← Prev in month
← Prev in thread
Next in thread →
Next in month →
MEETING MINUTES -- 11 JAN 2005 -- DITA TECHNICAL COMMITTEE
posting for France, minute taker: ** AGENDA ** ------------ 1. Roll call 2. Review/approve minutes from 14 December http://lists.oasis-open.org/archives/dita/200412/msg00033.html and 04 January http://lists.oasis-open.org/archives/dita/200501/msg00032.html 3. Affirm tentative decisions from last week: - Namespace URI: dita.oasisopen.org/1.0/ - Naming conventions will be removed for shell doctypes. - Need to add query, translate and xml:lang attributes in the formal update of the map dtd. 4. Specification status As of last week: - namespaces in design: to be written - developing processes: to be written - CSS fallback support: Don to get back - namespaces in processing: to be written Related list issues: - otherprops syntax - URIs and versions 5. Bug lists (triage): http://lists.oasis-open.org/archives/dita/200411/msg00023.html 1. Make the body element optional in base topic as well as for task, concept, and reference 2. Make the keyref attribute CDATA instead of NMTOKEN so we can use URIs as keys 3. Retire the boolean element 4. Insert <desc> into <xref> -- for equivalence between xref and topicref 5. Choice is missing a number of elements -- should have same content model as <li> 6. Expand map DTD so meta elements are equivalent to topic meta (content elements in map should match same elements in topic: shortdesc, searchtitle, linktext) http://lists.oasis-open.org/archives/dita/200412/msg00031.html (rationale) http://lists.oasis-open.org/archives/dita/200501/msg00026.html (markup) 7. Add mapref to map http://lists.oasis-open.org/archives/dita/200412/msg00028.html (issue 8 fixes a syntax error introduced by the fix for #9) 8. Change the syntax for the keyword definition to parameterize the "tm" element http://lists.oasis-open.org/archives/dita/200501/msg00004.html (June era toolkit fixes that were not reported at the time) 9. Add tm into keyword content model to enable proper trademarking of keyword content. 10. Map and topic cannot share same keyword definition: recommend cutting the keyword element declaration from meta_xml.mod and paste separate defs into map.mod (no tm) and topic.mod (with tm). 11. Elements derived from keyword must have same content models-- need to revert words.cnt back to "PCDATA" for: msgnum, cmdname, varname option, parmname, apiname, kwd wintitle 12. Default attribute values must be declared with " delimiters http://lists.oasis-open.org/archives/dita/200501/msg00006.html 13. Provide more equivalence between topicref/link atts: @query, @translate, @xml:lang If time permits--statements, at least, on these questions from users. - Should <tm> allow images or logoized content? http://lists.oasis-open.org/archives/dita/200411/msg00024.html - Should <keyword> be allowed to nest? http://lists.oasis-open.org/archives/dita/200411/msg00025.html 6. AOB? ** MINUTES ** ------------- 1. Roll call - See separate attendance document. 10/19--met quorum 2. Review/approve minutes from 14 December http://lists.oasis-open.org/archives/dita/200412/msg00033.html and 04 January http://lists.oasis-open.org/archives/dita/200501/msg00032.html - Review decisions and minutes of December 14th Michael P motion to approve. Second: Yas. Accepted - Review minutes of Jan 4th Review, update: Chris Wong instead of ??? as speaker. - Correction: namespace should be "oasis-" in tentative decision. - Motion to approve: MPriestley Second: Eliot. 3. Affirm tentative decisions from last week: - Namespace URI: dita.oasis-open.org/1.0/ (with hypen between oasis and open) - Discussions: Erik Hennum - Sent attempt to summarize on the list. First half about file URIs… second half is namespace. Paul thought 1 namespace in 1.0. Erik Hennum: decided to have required namespace for dita arch attr Optional for document types so that they would be std namespace in environ that require them Paul: A namespace for each doctype? EH: Yes but optional in 1.0. Will go on architecture att not the class att. EH: It would apply to the root element within the doctype. It's a namespace for the entire document. Namespace in different doctype are diff. Paul: topic is a topic, it should be the same EH: Elements within each elements would be different ?? Would you still be able to reuse by conref? Paul: Not really. Paul. Do we need to do that for 1.0? EH Some people on the list said they absolutely needed some namespaces. Paul. Thought class att namespace was enough for 1.0. Don: can someone summarize how we moved from last week simple proposal to more complex. EH: Minutes did not actually recorded or reflected the fact that what was approved in Nov was namespace for the (…) doctype and not the class. If we have to repopen discussion we should. If we have people that say that dita cannot be used with Word unless you use an actual namespace, are we ignoring our charter if we pay attention to that? Don: Paul I think I look at your experience with TC process and actual info and tools. Where are we missing the namespace requirement for oasis? Would it be satisfy users to simply have a arch namespace ???: Not going to solve the word issue. Don. What about the reuse problem that Paul mentioned. Paul. Does that make sense to have element reused. Namespaces are creating a bunch of names, if I have 2 namespaces I have 2 different names, I don't think that's what dita was trying to do, cause then you can't reuse them EH: Should we take interested parties off line? TODO : TO bring this discussion offline: Who should be lead in setting up teleconf? EH. Others to include: Paul, Eliott, Bruce, Michael P., Eric Sirois. Will bring this subject to next week meeting. - Naming conventions will be removed for shell doctypes. Motion to approve: Michael. Second: France. - Need to add query, translate and xml:lang attributes in the formal update of the map dtd. Accept proposal 13 (from something??? 