Re: Changing definition of "Large text" to use px rather than pt

Additionally, worth pointing out: this exact problem (the inability of 
an author to have influence over the exact physical size as measured on 
screen of what they define in CSS etc) will also come around the same 
way if the Low Vision TF tries to define a baseline minimum font size. 
So having a shared understanding of what can actually be mandated in the 
SC (rather than running the danger of mandating something that simply 
can't be consistently achieved/tested by authors) is essential even 
here, or we'll be circling the same discussion at that point too 
(similarly, I've already been around this whole discussion in the mobile 
a11y TF when defining touch target sizes and adding qualifiers to any 
measurements there).

P

On 25/04/2016 18:05, Patrick H. Lauke wrote:
> Copying this to the list, rather than making an off-list discussion of
> this:
>
> On 25/04/2016 17:11, Wayne Dick wrote:
>> Dear Patrick,
>> I do understand the current need to use relative font sizes, but
>> legibility cannot be addressed with relative sizes. For normal readers
>> you need to fit between 7-15 letters in the foveal region of the retina.
>> For readers with partial signt you usually need much higher settings to
>> make up for not using the fovea effectively. So to get projections of
>> that size on the retina from 40cm you need specific physical sizes.
>>
>> The contrast settings are based on visual acuity measures that influence
>> an individual's contrast sensitivity. Again this requires specific
>> angles to subtend the retina.
>>
>> So, I our next go around we do need to include physical size of the
>> print or our contrast ratios are meaningless. We may conclude that for
>> certain screen sizes and dpi there is an acuity range necessary for
>> visual access.
>
> My point is that there is no way that an author can know, or even set,
> the physical size that any measurement they set (in their CSS, whether
> using px or pt or mm or any other unit of measure) will actually render
> as in physical sizes.
>
> So we need to anchor this somewhere.
>
> The assumption needs to be made that the default user agent base font
> size is baseline readable (otherwise the user will need to use a
> different device, change user agent, change their OS). If that is not
> the case, and the user has a device where 16px (the common default user
> agent font size) is not legible, then that cannot be the concern of the
> author (as the author has no way of knowing this, nor influencing this
> only for that particular user).
>
>> Here are some guidelines. Depending on the language, words have an
>> average length of from 10 to 15 letters with longer words occuring about
>> 2% of the time. A device should fit most words on one line for genuine
>> readability. Thus if you read English and need letters of 1.25cm (36pt)
>> letters a you need at least a screen width of 12.5cm (5in). That means a
>> diagonal of 25cm (10in)
>
> This does not take into consideration user agent or OS settings for
> zoom, screen size, dpi adjustments that are made automatically, or
> particular viewport settings specified by the page and how the UA/OS
> react to them.  You can't have an SC that
>
>> My mobile phone is 15cm diagonal and 7.5 on its portrait width. I can
>> use it for visual reading some. I cannot read most web pages visually.
>
> This is anecdotally interesting, but it would be more helpful to see
> what the exact font size of those pages that you can't read is set to,
> and whether or not those pages set the browser into using the ideal
> viewport (with viewport meta of width=device-width).
>
> P


-- 
Patrick H. Lauke

www.splintered.co.uk | https://github.com/patrickhlauke
http://flickr.com/photos/redux/ | http://redux.deviantart.com
twitter: @patrick_h_lauke | skype: patrick_h_lauke

Received on Monday, 25 April 2016 17:14:55 UTC