Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-03-13 Thread Richard Moore
On Thu, Feb 10, 2011 at 4:39 AM, Ted Gould t...@ubuntu.com wrote:
 1.  Expert screenreader users will, if they can, save time by listening
    to only the first part of an item's label before navigating
    elsewhere. So an accessible label may put variable information
    first (for example, 22 percent battery), while the tooltip may be
    better presenting the same information in key-value form (for
    example, Battery (22%)).

This is a great example, and suggests that we also need this
information included within the spec - ie. the distinction needs to be
made clear to developers using it. If we don't make it clear then
we'll end up with developers simply reusing the tooltip text anyway!

Cheers

Rich.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-03-11 Thread Matthew Paul Thomas
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Aaron J. Seigo wrote on 03/03/11 19:53:

 On Thursday, March 3, 2011, Matthew Paul Thomas wrote:

 Aaron J. Seigo wrote on 03/03/11 06:15:

 On Tuesday, March 1, 2011, Matthew Paul Thomas wrote:
...
 But in general, while all interactive graphic-only elements should
 have accessible labels, not all of them need tooltips.

 example?

 Icon-only labels at opposite ends of a slider: for example, muted at
 one end and maximum-volume at the other, or low-zoom at one end and
 high-zoom at the other. The icons should have accessible labels, but
 to a sighted user they are obvious enough that tooltips would be
 silly. (If they weren't obvious enough, the correct fix would be to
 make the labels plain text, as slider labels often are, not to give
 them tooltips.)
 
 i was hoping for examples relevant to the topic at hand, namely status 
 notifiers / app indicators. :)

You asked for an example of my in general point, not of status
notifiers in particular.

For status notifiers in particular, as you may know, Ubuntu's design
position is that the answer to which app indicators should be obvious
enough not to need a tooltip is all of them.

...
 If the syntax had been img src=...alternate/img, it would have
 been much more obvious that an alternate was expected and what it was
 for. Sure, some people would still have routinely written
 img src=.../img. But fewer people would have, because it would
 have been more obvious that that was missing something than that
 img src=... is missing something.
 
 img src=http://idontcare.com/haha.png; /
 
 :)
 
 most API is abusable, and app developers will do just that. a truly
 crappy part of reality.

We're not disagreeing. The fact remains that a well-designed API will be
used in good ways more often than a badly-designed API. An API that
makes accessibility obvious will be used accessibly more often than one
that makes accessibility non-obvious.

- -- 
mpt
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk16WSYACgkQ6PUxNfU6ecrctQCeMipT99JLeyPtzJd+a3NINzcO
G9EAn0Kp/KOVnfGFoyUpIGEfR9WLI0x4
=fzlb
-END PGP SIGNATURE-
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-03-03 Thread Matthew Paul Thomas
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Aaron J. Seigo wrote on 03/03/11 06:15:

 On Tuesday, March 1, 2011, Matthew Paul Thomas wrote:
...
 But in general, while all interactive graphic-only
 elements should have accessible labels, not all of them need tooltips.
 
 example?

Icon-only labels at opposite ends of a slider: for example, muted at one
end and maximum-volume at the other, or low-zoom at one end and
high-zoom at the other. The icons should have accessible labels, but to
a sighted user they are obvious enough that tooltips would be silly. (If
they weren't obvious enough, the correct fix would be to make the labels
plain text, as slider labels often are, not to give them tooltips.)

...
 In a survey by Opera Software of 3,219,487 Web pages that used the
 img element, 2,520,939 of them (78%) used alt=, while 367,132 (11%)
 used title=.
 http://dev.opera.com/articles/view/mama-images-elements-and-formats/#img

 As far as I know, there are no public statistics on how often img
 elements have distinct alt= and title= values. But we can conclude, at
 least, that Web authors care enough about accessibility to provide
 accessible equivalents most of the time.
 
 after years of getting it pounded into their heads and with the added
 bonus that this text is often used for things like tooltips ... they
 still fail 11% of the time.
...

That is because the img API was apparently designed without any
thought for accessibility.
http://diveintomark.org/archives/2009/11/02/why-do-we-have-an-img-element

If the syntax had been img src=...alternate/img, it would have
been much more obvious that an alternate was expected and what it was
for. Sure, some people would still have routinely written
img src=.../img. But fewer people would have, because it would
have been more obvious that that was missing something than that
img src=... is missing something.
http://sweng.the-davies.net/Home/rustys-api-design-manifesto

So I think it is important to bake accessible labels into the API in a
way that makes it obvious that they are needed, even for things that
shouldn't have tooltips.
http://behindthecurtain.us/2010/06/12/my-first-week-with-the-iphone/

- -- 
mpt
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk1v5oEACgkQ6PUxNfU6eco0iwCgguPHPPa99u+NupQP7VF3F0Bb
Gt0An3QkHeQgxUYM2Fl4A+o77cdg1rCr
=yE+G
-END PGP SIGNATURE-
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-03-03 Thread Aaron J. Seigo
On Thursday, March 3, 2011, Matthew Paul Thomas wrote:
 Aaron J. Seigo wrote on 03/03/11 06:15:
  On Tuesday, March 1, 2011, Matthew Paul Thomas wrote:
 ...
 
