
  date    "2024-03-12" .
  release "1.0" .)

Is there a technical problem that needs to be solved here?

                                        Be seeing you,
                                          norm

[1] https://github.com/invisibleXML/ixml/issues/235#issuecomment-1978366604

--
Norm Tovey-Walsh
Saxonica