See also: IRC log
<trackbot> Date: 13 September 2012
<mhakkinen> will be calling in but wrapping up another call first.
Kelly: levels discussion -- try
and make progress and close down on some of that. We tried a
couple of approaches. We had a spreadsheet we were trying to
use to list our various rules for setting out criteria for
levels and we've had similar discussions around how we're going
to set the levels. We need to revisit the levels in order to
call ourselves done and make sure we feel comfortable...
... with them.
... Gut check from everyone -- how do you feel about the levels in the document today-- do you feel everything's pretty close, not, or no clue
Jan: haven't read it for the levels in a while, so don't know
Simon: same for me
Jeanne: WICAG does not have standards for levels, and that has caused problems, lawmakers need something to point to
Jan: ATAG we don't -- just talk
about the kinds of factors that go into the decision
... it's really complicated and sometimes different for different SC's
... what they did in Ontario -- takes all of WICAG up to double-A, but dump out live captions and video descriptions
... if we know that governments will be balancing different things -- our levels may not exactly line up with what they might have in mind for their legislative priorities
Jeanne: if WICAG had had something well written that could have been pointed to that might not have happened. If we can write it we should. I'd like to try
Jan: the wall that WICAG ran into
use somewhere you're going to get where you have to balance
different groups of people against each other -- you may say
something is really important for this group, maybe even low
hanging fruit, but then this thing that's really important for
another group of people, maybe a small group of people but it's
really really expensive were going to say you don't...
... have to do that, and it's going to be right there in black and white
Jim: technical ability versus needs and there's got to be a give somewhere and it can be ugly when you put it down on paper
Kelly: here's where the challenge
comes in -- the biggest thing that every time I've been
involved in any effort to set accessibility priorities, which
is really what we are trying to do here -- there's two things
that I see happen. The disability constituencies in the group
that are setting the priorities tend to get more of their needs
rated higher. Two, overall the priorities never are...
... willing to put the hard decisions down on paper that say basically we're making a choice to cut this group of people off because were going to lower priority for this item and put in those stark of terms --
... because as soon as you do using that small population, I can just as easily make that decision and say the whole population is small
... we spent a lot of time talking about it and came up with an elaborate formula
Kim: five criteria from the spreadsheet
Kelly: I'm not trying to be a naysayer, but... for me, and you can't defend it to a lawmaker, for me the way priorities get set most often is take a group of people who really understand the subject matter and make a judgment. Maybe that doesn't work. But that's how everything I've ever done ultimately -- I don't care what rules
Jan: that's ATAG tried to do
Jeanne: I think that's what we should do -- these are the factors we are trying to balance
<jeanne> 5 - no access, 4 -requires expert knowledge to access, 3 -hard or slow to accomplish task, 2 - makes it easier or more efficient, 1 - nice to have
<jeanne> Deterministic vs. Inferntial/ minor or major change/ number of current implementations
Jeanne: that's what we were
trying to balance when we did our spreadsheet.
... so level A if there requires access or if it requires expert knowledge to access , it's deterministic and it's a minor change and there are current implementations
Kelly: at least one problem with
even at that we have to address. When you say no access there's
an implied clause there that isn't in writing -- no access for
... is that really something you're willing to be put in writing
Jeanne: I think we have to be careful with how we write it -- the way we were scoring itno access for a number of groups of disabilities not how many people, but how many groups of people with disabilities did it affect
Kelly: is 100 groups of one more important than one group of 100?
Jeanne: yes because we are trying to protect the minorities
Jan: to rephrase Kelly's question 100 groups of one versus one group of 1 million, I guarantee that those 100 groups of one, were probably not going to hit too many of those. They are pretty big outliers.
Kelly: I recognize that, using that as an extreme example
Jeanne: I would prefer to dodge that as much as possible, but it is our responsibility to look out for minority groups
Jan: let's just go through the document rather than ahead of time spending time on the words that will justify our actions
Kelly: my real concern is that I'm not sure we can put in writing anything that we won't spend a lot of time revisiting. What's the total number of success criteria?
Kelly: What if we look at each one for one minute and made quick decisions -- it would take one meeting.
Kim: quickly go through, useful for sharing why's
Kelly: spreadsheet was complicated. we already have the why's, that's the whole point of the iintents
Jan: what's written down in the
intents are good, spreadsheet complicated
... if we can go through each one, take a minute for each one
Kelly: how would people feel if we took 30 minutes trying that approach once?
