The idea of the patch files in W3C/OpenDDR was always to allow custom
improvement of detection, e.g. if a large company has mostly 1600x900 desk
screen, that would go there. Providing a default common value out of the
box is a good idea, but it should be possible for those companies to
override it in *Patch.

Will try it should work, at least using SimpleDDR. For the other clients,
it also depends what loading mechanism you chose. If the source is inside a
JAR or from a remote service, then the only way to offer patching would be
a "chain" of sources, i.E. load the default from a service, but offer a
patch locally in the application's classpath or file system. Maybe
something to consider or put in the backlog...

Werner

On Thu, Jul 31, 2014 at 11:37 AM, Werner Keil <[email protected]> wrote:

> Sounds good, will have a look.
> Note, the "vendor" field defined by Core DDR is normally the DEVICE
> vendor, so except XBox or latest Nokia([?]) that would not be Microsoft.
> There is a separate ODDR introduced "os_vendor", that would work for
> Microsoft, but except in Apple's case the OS and device vendors are rarely
> the same. Another custom property is something like "marketing_name" that
> applies to e.g. some of the "Nexus" devices, where the actual device vendor
> is often ASUS, LG or Samsung. That's a grey zone, but also for Android at
> most Google could be considered OS_Vendor, not device vendor in most cases.
>
> Werner
>
> On Thu, Jul 31, 2014 at 2:42 AM, Reza <[email protected]>
> wrote:
>
>> I went ahead and moved the generic devices into the core device file:
>>
>>
>> http://svn.apache.org/viewvc/incubator/devicemap/trunk/data/device-data/src/main/resources/devicedata/DeviceDataSource.xml?r1=1614520&r2=1614797
>>
>>
>> So go ahead and make sure all the attributes look good and please feel
>> free to make changes as you see fit. Example, for a windowsDesktop, I put
>> the vendor down as Microsoft, not sure if that is correct. Once this is all
>> set, I will go ahead and link the patterns back to these devices.
>>
>> Also, for the screen resolution, I moved it back to 800x600. When I
>> thought about it, it might be better to have the value be the *MINIMUM*
>> resolution for a desktop just incase there is logic which decides on a
>> layout based on minimum width (I have seen this before). If you want to 0
>> it out, no biggie.
>>
>
>

Reply via email to