See also: IRC log
-> http://www.w3.org/XML/XProc/2015/05/27-agenda
Accepted.
-> http://www.w3.org/XML/XProc/2015/04/29-minutes
Accepted.
-> https://github.com/xproc/specification/issues
Roughly: I think Wednesday morning for agenda planning; some mixture of discussion and drafting.
Henry reports he's unavailable from 13:00-14:30 on Wednesday 10 June.
Norm: What time do we want to start?
Henry: Doors open at 8:30.
Norm: I propose we start at 9:00.
<ht> I propose we convene at the Police Box coffee kiosk, some 50 metres or so West of the front door, as they have good coffee. . .
Alex: Some of the issues are nicely compartmentalized; others are central to a better draft. We should try to organize that list a little better.
Norm: Perhaps we should divide the actions per spec to keep the step actions apart from the language actions.
Regrets from Norm and Henry for 3 June; next meeting is the face-to-face, 9:00a on Wednesday 10 June.
-> https://github.com/xproc/specification/issues/159
<ht> Sigh
Norm waffles about reading text files in known encodings from the filesystem; it's another crack between http-request and the filesystem.
Alex: We could add a property.
<ht> Call it file-media-type
Norm: Where would you add the property?
Alex: The request.
<ht> my mistake
Norm: Oh, ok, on the request not a document property in the "document properties" sense.
<alexmilowski> text/plain; encoding=ascii
Alex: Or we could add the encoding to the override-content-type as a parameter.
<ht> That is definitely a good way to do it
Norm: Yes, if you know the encoding you probably no the content-type.
Alex: We could spec it..."If you specify the encoding and it's not provided by the system, then that's the encoding used..."
<scribe> ACTION: Norm to spec the encoding parameter on override-content-type as a way to specify the encoding of documents. [recorded in http://www.w3.org/2015/05/27-xproc-minutes.html#action01]
<Zakim> ht, you wanted to note the different semantics, however. . .
Henry: Well, we spoke of this in terms of only applying when the encoding wasn't provided.
Norm: True.
Henry: I think the server encoding is more likely to be correct.
Norm mutters on about whether overriding the encoding from HTTP makes sense.
Norm: Perhaps we should document it as an override, use with care.
-> https://github.com/xproc/specification/issues/146
Norm: Aside from the microparsing of the attribute value, I'm sympathetic.
Alex: We could add attribute sets
Henry: That's just as complicated
as using p:set-properties
... It's premature optimization
Norm: Defer until we have more experience?
Alex: Are there certain
properties that we want to be able to set universally?
... Like content-type.
-> https://github.com/xproc/specification/issues/156
Norm: I have reservations.
Alex: I'd like to see the
pipeline that does this with directory listing.
... And see how that would be simplified by this change. Maybe
there are other things we can do to make this simpler.
... I think it's a good problem. I think we should look at the
pipeline instead of adding a feature to load now.
Norm: I just don't have to do this very often.
Alex: You do this with ebooks all the time.
Norm: Ah, true.
Alex: A few good examples would help us understand the problem better.
Alex: Can we take this to the list and have a conversation about agendas there.
Norm: I'd be happy to have some help.
Alex: Start with dividing the issue list into two sets and add priorities.
Norm: Ok. Modulo the fact that there isn't complete parity between the action list and the issue list.
Adjourned.