  But in general, while all interactive graphic-only
  elements should have accessible labels, not all of them need tooltips.
  
  example?
 
 Icon-only labels at opposite ends of a slider: for example, muted at one
 end and maximum-volume at the other, or low-zoom at one end and
 high-zoom at the other. The icons should have accessible labels, but to
 a sighted user they are obvious enough that tooltips would be silly. (If
 they weren't obvious enough, the correct fix would be to make the labels
 plain text, as slider labels often are, not to give them tooltips.)

i was hoping for examples relevant to the topic at hand, namely status 
notifiers / app indicators. :)

  In a survey by Opera Software of 3,219,487 Web pages that used the
  img element, 2,520,939 of them (78%) used alt=, while 367,132 (11%)
  used title=.
  http://dev.opera.com/articles/view/mama-images-elements-and-formats/#im
  g
  
  As far as I know, there are no public statistics on how often img
  elements have distinct alt= and title= values. But we can conclude, at
  least, that Web authors care enough about accessibility to provide
  accessible equivalents most of the time.
  
  after years of getting it pounded into their heads and with the added
  bonus that this text is often used for things like tooltips ... they
  still fail 11% of the time.
 
 ...
 
 That is because the img API was apparently designed without any
 thought for accessibility.

true ... but:

 If the syntax had been img src=...alternate/img, it would have
 been much more obvious that an alternate was expected and what it was
 for. Sure, some people would still have routinely written
 img src=.../img. But fewer people would have, because it would
 have been more obvious that that was missing something than that
 img src=... is missing something.

img src=http://idontcare.com/haha.png; /

:)

most API is abusable, and app developers will do just that. a truly crappy 
part of reality.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-03-02 Thread Aaron J. Seigo
On Tuesday, March 1, 2011, Matthew Paul Thomas wrote:
  I understand that it's suboptimal (though thanks, Matthew for laying it
  out in greater detail), my reservation was that many developers simply
  don't care about accessibility enough to add another string, so it
  probable makes sense if the accessible label falls back to the tooltip
  when the developer forgot to specify the accessible label.
 
 In the particular case of StatusNotifierItem, KDE developers may decide
 for consistency that they should always, or (as in Ubuntu) should never,
 have tooltips. 

irrelevant, as this is something decided on the visualization side, not 
something the application developer needs to worry about.

 But in general, while all interactive graphic-only
 elements should have accessible labels, not all of them need tooltips.

example?

  Images in HTML. After a long struggle, accessibility advocates finally
  got most browsers to treat alt= (the accessible equivalent) and title=
  (the tooltip) differently, to help Web authors understand that what's
  good for one is rarely good for the other.
  
  Exactly, and how many websites end up with this distinction correctly
  implemented?
 
 ...
 
 In a survey by Opera Software of 3,219,487 Web pages that used the img
 element, 2,520,939 of them (78%) used alt=, while 367,132 (11%) used
 title=.
 http://dev.opera.com/articles/view/mama-images-elements-and-formats/#img
 
 As far as I know, there are no public statistics on how often img
 elements have distinct alt= and title= values. But we can conclude, at
 least, that Web authors care enough about accessibility to provide
 accessible equivalents most of the time.

after years of getting it pounded into their heads and with the added bonus 
that this text is often used for things like tooltips ... they still fail 11% 
of the time.

colour me unconvinced :)

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-03-01 Thread Matthew Paul Thomas
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Sebastian Kügler wrote on 10/02/11 17:22:

 On Thursday, February 10, 2011 05:39:08 Ted Gould wrote:
...
 I forwarded the discussion that was had about using the title in the
 tooltip for the accessible label.  While I felt that it was not
 adequate I couldn't express why.  As per usual, mpt says it better
 than I, here is his response:
...
 I understand that it's suboptimal (though thanks, Matthew for laying it
 out in greater detail), my reservation was that many developers simply
 don't care about accessibility enough to add another string, so it
 probable makes sense if the accessible label falls back to the tooltip
 when the developer forgot to specify the accessible label.

In the particular case of StatusNotifierItem, KDE developers may decide
for consistency that they should always, or (as in Ubuntu) should never,
have tooltips. But in general, while all interactive graphic-only
elements should have accessible labels, not all of them need tooltips.
So thinking about it as add another string could be a bit misleading.

...
 Images in HTML. After a long struggle, accessibility advocates finally
 got most browsers to treat alt= (the accessible equivalent) and title=
 (the tooltip) differently, to help Web authors understand that what's
 good for one is rarely good for the other.
 
