Re: Features and fixes incompatible with backward compatibility

On 2/20/2012 11:52 AM, Brad Kemper wrote:
> On Feb 20, 2012, at 11:14 AM, Charles Pritchard<chuck@jumis.com>  wrote:
>
>> On Feb 20, 2012, at 9:26 AM, Brad Kemper<brad.kemper@gmail.com>  wrote:
>>
>>>> Maybe if we had only "px" and "truemm" (with the latter very rarely used), things would be better because I'd be less frequently accused of destroying the metric system :-).
>>> I think anyone who would have typed mm or truemm and expected accurate millimeters in anything other than print would still be very disappointed most of the time. Not because of incorrect implementations, but because there are so many reasons why the display technology wouldn't or couldn't set that measurement accurately.
>>
>> What is an acceptable success ratio?
>>
>> I'm pretty sure we can get 100% of every iPhone and iPad when viewed on the screen.
> How do you figure? Web pages on iOS are normally zoomed to something other than 100%, with no user controlled way to show at exactly 100%, so most of the time, truemm would not be accurate.

This is a niche feature. Web app authors disable zoom when it suits 
them. The general web won't be on the truemm train. Double-tap zooms 
back to normal.

I was referring to the nature of the iOS devices: they are very much 
standardized. This is in response to your concern that: "display 
technology wouldn't or couldn't set that measurement accurately".


>> If it works for 95%, is that good enough? Is any number good enough?
> I don't think it would be good enough for general use, and if introduced it would be abused. It would mean that putting a web page onto a jumbotron would result in unreadably small text, and text on an eyeglasses display would be unreadably large. And 95% is a pipe dream, even for desktop displays.

Eyeglasses and jumbotron are a dead horse at this point; or a red 
herring. I don't know. We've got media queries to adjust for jumbotron 
and eye glasses.

If someone is programming their jumbotron and wants to say that they are 
looking for text to be 2 meters high, well it'd be swell if they could 
just write that.
span { font-size: 2000truemm; }

This feature isn't intended for general use, IMO. It's intended for 
niche uses. I think a jumbotron is a niche use; so maybe that is an 
actual use case.
In the meantime, the "ruler-widget" use case still stands. The @print 
use case still stands.

And the "will be abused" sentiment is just the way of the world... It's 
not a security risk. It's a usability risk. And it's well within normal 
range. If authors are neglecting WCAG, that's just a shame, but it's not 
a deal breaker.

Received on Monday, 20 February 2012 20:13:50 UTC