On Mon, Nov 1, 2010 at 7:25 AM, Henrik Rydberg <rydberg@xxxxxxxxxxx> wrote: > On 10/31/2010 05:00 AM, Ping Cheng wrote: > >> The "Multi-touch (MT) Protocol" says: >> >> "The slot protocol requires the use of the ABS_MT_TRACKING_ID, either >> provided by the hardware or computed from the raw data". >> >> We get the IDs from Bamboo touch device. Let's use them instead of >> compute our own. >> >> Signed-off-by: Ping Cheng <pingc@xxxxxxxxx> > > > Hi Ping, > > I share the same concern as Dmitry here. This patch weakens the semantics of the > current tracking id definition. As it seems, only to allow the tracking id to be > used in a different, undocumented or hardware-specific way. This is not a good > way to go - neither from a kernel maintenance perspective, nor from a user > experience perspective. After all, the kernel is about unifying the experience > of different hardware. I think I got the point now. The tracking id is aimed at an unified touch/finger tracking in the kernel. Not just to pass whatever hardware provides. I was stuck at the phrase of "provided by the hardware". With your explanation, I see the rationale behind the use of "trk_id". This discussion made it clear how I need to convince app developers. Please ignore this patch. I see a potential in merging type A into type B with the introduction of trk_id. The type A drivers may need to tweak their x/y/..values a bit to fit into the protocol. But the benefit of tracking ID should be universal. Trust me, this is all technical. Wacom does not even have a type A device. It is not my job to care about type A. >From a pure developer point of view, if our goal is to unify the experience of different hardwares, merging type A to type B may not be a bad idea. However, I do know there was issues preventing us from merging them when we started in June. I bring it up now to see if things has changed in our favor or not. Thank you Henrik for your reply and comments. It helped me a lot. I hope it does the same for others too. Ping -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