 Exactly, and how many websites end up with this distinction correctly 
 implemented?
...

In a survey by Opera Software of 3,219,487 Web pages that used the img
element, 2,520,939 of them (78%) used alt=, while 367,132 (11%) used
title=.
http://dev.opera.com/articles/view/mama-images-elements-and-formats/#img

As far as I know, there are no public statistics on how often img
elements have distinct alt= and title= values. But we can conclude, at
least, that Web authors care enough about accessibility to provide
accessible equivalents most of the time.

- -- 
mpt
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk1s6NkACgkQ6PUxNfU6ecovQACfQxQArrvx3Hd/8iirN0z+XvQK
AZIAnjVqiyp071ZLORODYLnkY2jjc6vI
=i+4S
-END PGP SIGNATURE-
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-02-10 Thread Sebastian Kügler
On Thursday, February 10, 2011 05:39:08 Ted Gould wrote:
   Plasma folks,
 
 I forwarded the discussion that was had about using the title in the
 tooltip for the accessible label.  While I felt that it was not adequate
 I couldn't express why.  As per usual, mpt says it better than I, here
 is his response:
 
  Forwarded Message 
 From: Matthew Paul Thomas m...@canonical.com
 
 Not enough information to check signature validity.  Show Details
 
   Ted Gould wrote on 08/02/11 20:23:
 ...
 
  Do you have thoughts on this from the a11y perspective?  They want to
  use the title of the tooltip for the accessible label.  It seems to me
  that would be sub-optimal, but if so I need a good reason :)

I understand that it's suboptimal (though thanks, Matthew for laying it out in 
greater detail), my reservation was that many developers simply don't care 
about accessibility enough to add another string, so it probable makes sense 
if the accessible label falls back to the tooltip when the developer forgot to 
specify the accessible label.

 
 1.  Expert screenreader users will, if they can, save time by listening
 to only the first part of an item's label before navigating
 elsewhere. So an accessible label may put variable information
 first (for example, 22 percent battery), while the tooltip may be
 better presenting the same information in key-value form (for
 example, Battery (22%)).
 
 2.  A smart human reading something aloud will often expand
 abbreviations. A screenreader often won't understand the context
 well enough to do this, so an accessible label may pre-expand
 abbreviations instead. For example, whereas a tooltip may compactly
 say Battery (4:05 remaining), an accessible label might better
 say 4 hours 5 minutes battery remaining. (Luke may want to
 correct me on this.)
 
 3.  Some things that need to be conveyed in accessible labels would be
 utterly redundant in tooltips. For example, something like Ubuntu's
 session menu needs an accessible label Session to briefly
 identify it. But if we were giving our menus tooltips, Session
 would be rather useless; a better tooltip would be something like
 Commands for exiting Ubuntu.
 
Any
 
  examples anywhere I could point to?
 
 ...
 
 Images in HTML. After a long struggle, accessibility advocates finally
 got most browsers to treat alt= (the accessible equivalent) and title=
 (the tooltip) differently, to help Web authors understand that what's
 good for one is rarely good for the other.

Exactly, and how many websites end up with this distinction correctly 
implemented?

Anyway, I don't have any objections, just the above reservation and possible 
workaround.
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: [Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-02-10 Thread Marco Martin
On Thursday 10 February 2011, Sebastian Kügler wrote:
 On Thursday, February 10, 2011 05:39:08 Ted Gould wrote:
Plasma folks,
  
  I forwarded the discussion that was had about using the title in the
  tooltip for the accessible label.  While I felt that it was not adequate
  I couldn't express why.  As per usual, mpt says it better than I, here
  is his response:
  
   Forwarded Message 
  From: Matthew Paul Thomas m...@canonical.com
  
  Not enough information to check signature validity.  Show Details
  
Ted Gould wrote on 08/02/11 20:23:
  ...
  
   Do you have thoughts on this from the a11y perspective?  They want to
   use the title of the tooltip for the accessible label.  It seems to me
   that would be sub-optimal, but if so I need a good reason :)
 
 I understand that it's suboptimal (though thanks, Matthew for laying it out
 in greater detail), my reservation was that many developers simply don't
 care about accessibility enough to add another string, so it probable
 makes sense if the accessible label falls back to the tooltip when the
 developer forgot to specify the accessible label.

yeah, i have this concern as well, so let's go for:
search AccessibleItemLabel
if not found, fall back to tooltip title?

and i would even specify in the spec that falling back is intended and fine, 
only labels that really have to be different from tooltips for some reason 
should overwrite the tooltip

Cheers,
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


[Fwd: Re: [Fwd: Re: New properties for StatusNotifierItem: Accessible Label (1/3)]]

2011-02-09 Thread Ted Gould
Plasma folks,

I forwarded the discussion that was had about using the title in the
tooltip for the accessible label.  While I felt that it was not adequate
I couldn't express why.  As per usual, mpt says it better than I, here
is his response:

 Forwarded Message 
From: Matthew Paul Thomas m...@canonical.com

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Ted Gould wrote on 08/02/11 20:23:
...
 Do you have thoughts on this from the a11y perspective?  They want to
 use the title of the tooltip for the accessible label.  It seems to me
 that would be sub-optimal, but if so I need a good reason :)

1.  Expert screenreader users will, if they can, save time by listening
to only the first part of an item's label before navigating
elsewhere. So an accessible label may put variable information
first (for example, 22 percent battery), while the tooltip may be
better presenting the same information in key-value form (for
example, Battery (22%)).

