OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

dita message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [dita] Scenario for cross-deliverable referencing

Hi Richard,

If the keys are not unique, we will need a key scoping mechanism. In my original note I called that out as a requirement.  If the keys are unique, no scoping mechanism is required.

The issue of key scoping is separate from the issue of cross-deliverable references. You can have issues with duplicate keys in a single deliverable as well. Whatever solution we come up with for key scoping will have to cope with both separate deliverable and combined deliverable use cases across the same content source. That's why I provided an example that showed both.

Eliot wants to have scoping by map - I have concerns with this as an approach, like the need to require source map awareness in the context of assembled map key spaces (like A+B getting built together into a single PDF); but maybe these could be figured out. We'd also want to figure out how things behave when conref'd, etc. All this could be done as part of the proposal for handling key scoping - maybe map-based scoping will turn out to be the best way to do it after all.

But map-based scoping is:
 1) not a requirement for cross-deliverable referencing (any more than it is a requirement for building single deliverables that include multiple maps)
2) only one of several possible ways that keys could be scoped, which I would like to leave to the other proposal to work out.

My main concern is that we separate the issues: scoping by map does not solve the issue of cross-deliverable addressing, it only solves the issue of duplicate key resolution (and it creates new issues for key resolution in the context of same-deliverable addressing).

Michael Priestley, Senior Technical Staff Member (STSM)
Lead IBM DITA Architect

From: Richard Hamilton <hamilton@xmlpress.net>
To: Michael Priestley/Toronto/IBM@IBMCA
Cc: Eliot Kimber <ekimber@reallysi.com>, dita <dita@lists.oasis-open.org>
Date: 09/06/2011 05:01 PM
Subject: Re: [dita] Scenario for cross-deliverable referencing


On Sep 6, 2011, at 8:11 AM, Michael Priestley wrote:

> Hi Eliot,
> I believed your scenario was covered in my case of multiple chapters being produced as separate PDFs.
> In that scenario, the author of Map A does not define a key that references a key defined in map B - they simply include map B's definition of map B's keys, after it has gone through processing to produce a deliverable-specific version of that key definition.
I'm wading in deep water here:-) and may very well have missed a critical point, but it seems like this is a markup issue that isn't resolve in the general case by simply merging the two (or more) sets of key definitions (which I think is what you're saying above).

If the keys are unique across all of the maps, then yes, you could simply include map B's definitions in map A and use them directly. But, if they are not unique, then don't you still need markup that will allow you to uniquely identify the key you want? In which case, you're back to the problem Eliot stated originally (at least as I understand his original statement). Also, don't you still need a method for identifying the universe of maps that are in play?

Best Regards,
Richard Hamilton
XML Press
XML for Technical Communicators

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]