See also: IRC log
No regrets given
Norm: I'll get a new version out
RSN with the new function we added.
... I saw a few editorial issues on the lists, I'll try to fix those too.
... I know we've all read the spec 100 times, but it would be nice to read it through one more time.
<scribe> ACTION: Mohamed volunteers to read the CR draft again [recorded in http://www.w3.org/2009/06/11-xproc-minutes.html#action01]
Norm: I don't have an official
report, but it's definitely less than 100%
... Does anyone have any ideas on how we might improve that?
Norm: Missing: escape-markup (serialization in general)
Vojtech: We also need to test p:store, probably the same way
Norm: That really can't be tested
in the current test harness, it'll need to be done by comparing
files on disk, I think.
... Really, if you overlook the serialization options, we're not in bad shape
Mohamed: It's hard to test serialization in all cases, consider p:exec
Norm: True, but if we get good results for, say p:store, I'm willing to argue that impls are likely to get it right in the other cases.
Norm: I think Vojtech and I are
both striving for 100% coverage.
... That's the bare minimum needed to declare victory.
Mohamed: I've heard of another one in google code.
Some discussion of doing XProc using GWT to get it into the browser.
Norm: Anything else to be said about implementation coverage?
Norm: In Henry's absence, I'm inclined to leave this until next week, but does anyone have anything they'd like to discuss?
Norm: Let's postpone this until Henry is present