2.  A smart human reading something aloud will often expand
abbreviations. A screenreader often won't understand the context
well enough to do this, so an accessible label may pre-expand
abbreviations instead. For example, whereas a tooltip may compactly
say Battery (4:05 remaining), an accessible label might better
say 4 hours 5 minutes battery remaining. (Luke may want to
correct me on this.)

3.  Some things that need to be conveyed in accessible labels would be
utterly redundant in tooltips. For example, something like Ubuntu's
session menu needs an accessible label Session to briefly
identify it. But if we were giving our menus tooltips, Session
would be rather useless; a better tooltip would be something like
Commands for exiting Ubuntu.

   Any
 examples anywhere I could point to?
...

Images in HTML. After a long struggle, accessibility advocates finally
got most browsers to treat alt= (the accessible equivalent) and title=
(the tooltip) differently, to help Web authors understand that what's
good for one is rarely good for the other.

- -- 
mpt
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk1SjcsACgkQ6PUxNfU6ecptswCfcvg80at8zAnXo+oGfyRpgldb
GfIAoMFC4NPEJr1HFBWXy/hhesW+zP9V
=A97b
-END PGP SIGNATURE-



signature.asc
Description: This is a digitally signed message part
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-08 Thread Aurélien Gâteau
(resending, cc'ing Ted)
  
  What would help is if the tooltip was using semantic markup, instead of a
  subset of html. Think about something like this:
  
  tooltip
titleAmarok/title
icon src=amarok.png alt=Amarok/
bodyCurrently playing: Qt 4 dance by Trolltech/body
  /tooltip
  
  With such a format, it would be possible to extract meaningful information
  for a11y purpose.
  
  Heck, this could even be somehow standardized to also work for
  notifications (/me looks at this thread slowly drowning into markup
  bikeshed)
 
 title and icons are already outside the body in the spec and the title can't 
 have html
 th tooltip is a single proprty, that is a struct unfolded as this
 icon (name or binary dump as usual)
 title (short, plain text, must be meaningful and always present)
 subtitle (optional, can have html)

Ohhh... my memory is failing me. Indeed, tooltips are actually not a single 
field. With this in mind, I agree using the tooltip title should be good 
enough. What do you think, Ted?

Aurélien
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Aurélien Gâteau
 On Thursday, February 3, 2011, Aurélien Gâteau wrote:
  There are ways to strongly suggest application developers to define such
  strings: for example outputing warnings on stderr when a KSNI goes live
  without having proper a11y properties set.
 
 catching and punishing sins (though a warning is hardly a punishment unless 
we 
 email the devs every time it happens .. yes, i'm joking ;) is not a good 
 approach compared to being able to prevent the sins in the first place.
 
 developers have every motivation to add tooltips and they usually do. it's 
 also easily tested by them: hover the entry with their mouse. if we use that 
 same data for a11y, we prevent the sins.

Using the tooltip is a handy fallback, but it will most likely be less adapted 
to a blind reader, especially since it may contain complex HTML content 
(reminds me of another discussion...), making it much more work to read for a 
blind user than a simple, text-only, label.

I suggest we add the property Ted is proposing and specify that tooltips will 
be used as a fallback, but strongly recommends devs to set appropriate a11y 
labels (and output warnings in KDE implementation when they are not set).

Aurélien
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Aurélien Gâteau
  Yeah, in talking to the a11y folks they like tooltips, but they really
  wanted another label as sometimes the tooltips don't describe the icon
  enough for a blind user to know what it means. 
 
 this is true for traditional tooltips, but the ones in a SNI are not like 
 those tooltips at all: they have titles, subtext, etc. they are supposed to, 
 in an SNI, be fully descriptive. are the a11y folks aware of those
 differences between SNI tooltips and normal ones on  e.g. pushbuttons?
 
 what this denotes to me is that maybe we need to revamp the documentation 
 around tooltips (and as marco pointed out on irc earlier today, perhaps that 
 was an unfortunate name to choose in the spec :/ ) so that this is very
 clear to app devs too.

What would help is if the tooltip was using semantic markup, instead of a 
subset of html. Think about something like this:

tooltip
  titleAmarok/title
  icon src=amarok.png alt=Amarok/
  bodyCurrently playing: Qt 4 dance by Trolltech/body
/tooltip

With such a format, it would be possible to extract meaningful information for 
a11y purpose.

