sXBL initial remarks on language

I'm shortly producing a draft of section 10 of the sXBL
spec (comparison with Mozilla's XBL). Before I get that far,
however, I've done a pass over the existing content in case
there's something I can add.

This email covers picky little things which I hope aren't
contentious. If there's a lack of objection in this forum,
I'll include those changes with my others for review. If
there is objection, let's have it.

If the public forum is a bad place for my remarks, given that
some of the material I'll produce is not yet first draft,
then point me to a better one.

[1] §1. The first section states what XBL is, but not what it
     is good for. A brief remark is apropos.

[2] various minor grammar problems I'll pick up: dangling
     clauses, unpopular split infinitives, circuitous language,
     object-subject confusions, etc. My own grammar problems
     I leave it up to you to spot :-).

[3] §1.2 Not all definitions are collected here. I'll collect
     them here and provide meaning reminders where they appear
     far from this section when first used.

[4] Confusion about conformity of specific syntax possibilities.

     The spec abtrusely defines the idea of badly constructed
     XBL syntax. Is it or isn't it XBL if it's badly constructed
     and non-conforming to the spec? It isn't. I propose that
     the idea of "syntax attempting to be XBL" (or similar words)
     be used whereever such syntax is remarked upon.

[5] Bindings have no active behaviour.

     It is stated in several places that bindings "attach themselves"
     or equivalent active language. Bindings are specifications.
     It is the implementation / UA that attaches them. I could
     parse a binding with an XML parser and no attachment would
     occur.

[6] Overlapping meaning

     The words "applied", "attached", "bind", "load" and "bound"
     are thrown about a bit in the first few sections and are
     a bit confusing. I propose to clean that up a little. I propose
     to reduce the uses of "binding" to a noun alone, and to remove
     all other forms of "bind" except with respect to event names.
     This will save the amount of meaning overload from which
     "bind" and "binding" currently suffer. I have found no
     English language use for the word "bind" in this spec.
     "Attach" seems to hold the most meaning as a verb.

regards, Nigel.

-- 
---------------------------------------------------------------------
Nigel McFarlane                                   nrm@kingtide.com.au
Services:                   Analysis, Programming, Writing, Education	
Expertise:            Software, Telecommunications, Internet, Physics
"Rapid Application Development with Mozilla" / www.nigelmcfarlane.com

Received on Sunday, 17 October 2004 06:04:27 UTC