Re: [dita] Some rambling thoughts about the domains attribute

[Bob Thomas <bob.thomas@tagsmiths.com>]

Despite the warnings, I went ahead and read the analysis in its entirety. I am left with the following impressions:

• The domains attribute is a metalanguage that attempts to describe which types and instances of shared vocabulary components are present in a shell
• The metalanguage is an insufficient foundation for fully implementing many of its use cases
• The "build it and they will come" philosophy behind some of the uses cases has yet to happen

My interest in this topic is primarily with respect to constraints. Not only does @domains fail to provide enough information, it can also be ambiguous. Consider this example from spec section 2.6.3.7 DTD: Coding requirements for constraint modules:

@domain value (topic hi-d basic-HighlightingDomain-c) is associated with this content-model fragment: <!ENTITY % HighlightingDomain-c-ph "b | i" >. But, who is to say that in the land of limitless interchange, there isn't some other Information Architect who has decided that the content-model fragment is this: <!ENTITY % HighlightingDomain-c-ph "sub | sup" >. Now, when it comes time to interoperate, the names that were dictated by 2.6.3.7 mean entirely different things. This deficiency means that the current @domain naming conventions for constraints are insufficient. At best, they say that constraints have been applied and things may or may not work. I don't believe that we could come up with a token-based grammar for @domains that would solve these problems. Given that, why have any constraint tokens at all?

Best Regards,

Bob Thomas

Tagsmiths, LLC

On Wed, Jan 4, 2017 at 10:24 AM, Robert D Anderson <robander@us.ibm.com> wrote:

Over time I've come to see less and less benefit + more and more pain from the domains attribute. (Not from domains themselves - repeat, not from domains themselves - just the tokens in that attribute). I think they stand out as one of those DITA things with high level of complexity, little actual benefit, and (outside of attribute domains) few or no repercussions if you mess it up.

I started writing out all the arguments I've heard in favor of the attribute, and why I think most of them are no longer reasonable. Eventually I ended up with a small book. It's way too much for an email thread, so I've posted the resulting thoughts on a blog, and am curious what people think:
http://metadita.org/toolkit/nonononodomains.html

Warning, you may want to fill up on coffee before taking a look.

Regards, Robert D. Anderson DITA-OT lead and Co-editor DITA 1.3 specification, Digital Services Group
E-mail: robander@us.ibm.com Digital Services Group
11501 BURNET RD,, TX, 78758-3400, AUSTIN, USA

--

Bob Thomas

+1 720 201 8260

Skype: bob.thomas.colorado

Instant messaging: Gmail chat (bob.thomas@tagsmiths.com) or Skype

Time zone: Mountain (GMT-7)