Heck, this could even be somehow standardized to also work for notifications 
(/me looks at this thread slowly drowning into markup bikeshed)

Aurélien
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Marco Martin
On Monday 07 February 2011, Aurélien Gâteau wrote:
   Yeah, in talking to the a11y folks they like tooltips, but they really
   wanted another label as sometimes the tooltips don't describe the icon
   enough for a blind user to know what it means.
  
  this is true for traditional tooltips, but the ones in a SNI are not like
  those tooltips at all: they have titles, subtext, etc. they are supposed
  to, in an SNI, be fully descriptive. are the a11y folks aware of those
  differences between SNI tooltips and normal ones on  e.g. pushbuttons?
  
  what this denotes to me is that maybe we need to revamp the documentation
  around tooltips (and as marco pointed out on irc earlier today, perhaps
  that was an unfortunate name to choose in the spec :/ ) so that this is
  very clear to app devs too.
 
 What would help is if the tooltip was using semantic markup, instead of a
 subset of html. Think about something like this:
 
 tooltip
   titleAmarok/title
   icon src=amarok.png alt=Amarok/
   bodyCurrently playing: Qt 4 dance by Trolltech/body
 /tooltip
 
 With such a format, it would be possible to extract meaningful information
 for a11y purpose.
 
 Heck, this could even be somehow standardized to also work for
 notifications (/me looks at this thread slowly drowning into markup
 bikeshed)

title and icons are already outside the body in the spec and the title *can't* 
have html
th tooltip is a single proprty, that is a struct unfolded as this
icon (name or binary dump as usual)
title (short, plain text, must be meaningful and always present)
subtitle (optional, can have html)

if i read the titles and subtitles of all the icon tooltips i have in my 
systray i have
* KGpg - encryption tool
* Kopete: this one is bad since the detailed status info is in the subtitle as 
html, but reading the plain text should convey enough info like jabber: 
notm...@gmail.com (offline) MSN: notm...@gmail.com (offline) etc..
* clipboard contents   -  first line in the clipboard
* Volume at 70%
* Keyboard layout  - Italy
* KDE wallet  -  a wallet is open

all of them, seem to convey quite all the informations i need if are read out 
loud.
I would see maybe an optional field to be used instead of the tooltip *if not 
possibleto put enough info in the tooltip* rather the other way around.

Cheers,
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Marco Martin
On Monday 07 February 2011, Aurélien Gâteau wrote:
  On Thursday, February 3, 2011, Aurélien Gâteau wrote:
   There are ways to strongly suggest application developers to define
   such strings: for example outputing warnings on stderr when a KSNI
   goes live without having proper a11y properties set.
  
  catching and punishing sins (though a warning is hardly a punishment
  unless
 
 we
 
  email the devs every time it happens .. yes, i'm joking ;) is not a good
  approach compared to being able to prevent the sins in the first place.
  
  developers have every motivation to add tooltips and they usually do.
  it's also easily tested by them: hover the entry with their mouse. if we
  use that same data for a11y, we prevent the sins.
 
 Using the tooltip is a handy fallback, but it will most likely be less
 adapted to a blind reader, especially since it may contain complex HTML
 content (reminds me of another discussion...), making it much more work to
 read for a blind user than a simple, text-only, label.
 
 I suggest we add the property Ted is proposing and specify that tooltips
 will be used as a fallback, but strongly recommends devs to set

not the title, the tooltip title never has html.

i can see at limit another property that is used as fallback, instead of the 
tooltip title if is not possible for some reason  to encode the proper 
informtion in the title (thing that i higly doubt by the way) rather than 
having the tooltip as fallback of another property

Cheers, 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Aurélien Gâteau
  
  What would help is if the tooltip was using semantic markup, instead of a
  subset of html. Think about something like this:
  
  tooltip
titleAmarok/title
icon src=amarok.png alt=Amarok/
bodyCurrently playing: Qt 4 dance by Trolltech/body
  /tooltip
  
  With such a format, it would be possible to extract meaningful information
  for a11y purpose.
  
  Heck, this could even be somehow standardized to also work for
  notifications (/me looks at this thread slowly drowning into markup
  bikeshed)
 
 title and icons are already outside the body in the spec and the title can't 
 have html
 th tooltip is a single proprty, that is a struct unfolded as this
 icon (name or binary dump as usual)
 title (short, plain text, must be meaningful and always present)
 subtitle (optional, can have html)

Ohhh... my memory is failing me. Indeed, tooltips are actually not a single 
field. With this in mind, I agree using the tooltip title should be good 
enough. What do you think, Ted?

