Re: Pre-agenda questions for ixml 26 April meeting

On Monday 25 April 2022 08:39:30 (+02:00), Norm Tovey-Walsh wrote:

> Hi Steven,
> I don’t think you’ve had a chance to reply to any of the recent messages
> on the mailing list. 
I've been away all week, and leave again tomorrow, so I haven't been online much.

> In preparation for tomorrow’s meeting, are you
> content to agree that we have consensus on these issues?
> * Changing ~ to !

No. I would be willing to accept adding ! for those people who are unable to accept that different languages have different syntaxes, but ! doesn't mean 'not' to me, and it isn't used as 'not' in ixml either, so I see it as misleading.

> * Using = and | exclusively

No. I added = and | as a sop to those people who as above  are unable to accept that different languages have different syntaxes, but it isn't an equality relation, and just as many other languages use : for definition as other characters.

> * A version declaration

No problem.

> * Using + for insertions

I missed the argument for why ^ was problematic. I need to read up. I don't have a great problem with it, though I do still feel that ^ was a good choice. + is already used in two different ways in the spec, and this is a completely different use. On the other hand ^ is closely related to how it is used now, as the text now reflects.

> * Introducing error codes parenthetically (“Option 3”[1])

It does spoil the readability of the spec. I don't see why users should be exposed to these details: it is just not interesting. No objection to having error codes, but textually it is really ugly, and non functional except to a tiny group of people.

> There was some discussion about the simple namespaces proposal in email
> and in the issue[2] commments. Where do you think we stand on that and
> do you have any suggestions about what features of the proposal are most
> in need of further discussion?

If a person used my suggested version, it would still need to work, right? In which case, I don't see the added value of a second version of doing the same thing.
If it wouldn't do the same thing, then why not?

Sorry to be difficult, but I thought this group was going to fix up the last bits of the spec, and quickly publish. Now we are changing all sorts of things.


> Assuming you’ve addressed issues #25 and #26, and accepting consensus on
> all of the issues above, I think the simple namespaces proposal would be
> the only remaining technical change proposal to resolve.
> Be seeing you,
> norm
> [1]
> [2]
> --
> Norm Tovey-Walsh
> Saxonica

Received on Tuesday, 26 April 2022 13:57:00 UTC