-
Notifications
You must be signed in to change notification settings - Fork 95
Meeting minutes 2017 03 02
- Attendance
- OT project status updates
- New issues since last call
- Fixed issues since last call
- Project planning board review
- Discussion
- Peter
- Roger
- Shane
The DITA-OT contributor call was held immediately prior to the docs call. Meeting minutes are available at https://github.com/dita-ot/dita-ot/wiki/Meeting-minutes-2017-03-02.
- DITA-OT 2.4.4 was released on February 26. See http://www.dita-ot.org/2.4/release-notes/. For docs changes, see https://github.com/dita-ot/docs/compare/2.4.3...2.4.4.
5 new issues & pull requests were created since the last call.
- #128 – Is the --filter command path absolute or relative ?
- #129 – [Bug] dita --install syntax is incorrect
- #130 – Sample Ant builds contain obsolete information
- #131 – Document new Java API for 2.5
- #132 – Description of HTML5 parameters is not equal to the HTML-based output parameter description
A total of 3 issues were closed.
- #108 – Update information on “DITA for Print”
- #129 – [Bug] dita --install syntax is incorrect
- #130 – Sample Ant builds contain obsolete information
The docs issue tracker at https://github.com/dita-ot/docs/issues currently lists 18 open issues, 75 closed.
Our GitHub Projects boards show the status of issues currently associated with each release milestone and serve as the primary planning overview for upcoming releases:
- Send Peter Slack invitation
Proposals for resolving HTML5 duplicate parameter description (#132)
- Shane suggests generating a single warehouse topic via XSLT & filtering that, rather than pushing to existing topics
- If each parameter is created w/
@productor some other attribute, we could use that to filter
- Create #docs-indexing channel (or better: thread)
Roger mentions GDS article: https://gdsdata.blog.gov.uk/2017/01/12/using-data-science-to-build-a-taxonomy-for-gov-uk/
Shane suggests our docs are small enough that manual approach might be more effective than investing time to configure any automated tools for our purposes.
Peter says indexing is very labor intensive, points to guidelines in IBM Style Guide, checklist for quality assurance of indexes, controlled vocabulary, taxonomy, etc.
Peter points out that assigning groups of topics to different students could produce very different results. Best results are typically obtained when a single indexer is responsible for the overall output (more consistent). Shane points out that we'd need their permission to use the results of their efforts, make it clear to students that this is a real-world problem.
-
Give more than one group the same set of topics to see what they come up with, use that as a basis for discussion.
-
Peter likes the idea of a card sorting exercise as starting point for indexing.
-
Use student input as taxonomy proposal for us to discuss & iterate.
Discussing Shane's spreadsheet, he confirms that he deliberately excluded the existing hierarchy from the topic list to make sure we're not too attached to existing hierarchy (best to start all over when designing the taxonomy).
Peter favors a "common reference table of contents" ToC as one special case of a taxonomy (one type of "page one").
Shane suggests if we have different views on hierarchy, we could resolve competing hierarchies when both have value by providing a supplementary hierarchy.
We now group all parameter topics in one group by the plug-in that supplies them, what other groupings might be useful? (Like extension points, where we group them once by what they do, and then again by their plug-in source.)
Peter points out that facets are a key solution for providing multiple entry points to the same content, essential prerequisite for dynamic content delivery.
https://github.com/dita-ot/docs/wiki/Meeting-minutes-2017-03-02
Created 2017-03-02 15:57
View the latest DITA Open Toolkit docs at www.dita-ot.org/dev.