Jim: just as long as we go through them quickly
reading through SC's, noting only if anyone thinks something should be changed
Starting at 1.1.1
Jeanne: note on 1.1.3 there are some media where the captions are built-in like subtitles where they are in the movie and can't be moved
will discuss later
Jeanne: level problem -- maybe AAA
Jan: not all of the things that need to be underlined are in the draft yet, so maybe one recognized is underlined that might be enough
Starting at 1.2.1
<mhakkinen> no objection
Jan: this is repair -- AA?
Kelly: I could go down to AA to
1.1.2 should go down to AA
Correction: 1.2.1 should go down to AA
Jan: you're going to infer the relationship should go down to AA
Kelly: any objections:
1.2.2 should go down to AA
1.3.1 -- no change
Jan: the issue there is for elements with alternative content should it fail if it doesn't provide borders with alternative thickness?
Kelly: do we separate them or lower the whole priority?
Jim: could go with AA
1.3.2 should go down to AA
1.4.1 -- no change
1.4.2 -- no change
rethinking 1.4.2 Jan: would a system fail if it didn't wipe out the distinctions in the size of header text -- because I would be against that. I don't think it's a level a that user agents have to have a mode in which the zoom does not have to preserve size distinctions, so I think we have to remove the whether or not to keep it a level a? Is there a reason for washing away the size distinction?
Jeanne: yes that's what Wayne was talking about -- he redefines what the style looks like -- people with very large type it won't even fit into the screen
Jan: it doesn't have to be an option checkbox somewhere in the options
Jim: so where does the responsibility fall -- browser or user
Jan: we already have user stylesheets
1.4.2 Should discuss later -- maybe with Wayne's input
1.5.1 -- no change
1.6.1 -- no change
MH: are there some higher-quality engines that don't allow you to adjust pitch? I'm almost afraid we would downgrade some of the better voice options because of that
Jan: engines provided by platform services
MH: do some quick research on,
and maybe just add a note to it
... I think we can leave it as long as we qualify or not excluding some high quality engines because of that
1.6.3 we could use that language in a note for 1.6.2
<jeanne> ACTION: jeanne to add a note to 1.6.2 to insure that high quality speech engines are not excluded for pitch. Assigned to Mark [recorded in http://www.w3.org/2012/09/13-ua-minutes.html#action01]
<trackbot> Created ACTION-758 - Add a note to 1.6.2 to insure that high quality speech engines are not excluded for pitch. Assigned to Mark [on Jeanne F Spellman - due 2012-09-20].
1.6.3 -- no change
1.6.4 -- no change
1.7.1 -- no change
1.7.2 -- no change
1.7.3 -- no change
1.7.4 -- no change
1.8.1 -- no change
Jan: 1.3.2 AA, 1.8.1 single A combines that. If current focus was one of the 1.3.1 highlighting items than 1.3.2 would refer to it
Jim: but if it's up to the user to decide -- white background etc. they can do that, but if you put viewport up in 1.3.1 and 1.3.2 then we can eliminate
Jan: but viewports are a
... is there reason for the second sentence -- nested containers -- what does that add?
Jeanne: I think it was just there to clarify what is nested
1.8.1 -- no change
1.8.2 -- no change
Kelly: in roughly 45 minutes we
move through a fair number of these -- do we want to use more
meeting time to do this or do people want to read them
... probably going to take one more full meeting and probably a little of another to get through all of them
one sixth of the document in half the meeting time
Kelly: everyone good with it?
This is scribe.perl Revision: 1.136 of Date: 2011/05/12 12:01:43 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) No ScribeNick specified. Guessing ScribeNick: KimPatch Inferring Scribes: KimPatch WARNING: Dash separator lines found. If you intended them to mark the start of a new topic, you need the -dashTopics option. For example: <Philippe> --- <Philippe> Review of Action Items Default Present: +1.425.883.aaaa, Jim_Allan, +1.425.883.aabb, Kim_Patch, Jan, Jeanne, sharper, +1.609.734.aacc Present: Kelly Jan Jim Kim Jeanne Simon Mark Regrets: Greg Found Date: 13 Sep 2012 Guessing minutes URL: http://www.w3.org/2012/09/13-ua-minutes.html People with action items: jeanne[End of scribe.perl diagnostic output]