Meeting minutes
<Steven> *Is
Now
<Steven> Because now is 14:00 UTC
<Steven> and the agenda says 15:00 UTC
Well - there is the QT4 meeting at 16:00 BST
<Steven> I only turned up in case the agenda was wrong!
rrsagent create minutes
Presend+: Michael
Previous Actions
ACTION 2023-01-10-b: Michael to take another pass on the EBNF to BNF document -- still continuing
ACTION 2023-01-10-d: Michael to work out the metadata for identifying the Unicode version(s) associated with a test --- could do with further discussion
ACTION 2023-01-10-f: Norm and Michael to do a bit of revision to improve the draft documentation of the XML vocabulary (see #137) for further discussion -- continues
<Github> invisibleXML/
ACTION 2023-01-10-h: Michael to review RELAX NG rules and report back - continues
ACTION 2023-02-07-a: Steven to make Pull Request for Unicode test - continues
ACTION 2023-02-07-b: Michael to think about adding to the test suite schema to cover browser (etc) version - discuss later
<john> s/dcsuss/discuss/
ACTION 2023-02-07-e: Bethan order stickers (for Markup UK) - proofs done but not yet ordered - continuing
Status reports
Status of implementations: CoffePot 2 released
Status of testing and test suites: discuss re the schema
Publication Plans
Steven: Link for community group report: https://
ACTION: Norm to investigate production of rules-conformant version of spec, for publication as CG Report
Norm: Proposes to publish the 1.0 erratum spec as the group report
Review and resolution of bug reports and technical issues
Michael: How do we detect and allow for platform variation in the test descriptions, e.g. Unicode version, browser, versions of Xengine....
Michael: There are additional effects, e.g. checking unreachability. Environmental issues can be either implementation defined or platform dependant.
Michael: Uses an 'appInfo' element in the test to predicate test result applicability
Michael: for issues like Unicode version, the tests are duplicated with predication at the top level.
John: Look at parallels with 'evironment/feature' predication in the QT3 test suites
Michael: options in a test case list a set of possible test results.
ACTION: Michael to explore the use of named environments in the test suite
Issue #139 Sample grammars for IRIs and URIs - contiues - make issue for next time
<Github> invisibleXML/
<cmsmcq> John reported that he has done work on prototype implementations of both grammar import and nonterminal renaming.
<cmsmcq> Grammar import can borrow a lot from the XSLT packaging mechanism. Requirements include specifying where the grammar to import can be found,
<cmsmcq> specifying which parts of it you want (JL proposes to interpret that list as 'these and everything reachable from them'),
<cmsmcq> also overriding rules in the imported grammar
<cmsmcq> and supplementing them.
<cmsmcq> Supplements might include adding an alternative.
<cmsmcq> SP says he would really like a prose description of what this would all mean, and why particular features are thought to be required.
ACTION: John to write up his work on grammar import.
<cmsmcq> On nonterminal renaming, John says he has found it trickier than it looks. Especially dynamic renaming.
<cmsmcq> He has come to believe that in many cases, all the action affects the production of output -- in this respect, renaming is an extension of the existing set of marks.
<cmsmcq> The trickiness is less in the implementation than in making clear exactly what effect a directive like this will have on the grammar.
<cmsmcq> JL notes that current marks can occur either on definitions or on uses. For renaming, a similar dichotomy arises, but the reference seems particularly important.
<cmsmcq> If you just want to rename it, why not just name the nonterminal differently?
<cmsmcq> SP and NDTW answer: because you've already got one with that name.
<cmsmcq> NDTW says that if we are going to discuss grammar composition, we should try to schedule some time on the use cases document.
ACTION: John to write up his work on nonterminal renaming.