Aurélien
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Aaron J. Seigo
On Monday, February 7, 2011, Aurélien Gâteau wrote:
  On Thursday, February 3, 2011, Aurélien Gâteau wrote:
   There are ways to strongly suggest application developers to define
   such strings: for example outputing warnings on stderr when a KSNI
   goes live without having proper a11y properties set.
  
  catching and punishing sins (though a warning is hardly a punishment
  unless
 
 we
 
  email the devs every time it happens .. yes, i'm joking ;) is not a good
  approach compared to being able to prevent the sins in the first place.
  
  developers have every motivation to add tooltips and they usually do.
  it's also easily tested by them: hover the entry with their mouse. if we
  use that same data for a11y, we prevent the sins.
 
 Using the tooltip is a handy fallback, but it will most likely be less
 adapted to a blind reader, especially since it may contain complex HTML
 content (reminds me of another discussion...), making it much more work to
 read for a blind user than a simple, text-only, label.

so then let's fix the problem instead of creating new ones.

the title doesn't have html (notmart already noted this). no problem there, 
and that's a perfect (literally) candidate for the main what this is a11y 
label.

then let's limit the html supported in the content. this can be a social 
control, just as the a11y controls would be (except in this case they'd be 
something people actually see). and if we look at all the content we have 
right now the most complex uses are of tables for aligning rows. that vast 
majority of tooltips are just simple strings.

so we might need a way to define multiple nicely aligned rows for content. 
otherwise, i don't see the problem in practice.

 I suggest we add the property Ted is proposing and specify that tooltips
 will be used as a fallback,

which will lead to the tooltips being used as the fallback in almost all 
cases. result? more dbus data, more spec to implement, nearly no real world 
use of that overhead. sounds like a horrible idea to me.

 but strongly recommends devs to set
 appropriate a11y labels (and output warnings in KDE implementation when
 they are not set).

you can keep wishing that this will work, but it won't. how many years have we 
been telling devs not put double margins in their dialogs? and that's 
something that they can see on their own scree. a11y is 100% invisible to 
99.some number of 9s% of devs.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-07 Thread Aaron J. Seigo
On Monday, February 7, 2011, Aurélien Gâteau wrote:
   What would help is if the tooltip was using semantic markup, instead of
   a subset of html. Think about something like this:
   
   tooltip
   
 titleAmarok/title
 icon src=amarok.png alt=Amarok/
 bodyCurrently playing: Qt 4 dance by Trolltech/body
   
   /tooltip
   
   With such a format, it would be possible to extract meaningful
   information for a11y purpose.
   
   Heck, this could even be somehow standardized to also work for
   notifications (/me looks at this thread slowly drowning into markup
   bikeshed)
  
  title and icons are already outside the body in the spec and the title
  can't have html
  th tooltip is a single proprty, that is a struct unfolded as this
  icon (name or binary dump as usual)
  title (short, plain text, must be meaningful and always present)
  subtitle (optional, can have html)
 
 Ohhh... my memory is failing me. Indeed, tooltips are actually not a single
 field. With this in mind, I agree using the tooltip title should be good
 enough. What do you think, Ted?

i don't believe Ted is on the list. you'll need to CC him, i think.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Marco Martin
On Thursday 03 February 2011, Ted Gould wrote:
 Hello,
 
 For an intro see the mail with the (0/3) in the subject line.
 
 One of the projects we're doing is to try and make Unity more
 accessible.  To that end we'd like to add properties to add accessible
 labels for the icons that could be given to screen readers.  We're
 planning on issuing warnings if an icon is set without an accessible
 label, it would be great if the KDE object could do the same.  Here are
 the properties that we're thinking about: IconAccessibleLabel and
 AttentionAccessibleLabel.  They should be strings describing the
 icons/movies, localized and change with the icon.  An example would be
 Battery 45%.
 
   --Ted
 
 PS - I don't think that OverlayAccessibleLabel makes sense, but I'd love
 to hear people's comments on that.

This is a really good idea, let's analyze it.

Basically, what the normal icon, the attention icon and the overlay indicate 
is a status, someting descriptive about what's going on (let's say battery 
45%)

now, we have a text representation, and that is what's written in the tooltip. 
this may or may not be enough for a screenreader to represent it textually...

so, we could have a staus for normal and active, always exposed to the 
bus, or a status text that is always the current, that is updated from the 
application, (and also when the icon status changes from normal to 
requestingattention for instance)
so a TextualStatus property could be enough...

ideas?

Cheers,
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Ted Gould
On Thu, 2011-02-03 at 14:37 +0100, Marco Martin wrote:
 On Thursday 03 February 2011, Ted Gould wrote:
  One of the projects we're doing is to try and make Unity more
  accessible.  To that end we'd like to add properties to add accessible
  labels for the icons that could be given to screen readers.  We're
  planning on issuing warnings if an icon is set without an accessible
  label, it would be great if the KDE object could do the same.  Here are
  the properties that we're thinking about: IconAccessibleLabel and
  AttentionAccessibleLabel.  They should be strings describing the
  icons/movies, localized and change with the icon.  An example would be
  Battery 45%.
  
