[whatwg] RWD Heaven: if browsers reported device capabilities in a request header

Matthew Wilcox mail at matthewwilcox.com
Tue Feb 7 08:58:33 PST 2012


I don't agree that it's a good idea to have to have the server query some
massive device database in order to find out what the requesting device
does and does not support. Device databases are, by their nature, going to
be hard to maintain, hard to support, and not tell us all we need to
know. How is a device database going to tell us the current connection
speed, or the user's zoom level, or device orientation, or anything else
dynamic that may be relevant to know on the server?

The database route is merely an extension of UA sniffing, which was about
detecting devices and assuming capabilities and settings. The problem with
that was inaccuracy, and ongoing laboured maintenance. I want us to detect
the actual capabilities directly - exactly how CSS works. Detect what's
there. Skip the middle-man. How does it make sense to resort to siloing
this information in a database, when we're the body that makes the
standards that would allow direct knowledge?

I'm less interested in current work-arounds (and that's what those are,
workarounds and best guesses destined to fail at some point) and more
interested in getting accurate, reliable, future-proof data exposed.


On 7 February 2012 16:46, Charles McCathieNevile <chaals at opera.com> wrote:

> On Tue, 07 Feb 2012 15:13:03 +0100, Matthew Wilcox <mail at matthewwilcox.com>
> wrote:
>
>  Personally, I think the issue of adapting resources to client
>> capabilities is here to stay.
>>
>
> For sure, although the mechanisms might evolve.
>
>
>  Devices of significantly varied size and performance are here to stay,
>>
>
> Yes...
>
>
>  and adapting images to suit is only one example of smart resource
>> adaption - it doesn't seem like it would become irrelevant until all
>> devices are memory and processor monsters over high speed connections.
>>
>
> Actually I suspect the end condition is "until memory and processing and
> bandwidth are all free" which is slightly different.
>
>
>  Which is going to take decades, if it ever happens.
>>
>
> I will be surprised if I see it.
>
> Anyway, I would really encourage you to look at
> http://en.wikipedia.org/wiki/**Device_Description_Repository<http://en.wikipedia.org/wiki/Device_Description_Repository>andperhaps something like
> http://www.openddr.org/ or the commercial solutions that deal with this
> area.
>
> cheers
>
>
> --
> Charles 'chaals' McCathieNevile  Opera Software, Standards Group
>    je parle français -- hablo español -- jeg kan litt norsk
> http://my.opera.com/chaals       Try Opera: http://www.opera.com
>



More information about the whatwg mailing list