16). Don: this is a Consistency fix. Motion to approve: Erik Hickson Second: Wendy Sh. - 4. Specification status As of last week: - namespaces in design: to be written - developing processes: to be written - CSS fallback support: Don to get back - namespaces in processing: to be written Related list issues: - otherprops syntax - URIs and versions - Don in dec we had approve 1 to 5 and 8 skip 6 and 7. this week we approved #13. Bugs 9 to 12 are late reporting. Bug fixes have been done. They are listed so they are on the table. I can back tehm out. But we are approving after the fact bugs that were known in June but not reported at that time. Propose acceptance on the list 9 to 12. Only 6 and 7 will be left to trade off. - Paul what is # 12? - Don there was an attribute in utilities the use of no key word in an attribute had a single quote and need double for some parsers. Since that change was made there's been no issues. -- Erick Hickson. Why not alow tm and keyword in map… concern with that. Maps generally allow to put stuff that you don't think is appropriate in topic. - ??? Adds that this is counter to direction in # 6. - Don: yes 6 is actually an attempt to formalize content model for shared element between maps and topics should ideally have the same content model - M.: Change # 10 to map and topic need to share and add tm to content model for map. - Don remove 10 from what we are aporoving and make 6 complete - M: I agree with 10 and 6 seperated. - Don proposal accept 9 11 12. 10 remain as bug with 6 and 7. - Micheal: motion. Sharon (Lasgo?) Second. - Michael lets deal with 7 now or discuss it on the list. - Don: M have you had any additional stuff on 7. - Michael let's explain why it's there. - M: The immediate need was to process multiple map at one go. You end up with a lot of maps. You want to build them all at once because you want. (…). It met a very specific problem. But we realize we have a non standard part. We should have a way of referencing map. So we want to get mapref in there to have a coherent architecture. In terms of ways to do that. Current proposal add mapref. Otherwise leave mapref as an odd child. If we say no to map ref, we will need a toolkit special doc on how to do it. Does affect the processing pipeline. Just a way to formalize stuff that was done informally in the past. - Don: What can we get out of this? - Eliot understands and agree that mapref is probably the best choice overall. - Paul: It seems to be the right technical solution. Still confuse about process of what 10.o is supposed to be - Don: loosing people so full count to vote is no there, so bring it up next week. - MP: Should we take it on the list? - Don: yes - MP: Also looking for feedback on what we already have. - Don: todo for everyone bring comments to Michael. - 5. Bug lists (triage): http://lists.oasis-open.org/archives/dita/200411/msg00023.html 1. Make the body element optional in base topic as well as for task, concept, and reference 2. Make the keyref attribute CDATA instead of NMTOKEN so we can use URIs as keys 3. Retire the boolean element 4. Insert <desc> into <xref> -- for equivalence between xref and topicref 5. Choice is missing a number of elements -- should have same content model as <li> 6. Expand map DTD so meta elements are equivalent to topic meta (content elements in map should match same elements in topic: shortdesc, searchtitle, linktext) http://lists.oasis-open.org/archives/dita/200412/msg00031.html (rationale) http://lists.oasis-open.org/archives/dita/200501/msg00026.html (markup) 7. Add mapref to map http://lists.oasis-open.org/archives/dita/200412/msg00028.html (issue 8 fixes a syntax error introduced by the fix for #9) 8. Change the syntax for the keyword definition to parameterize the "tm" element http://lists.oasis-open.org/archives/dita/200501/msg00004.html (June era toolkit fixes that were not reported at the time) 9. Add tm into keyword content model to enable proper trademarking of keyword content. 10. Map and topic cannot share same keyword definition: recommend cutting the keyword element declaration from meta_xml.mod and paste separate defs into map.mod (no tm) and topic.mod (with tm). 11. Elements derived from keyword must have same content models-- need to revert words.cnt back to "PCDATA" for: msgnum, cmdname, varname option, parmname, apiname, kwd wintitle 12. Default attribute values must be declared with " delimiters http://lists.oasis-open.org/archives/dita/200501/msg00006.html 13. Provide more equivalence between topicref/link atts: @query, @translate, @xml:lang If time permits--statements, at least, on these questions from users. - Should <tm> allow images or logoized content? http://lists.oasis-open.org/archives/dita/200411/msg00024.html - Should <keyword> be allowed to nest? http://lists.oasis-open.org/archives/dita/200411/msg00025.html 6. AOB? ** Todos from today's meeting ** -------------------------------- EH to bring the namesapce discussion offline. Others to include in discussion: Paul, Eliott, Bruce, Michael P., Eric Sirois. Everyone bring comments to Michael on what we already have. ** Decisions from today's meeting ** ------------------------------------ Approved minutes form Dec 14th and Jan 4th meetings with corrections. Naming conventions will be removed for shell doctypes. Need to add query, translate and xml:lang attributes in the formal update of the map dtd. Bug resoltutions: accept 9 11 12 as solutions. 10 remain as bug with 6 and 7.
← Prev in month
← Prev in thread
Next in thread →
Next in month →