  --Ted
  
  PS - I don't think that OverlayAccessibleLabel makes sense, but I'd love
  to hear people's comments on that.
 
 This is a really good idea, let's analyze it.
 
 Basically, what the normal icon, the attention icon and the overlay indicate 
 is a status, someting descriptive about what's going on (let's say battery 
 45%)
 
 now, we have a text representation, and that is what's written in the 
 tooltip. 
 this may or may not be enough for a screenreader to represent it textually...

Yeah, in talking to the a11y folks they like tooltips, but they really
wanted another label as sometimes the tooltips don't describe the icon
enough for a blind user to know what it means.  Apparently screen
readers can distinguish and make tooltips additional information as
well.  Also, sometimes tooltips can be too complex for screen readers to
do a good job with them.

 so, we could have a staus for normal and active, always exposed to the 
 bus, or a status text that is always the current, that is updated from the 
 application, (and also when the icon status changes from normal to 
 requestingattention for instance)
 so a TextualStatus property could be enough...
 
 ideas?

I thought about this, and I think that in general, this works as well.
But it seemed to me that it didn't align with the spirit of the spec
in that if we're describing icons, we should describe the icons in the
way that they're represented on DBus.

--Ted



signature.asc
Description: This is a digitally signed message part
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Sebastian Kügler
Hey,

On Thursday, February 03, 2011 15:29:38 Ted Gould wrote:
 On Thu, 2011-02-03 at 14:37 +0100, Marco Martin wrote:
  On Thursday 03 February 2011, Ted Gould wrote:
   One of the projects we're doing is to try and make Unity more
   accessible.  To that end we'd like to add properties to add accessible
   labels for the icons that could be given to screen readers.  We're
   planning on issuing warnings if an icon is set without an accessible
   label, it would be great if the KDE object could do the same.  Here are
   the properties that we're thinking about: IconAccessibleLabel and
   AttentionAccessibleLabel.  They should be strings describing the
   icons/movies, localized and change with the icon.  An example would be
   Battery 45%.
   
 --Ted
   
   PS - I don't think that OverlayAccessibleLabel makes sense, but I'd
   love to hear people's comments on that.
  
  This is a really good idea, let's analyze it.
  
  Basically, what the normal icon, the attention icon and the overlay
  indicate is a status, someting descriptive about what's going on (let's
  say battery 45%)
  
  now, we have a text representation, and that is what's written in the
  tooltip. this may or may not be enough for a screenreader to represent
  it textually...
 
 Yeah, in talking to the a11y folks they like tooltips, but they really
 wanted another label as sometimes the tooltips don't describe the icon
 enough for a blind user to know what it means.  Apparently screen
 readers can distinguish and make tooltips additional information as
 well.  Also, sometimes tooltips can be too complex for screen readers to
 do a good job with them.

On the other hand, it requires application developers to add another string in 
their code to all UI elements -- something which might be easily forgotten by 
those that don't care a lot about a11y (I fear that this applies to a lot of 
developers). Falling back to the tooltip sounds like something sensible for 
those cases.

In any case, we should clearly document why it makes sense (justifies the 
extra work for developers and translators), so people actually do it. 
Additionally, some guidelines how to make those accessible labels useful, i.e. 
not just repeat the tooltip, but making it convey the needed information in a 
form suitable for screen readers.

The overhead of such a thing is non-trivial, but I can't really judge the 
importance to a11y, so the above are just some thoughts about this proposed 
property.

Cheers,
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Aurélien Gâteau
 On the other hand, it requires application developers to add another string 
in their code to all UI elements -- something which might be easily forgotten 
by those that don't care a lot about a11y (I fear that this applies to a lot 
of developers). Falling back to the tooltip sounds like something sensible for 
those cases.

There are ways to strongly suggest application developers to define such 
strings: for example outputing warnings on stderr when a KSNI goes live 
without having proper a11y properties set.

Aurélien
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Aaron J. Seigo
first, thanks for working with us on this Ted. it really helps keep things 
moving forward in a coordinate fashion, and makes our lives (and hopefully 
yours) a lot brighter. :)

oh, and you don't need to CC Marco and I, we're on the plasma-devel list :)

On Wednesday, February 2, 2011, Ted Gould wrote:
 One of the projects we're doing is to try and make Unity more
 accessible.  To that end we'd like to add properties to add accessible
 labels for the icons that could be given to screen readers.  We're
 planning on issuing warnings if an icon is set without an accessible
 label, it would be great if the KDE object could do the same.  Here are
 the properties that we're thinking about: IconAccessibleLabel and
 AttentionAccessibleLabel.  They should be strings describing the
 icons/movies, localized and change with the icon.  An example would be
 Battery 45%.

when would the AttentionAccessibleLabel be used (as opposed to the Icon one)? 
are they ever used simultaneously? is the Icon label a more-or-less static 
description fo the icon, and the Attention label is more information for when 
something exciting happens?

can we re-purpose the tooltip for this?

