W3C home > Mailing lists > Public > www-style@w3.org > March 2013

Re: [css3-values] recommendation for pixel units

From: Simon Sapin <simon.sapin@exyr.org>
Date: Wed, 20 Mar 2013 21:55:29 +0100
Message-ID: <514A2241.8040604@exyr.org>
To: James Holderness <j4_james@hotmail.com>
CC: www-style <www-style@w3.org>
Le 20/03/2013 21:08, James Holderness a écrit :
> In section 5.2, with regard to lower-resolution devices, and devices with
> unusual viewing distances (which I assume would include most mobile
> devices), it is recommended that the pixel unit refer to the whole number of
> device pixels that best approximates the reference pixel.
>
> I take that to mean it is recommending that the device-pixel-ratio should
> always be a whole number. Am I reading that correctly?
>
> If so, is this perhaps worth reconsidering, given that it's not uncommon
> these days for mobile devices to default to a device-pixel-ratio of 1.5 or
> 0.75?
>
> I realise this text has been in the spec since the beginning of time, but if
> implementers don't think it makes sense anymore, and are just going to
> ignore it, is there any point in continuing to make the recommendation?

AFAICT, the only requirement in the whole section is the ratio between 
the different units. Implementations can anchors the units however they 
like and still be conforming.

Spec text:
> For such devices it is recommended that the pixel unit refer to the
> whole number of device pixels that best approximates the reference
> pixel.

I think that the intent of this sentence is:

1. It is recommended to approximate the reference pixel
2. It is acceptable to deviate, for example if an integer ratio (or a 
multiple of 0.25) makes implementation easier or gives better visual 
results or for whatever reason. Viewing distances on a given display 
vary anyway.

I don’t think there is an intrinsic value to an integer ratio. That’s 
really up to implementations.

Still, I suppose the wording could be improved a bit.

-- 
Simon Sapin
Received on Wednesday, 20 March 2013 20:56:02 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:21:07 GMT