RE: Proposal for device "enumeration"

> From: Martin Thomson [mailto:martin.thomson@gmail.com]
> 
> On 7 December 2012 09:43, Travis Leithead <travis.leithead@microsoft.com>
> wrote:
> > Instead of a deviceId, I want to use a deviceIndex
> 
> This would ensure that identifier stability exists only for the
> duration of a session.  How would you propose that removed/disabled
> devices be handled?

For the session, a removed device just creates a hole in the list of devices. Upon re-entering the next session, there would be no hole and some devices would have different indexes then they did before.

Disabled devices (e.g., user denied permission to it, or the device has an exclusive lock owned by another app), would still have an assigned index--you'd just never be able to connect the source at that index to a track.


> 
> > I think this is far easier to spec,
> 
> Yes.  But only marginally.
> 
> > less prone to fingerprinting,
> 
> No.  It's identical.

I say less because my exposure doesn't tie-in to any device-specific data to create the identifier. If the one-way hash you propose is ever compromised, then it might be possible to start to correlate specific devices. But other than that, yes, they are essentially the same.

> > and maintains almost all of the important use-cases.
> 
> That loses what I perceive to be the main advantage of having a
> persistent identifier: namely that I can configure an application to
> use the device that I prefer and that it will use that device each
> time that I return.

I understand this. I'm just not convinced that devices are going to change that often. Of all the devices in the world for consumers, how many of them use the plug-in media providing ones? I have one USB webcam in my office (I don't need two). And I have one scanner at home that I plug in occasionally. And for audio, it's the built-in mics and the input-line (for which different mics can be attached without actually being represented by different devices).

> I don't see there being any significant advantage to this other than
> the ability to control whether you get the same device or a different
> device when you ask a second time.  That's an improvement over status
> quo, but not an especially compelling one.

Received on Friday, 7 December 2012 18:31:10 UTC