"John L. Clark" <john.l.cl...@gmail.com> writes: > It seems that this is an innovation that happened between DocBook 5.1 > and DocBook 5.2, and I am curious to read the discussion around this > development. Does anyone know where I can read about the work that the > TC did in coming up with this enhancement? (Maybe on the mailing list > somewhere? It's a bit hard to search.)
It’s a bit impossible to search right now since OASIS seems to be
relying on MarkMail for this and MarkMail has been shut down!
I found the PR that did it[1], where I expected to find a pointer to the
discussion. (We do try to make the records easy to follow!)
Digging around in my personal email archive from around that time, I
discovered a private email discussion with a user where I suggested
using namespaced attributes and they reported that it didn’t work in
Oxygen.
That’s followed by some email with the editors (that really should have
been CC:’d to this list, apologies) where I said something like “I
thought we already allowed this, did I screw up?” and they all agreed
that they thought this was allowed (they were kind enough not to comment
on whether or not I’d screwed up).
So it was technically added in 5.2CR3, I believe, and appears explicitly
in the release notes for 5.2CR4.
Further searching didn’t reveal anything specific, but I didn’t go that
far back. My guess is that after we allowed elements from any namespace
in the info wrapper, it was a short step to allowing attributes in any
namespace on any element. DocBook doesn’t generally try to be overly
prescriptive.
> Also, I don't see this change
> listed in The Definitive Guide[2], but it seems valuable enough to
> list as one of the major changes. What do you all think about listing
> it there?
Sure[2].
Be seeing you,
norm
[1] https://github.com/docbook/docbook/issues/236
[2] https://github.com/docbook/defguide/issues/107
--
Norm Tovey-Walsh <n...@nwalsh.com>
https://norm.tovey-walsh.com/
> Some disguised deceits counterfeit truth so perfectly that not to be
> taken in by them would be an error of judgment.--La Rochefoucauld
signature.asc
Description: PGP signature
