[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [dita-lightweight-dita] RE: Minutes of the OASIS Lightweight DITA SC for Oct 31, 2016
A couple of changes requested:
· Change “It consists of a single XSL stylesheet and generates a valid LwDITA DITA template topic. “ to “It consists of a single XSL stylesheet that generates a RELAX-NG schema from a DITA template topic.”
· Change “The schema generation tool currently supports all of the specialized element and attributes excepting specrole.” to “The schema generation tool currently supports all of the specialization elements and attributes excepting specrole.”
Thanks,
Tim.
From: dita-lightweight-dita@lists.
oasis-open.org [mailto:dita-lightweight-dita@lists.oasis-open.org ] On Behalf Of Keith Schengili-Roberts
Sent: Friday, November 04, 2016 8:40 AM
To: dita-lightweight-dita@lists.oasis-open.org
Subject: [dita-lightweight-dita] Minutes of the OASIS Lightweight DITA SC for Oct 31, 2016
==============================
===================
Minutes of the OASIS Lightweight DITA SC
Monday, 31 October 2016
Recorded by Keith Schengili-Roberts
In attendance: Don Day, Kristen Eberlein, Carlos Evia, Mark Giffin, Tim Grantham, John Hunt, Michael Priestley, Keith Schengili-Roberts
Motion to accept last week’s meeting minutes (with amendments) was put forward. Tim Grantham and Don Day seconded this motion. The motion was passed.
==Status of the Specification==
Carlos talked about recent discussions centering around whether one version of LwDITA (such as XDITA) be considered the “master” format of the specification from which the others versions are derived. He has since come around to the idea that the versions of LwDITA should instead be considered equal to each other. Kris came up with a challenge to the group to discuss an element which is not common to the other formats, specifically suggesting that shortdesc be examined. A meeting is scheduled for Thursday November 3, 2016 for those who want to discuss options for this in Markdown and HTML5 at length.
Kris asked that it also be considered what the design intent for the non-XML formats: will there be a one-to-one _expression_ of all elements or will there be some which will not be covered? Michael responded by saying that the intent should be to push for as much correspondence between the LwDITA language-specific variants as possible, but if there is an element that cannot be mapped, then hypothetically we state that there are inherent limitations in that particular _expression_ of LwDITA.
Kris then went on to ask whether these three flavours of LwDITA should be considered equal, stating that people are interested in knowing this. Michael’s response was to say that while it is possible that one of the flavours will provide more functionality, that does not mean that another “flavor” of LwDITA has the highest functionality or should be considered the "mother" version (as the true "mother" in this case is full DITA). If there are limitations in the current state of the specification for any of the versions of LwDITA, then they need to be acknowledged.
The question of how other the other flavours of LwDITA would deal with specialization also came up. Carlos responded that within Markdown that this can be declared in the header, but as HTML5 does not inherit this may not be possible with an HTML5-based version of LwDITA.
Mark also noted that outputclass exists in some places within LwDITA that does not currently correspond to full DITA. This is something that would need to be brought up at a future DITA TC discussion for DITA 2.0.
==Status of Schema Generation==
Tim reported that email that he sent out last Friday (Oct 28) covers the current state of the template tool and describes how someone can use it its current state. It consists of a single XSL stylesheet and generates a valid LwDITA DITA template topic. The schema generation tool currently supports all of the specialized element and attributes excepting specrole.
Tim derived the RelaxNG version of LwDITA by taking the RelaxNG version of full DITA 1.3 and removed sections not used in LwDITA. Mark pointed out some corrections for elements that were derived from full DITA 1.2 but differ in DITA 1.3.
Tim also mentioned that the intent of the RelaxNG version of LwDITA is to handle specializations. He also reported that the Jim Clark’s tool can automatically generate a DTD and an XSD from the RelaxNG source. Mark volunteered to test this functionality.
For all of the differences the plan is to align with DITA 1.3, and for the any variants to punt this to DITA 2.0 (to be discussed with DITA TC).
Mark summarized the discrepancies between LwDITA and full DITA 1.2 as the following:
- Elements with sub-elements like dl/dlentry, ol/li, simpletable/strow etc. -- define the sub-elements as one-or-more, not zero-or-more.
- <prolog> for LwDITA needs to have the outputclass attribute for use with specialization templates, even though DITA 1.3 does not have it. If not outputclass, then an equivalent attribute. Discussion is needed with main DITA TC on best way to do this.
- Add props attribute to <topic>, <prolog>, <shortdesc>, <body>, <fig>, <desc>, <ph>, <image>, <alt>, <data>, <param>, and <xref>.
- Add outputclass attribute to <fig>.
There followed an impromptu discussion of GitHub repositories for: ultimately it was agreed that the SC needs to tell TC that a separate GitHub repository for different submitters and maintainers and possibly a third for holding transformations.
The remaining agenda items (“Roadmap to Deliverables” and “Poll Results Discussion”) will be discussed at the next meeting.
Cheers!
-
Keith Schengili-Roberts
Market Researcher and DITA Evangelist
IXIASOFT
825 Querbes, Suite 200, Montréal, Québec, Canada, H2V 3X1
tel + 1 514 279-4942 / toll free + 1 877 279-4942
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]