See also: IRC log
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0016
See: http://www.w3.org/2015/08/26-forms-minutes.html#item04
Clarification:
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0034.html
Erik: I did find an action item
about making the clarification about namespaces, that we would
have a new namespace
... but no action item for making the spec changes
... we need to agree on what changes are needed.
Steven: The minutes where we discussed this are here: http://www.w3.org/2015/08/26-forms-minutes.html#item04
ACTION 2025?
<ebruchez> http://www.w3.org/2015/10/14-forms-minutes.html
ACTION-2025?
<trackbot> ACTION-2025 -- Erik Bruchez to Deal with the naming over xpath versions -- due 2015-09-02 -- OPEN
<trackbot> https://www.w3.org/2005/06/tracker/xforms/actions/2025
Erik: Here is a place where we discussed the namespace: http://www.w3.org/2015/10/14-forms-minutes.html#item04
Steven: Is this the quote we need? "So it's like this - xf1: only contains all XForms functions, fn: only contains all XPath functions, and XF: contains both (less the clashes, where XPath functions are used, not the xf1 ones)"
Erik: I replied "so we put the
union of functions in the NEW namespace, which would be the
default namespace
... and then use xf: for the XForms-only namespace"
Steven: Right
... so we didn't give the action item for who was going to
write the spec text for that
<scribe> ACTION: Erik to provide new spec text for the two function namespaces [recorded in http://www.w3.org/2015/11/25-forms-minutes.html#action01]
<trackbot> Created ACTION-2041 - Provide new spec text for the two function namespaces [on Erik Bruchez - due 2015-12-02].
Steven: So does that solve it all?
Erik: Yes.
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0024.html
http://www.cwi.nl/~steven/forms/up-down.xml
Steven: Looks like a good suggestion; do we include it?
Alain: Sure.
Erik: Sure.
Steven: We were talking about patterns to put on the wiki, and this would be a good candidate for that as well.
<scribe> ACTION: Steven add up/down pattern to examples in spec [recorded in http://www.w3.org/2015/11/25-forms-minutes.html#action02]
<trackbot> Created ACTION-2042 - Add up/down pattern to examples in spec [on Steven Pemberton - due 2015-12-02].
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0021.html
Steven: 1 is obvious
<scribe> ACTION: Steven to make changes to JSON examples according to https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0021.html [recorded in http://www.w3.org/2015/11/25-forms-minutes.html#action03]
<trackbot> Created ACTION-2043 - Make changes to json examples according to https://lists.w3.org/archives/public/public-xformsusers/2015nov/0021.html [on Steven Pemberton - due 2015-12-02].
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0022.html
Erik: See
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0031.html
... for an alternative
... You know something is an empty array because it has
array="true"
... the empty top-level example is inconsistent
... it seems to me
Steven: Your proposal is to fix that inconsistency.
<ebruchez> https://gist.github.com/ebruchez/fd1c8d88f7966a9f0707
<ebruchez> https://gist.github.com/ebruchez/fd1c8d88f7966a9f0707#file-json-schemes-xml-L44
<ebruchez> https://gist.github.com/ebruchez/fd1c8d88f7966a9f0707#file-json-schemes-xml-L85
Erik: These examples show that it
is much easier to manage arrays in the new format
... not having an enclosing element makes it harder to do
... it also removes the need to have object="true" and
array="true"
... the XPath expressions we use look more natural, using the
names of the JSON keys in the path; this only changes that
slightly.
<ebruchez> I just lost you Steven
<ebruchez> did we lose you again?
<ebruchez> g[1]/_[2]
Steven: It is true that the initial design was to make the selecters as natural as possible, but if there are other constraints, we should take those into account
<ebruchez> g/_[1]/_[2]
Erik: It is unfortunate that the
expressions are less natural, but it means that arrays are
always consistently addressed
... But the fact that repeats and so on are much easier pushes
me over to the nested element version
Steven: I will work on the text then and submit a proposed new version before putting it in the spec
Erik: Ive played with all the examples, it round-trips fine and so on, it is not problematic, just the repeat/insert/delete is difficult to do
<scribe> ACTION: Steven t update JSON text to include nested array notation from https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0031.html [recorded in http://www.w3.org/2015/11/25-forms-minutes.html#action04]
<trackbot> Created ACTION-2044 - T update json text to include nested array notation from https://lists.w3.org/archives/public/public-xformsusers/2015nov/0031.html [on Steven Pemberton - due 2015-12-02].
Erik: It also simplifies the
extra attributes, since we can do it all with @type
... Alain, have you tried it yet?
Alain: I've tried the previous
version, I'm also implementing new node types,
... I'm also adding the CSV support
Steven: So you are happy with the changes Erik is suggestion
Alain: Yes
https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0030.html
Steven: Fine
Erik: Obvious
<scribe> ACTION: Steven to fix JSON according to https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0030.html [recorded in http://www.w3.org/2015/11/25-forms-minutes.html#action05]
<trackbot> Created ACTION-2045 - Fix json according to https://lists.w3.org/archives/public/public-xformsusers/2015nov/0030.html [on Steven Pemberton - due 2015-12-02].
Steven: Anything?
Oh congrats Erik btw on the South Tirol job
Another govt using XForms
[ADJOURN]
This is scribe.perl Revision: 1.144 of Date: 2015/11/17 08:39:34 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/abbout/about/ No ScribeNick specified. Guessing ScribeNick: Steven Inferring Scribes: Steven Present: Steven Alain Erik Regrets: Philip Nick Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2015Nov/0035 Found Date: 25 Nov 2015 Guessing minutes URL: http://www.w3.org/2015/11/25-forms-minutes.html People with action items: add down erik json pattern steven t text up update[End of scribe.perl diagnostic output]