<trackbot> Date: 30 March 2015

<mattking> scribe: Mmatt_king

<scribe> scribe:mattking

John's Spin button example http://jongund.github.io/aria-examples/slider/slider-1.html

JN: nit... headingsabove tables should be captions

jg: should you be able to use aria-labelledby to set captions

JN: yes; guess need to update my checking tool

jg: seems to be3 common pattern; preferred by devs over caption
... OAA also checks for these
... aria-described should map to summary?

JN: yes

jg: summary attribute is depricated in html5

mk: I think the question of what to do with aria-describedby on table needs to be raised with ARIA WG

AA: Why is it required on these tables?

JN: that is just what the tool we use requires

MK: I want to discuss keyboard behaviors
... is orientation specified

jg: required?

jn: not for a horizontal; it is the default

<jamesn> scribeNick:jamesn

MK: left decreases, right increases. I would expect down to increase and up to decrease but the opposite is happening
... down is moving further ahead in a document so would increase the value. I would have though right, down and pagedown and end would all move in the same direction

BG: that follows the paradigm I have used in the past

AA: That is exactly the opposite of what we have in the current authoring practices

JN: if we honeslty believe it is wrong I would like to get wider input

BG: if you have a vertical slider then down arrow moves the slider down. but if you have a horizontal slider
... it moves it opposite

MK: certain keys move you forward, and others move you backwards
... think of it in document context. Down and right move you later, up and left move you earlier
... what i am most concerned about is that those groups of keys are all alligned with each other
... I don't want to associate a direction with an increase of decrease

jn: leads to a potential conflict with spinbutton

<discussions on various controls....>

jn: what does input type=range

<mattking> scribe: matt_king

<mattking> JN: fre fox and chrome: up arrow moves right page up moves right; down arrow and page down move left

<mattking> jn: up moves right; page up not implemented

<mattking> jn: in IE, home goes to left and end to right

<mattking> jn: windows slider for fonts, opposite of browsers: in win 8.1

<mattking> jn: win 8.1 up moving left and down moving right

<mattking> jn: sound slider is opposite of the font, up moves it right

<mattking> mk: with all the current implementation, changing current practice to get consistency is probably not worthwhile battle

<mattking> jn: Add a note to the effect that real world example exist where pgup/pgdn/up/down are opposite of recommendation and there agood reasons for this in some circumstances

https://www.w3.org/Bugs/Public/show_bug.cgi?id=28376

<mattking> jn: are we comfortable putting this in apg even though it doesn't work in IOS?

<mattking> jn: role=slider is completely broken in ios; input type=range works

<mattking> mk: Yes, I think we should put in the guide; it is IOS bug

<mattking> jn: I am in process of creating ios bug

<Zakim> jamesn, you wanted to ask what does type=range do in the browsers

Example coding practices

<mattking> jg: AT testing results is advisory. I think this is useful. A lot of developers do not have much AT experience

<mattking> AA: maintenance night mare, perhaps

<mattking> jg: we do not have to change it as AT changes

<mattking> aa: if you call out problem with certain combo, you should date it

<mattking> mmk: I don't think we should call out AT problems

<mattking> jg: maybe we just log observed behaviors

<mattking> mk: Maybe we just put in info about observed behaviors for AT that work

<mattking> mk: not sure we should put in any issues like bugs

<mattking> mk: if we are going to do this, i think it should be consistent across all examples.

<mattking> jg: this info could be in json objects so it is reusable

<mattking> mk: not sure if that is worthwhile at this time; nice idea but not sure if the AT info would be relavant without the attached example

<mattking> jg: could help us with our own maintenance of the examples

<mattking> jg: I will next work on a vertical slider, e.g., thermostat

<mattking> mk: can we put role/property/state in one column

<mattking> AA: still lot of people do not know which is which

<mattking> jemma: i like sepaate column

<mattking> mk: could put something like slider (role) in a combined column

<mattking> mk: we should provide links to the spec for the roles, states, and properties

<mattking> mk: there is duplication with the guide itself

<mattking> jn: however every example does not implement all features so this helps people understand what is implemented

<mattking> bg: or in cases where there is more than one way, to use or not to use active descendant

<mattking> jn: really like the example and the supporting info

<mattking> jg: should we put the source in the page?

<mattking> jn: could you just link to it in github

<mattking> mk: if you link to it in github, then you get the whole page vs just the widget

This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/Meeting: Protocols and Formats Working Group Teleconference/Meeting: ARIA Authoring Practices Task Force/
Found Scribe: Mmatt_king
Found Scribe: mattking
Inferring ScribeNick: mattking
Found ScribeNick: jamesn
Found Scribe: matt_king
Scribes: Mmatt_king, mattking, matt_king
ScribeNicks: jamesn, mattking
Default Present: Jon_Gunderson, Matt_King, James_Nurthen, +1.512.459.aaaa, +1.217.244.aabb, AnnAbbott, Jemma, LJWatson, Bryan_Garaventa
Present: Jon_Gunderson Matt_King James_Nurthen +1.512.459.aaaa +1.217.244.aabb AnnAbbott Jemma LJWatson Bryan_Garaventa

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 30 Mar 2015
Guessing minutes URL: http://www.w3.org/2015/03/30-aria-apg-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.