W3C home > Mailing lists > Public > www-style@w3.org > March 2015

Re: [mediaqueries] Use-case for "viewing context" MQ

From: Florian Rivoal <florian@rivoal.net>
Date: Thu, 26 Mar 2015 17:44:24 +0100
Cc: Brad Kemper <brad.kemper@gmail.com>, www-style list <www-style@w3.org>
Message-Id: <096A4ADA-5AEC-4DA7-9352-5AB9552A98D2@rivoal.net>
To: "Tab Atkins Jr." <jackalmage@gmail.com>
> 
> On 26 Mar 2015, at 17:14, Tab Atkins Jr. <jackalmage@gmail.com> wrote:
> 
> On Thu, Mar 26, 2015 at 7:53 AM, Brad Kemper <brad.kemper@gmail.com> wrote:
>>> On Mar 26, 2015, at 1:52 AM, Florian Rivoal <florian@rivoal.net> wrote:
>>> Given that this is quite strongly tied to the nature of the device, this may
>>> be something we can reasonably expect a device vendor to be able to expose.
>>> 
>>> WBut I'm fairly reluctant about introducing this while:
>>> * TVs are the dominant lean back device on the market
>>> * Most TVs have a broken (way too large viewport)
>>> 
>>> Under these circumstances, introducing the (lean:back) media query is
>>> almost certain to be used by authors as a TV media type which they will
>>> then use use to show bigger buttons instead of relying on the css px doing
>>> its job (and screaming at TV vendors to fix their viewport sizes).
>> 
>> Can we agree that ideally, TV manufacturers should include a user changeable setting for dppx, to correspond to viewing distance? So, if the user says the viewing distance is 10 feet, then the TV will report having 2dppx, and the screen dimensions would be half what they would be if the user had the HDTV on a desktop (or turned sideways as an advertising display in a smaller room).
> 
> Indeed.

Agreed, and while the user has not provided this information, they should default to a typical viewing distance (which isn't a nominal arm’s length of 28 inches on a TV).

>> Maybe we should publish a technical note as guidance to TV manufacturers, and reference the definition of 'px' (in CSS Values and Units), as well as MQ and device adaptation specs.
>> 
>> I'm not convinced that 'lean: [ forward | back ]' descriptors would be valuable, or that authors would actually use them much on the Web.
> 
> I mean, I just provided a pointer to some use-cases.  I'm unsure if
> it's doable or worthwhile either, but hey, there are some reasons to
> do it.

I don't know either, but I agree that given the use cases, it's worth at least mentioning.

Even if it turns out to be a good idea, I am quite reluctant to put it out while TV browsers mismanage the size of their css px. Under the circumstance, a 'lean: forward | back' MQ is more likely to be abused than used, meaning that if the TV vendors ever get their px fixed, 'lean' would be unusable for its initial purpose due to legacy content / legacy UAs.

 - Florian
Received on Thursday, 26 March 2015 16:44:57 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:30 UTC