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Aaron J. Seigo
On Thursday, February 3, 2011, Aurélien Gâteau wrote:
 There are ways to strongly suggest application developers to define such
 strings: for example outputing warnings on stderr when a KSNI goes live
 without having proper a11y properties set.

catching and punishing sins (though a warning is hardly a punishment unless we 
email the devs every time it happens .. yes, i'm joking ;) is not a good 
approach compared to being able to prevent the sins in the first place.

developers have every motivation to add tooltips and they usually do. it's 
also easily tested by them: hover the entry with their mouse. if we use that 
same data for a11y, we prevent the sins.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Aaron J. Seigo
On Thursday, February 3, 2011, Sebastian Kügler wrote:
 On the other hand, it requires application developers to add another string
 in their code to all UI elements -- something which might be easily
 forgotten by those that don't care a lot about a11y (I fear that this
 applies to a lot of developers). Falling back to the tooltip sounds like
 something sensible for those cases.

the tooltip really ought to be the textul representation of the entry in the 
first place.

another benefit of re-using the tooltip is that it's easy for any developer 
(or user) to test: hover the entry with the mouse. a11y labels are going to be 
a lot harder for the average dev to test (well, plasmaengineexplorer makes it 
easy enough, but that's still a sort of backwoods for most developers)

 In any case, we should clearly document why it makes sense (justifies the
 extra work for developers and translators), so people actually do it.
 Additionally, some guidelines how to make those accessible labels useful,
 i.e. not just repeat the tooltip, but making it convey the needed
 information in a form suitable for screen readers.

+1

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Aaron J. Seigo
On Thursday, February 3, 2011, Ted Gould wrote:
 On Thu, 2011-02-03 at 14:37 +0100, Marco Martin wrote:
  On Thursday 03 February 2011, Ted Gould wrote:
   One of the projects we're doing is to try and make Unity more
   accessible.  To that end we'd like to add properties to add accessible
   labels for the icons that could be given to screen readers.  We're
   planning on issuing warnings if an icon is set without an accessible
   label, it would be great if the KDE object could do the same.  Here are
   the properties that we're thinking about: IconAccessibleLabel and
   AttentionAccessibleLabel.  They should be strings describing the
   icons/movies, localized and change with the icon.  An example would be
   Battery 45%.
   
 --Ted
   
   PS - I don't think that OverlayAccessibleLabel makes sense, but I'd
   love to hear people's comments on that.
  
  This is a really good idea, let's analyze it.
  
  Basically, what the normal icon, the attention icon and the overlay
  indicate is a status, someting descriptive about what's going on (let's
  say battery 45%)
  
  now, we have a text representation, and that is what's written in the
  tooltip. this may or may not be enough for a screenreader to represent
  it textually...
 
 Yeah, in talking to the a11y folks they like tooltips, but they really
 wanted another label as sometimes the tooltips don't describe the icon
 enough for a blind user to know what it means. 

this is true for traditional tooltips, but the ones in a SNI are not like 
those tooltips at all: they have titles, subtext, etc. they are supposed to, 
in an SNI, be fully descriptive. are the a11y folks aware of those differences 
between SNI tooltips and normal ones on  e.g. pushbuttons?

what this denotes to me is that maybe we need to revamp the documentation 
around tooltips (and as marco pointed out on irc earlier today, perhaps that 
was an unfortunate name to choose in the spec :/ ) so that this is very clear 
to app devs too.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Qt Development Frameworks


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


Re: New properties for StatusNotifierItem: Accessible Label (1/3)

2011-02-03 Thread Ted Gould
On Thu, 2011-02-03 at 16:24 +0100, Sebastian Kügler wrote:
 On the other hand, it requires application developers to add another
 string in their code to all UI elements -- something which might be
 easily forgotten by those that don't care a lot about a11y (I fear
 that this applies to a lot of developers). Falling back to the tooltip
 sounds like something sensible for those cases.
 
 In any case, we should clearly document why it makes sense (justifies
 the extra work for developers and translators), so people actually do
 it. Additionally, some guidelines how to make those accessible labels
 useful, i.e. not just repeat the tooltip, but making it convey the
 needed information in a form suitable for screen readers. 
 
 The overhead of such a thing is non-trivial, but I can't really judge
 the importance to a11y, so the above are just some thoughts about this
 proposed property.

I agree documenting the purpose and providing ways for developers to
realize when they're not given is key.  I think another thing that is
important is providing a documented path for adding them.  Then someone
who did use the strings could provide patches to projects which provided
inadequate descriptions.

--Ted



signature.asc
Description: This is a digitally signed message part
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel