On Sep 29, 2016 3:56 PM, "Gregg Reynolds" <dev at mobileink.com> wrote:
>
> On Sep 29, 2016 3:41 PM, "Thiago Macieira" <thiago.macieira at intel.com>
wrote:
> >
> > On quinta-feira, 29 de setembro de 2016 14:56:19 PDT Gregg Reynolds
wrote:
> > > What about a single physical device that hosts multiple OIC devices,
plus an
> > > on-boarding tool? Does unowned-device discovery need this?
> >
> > This is supposed to affect only queries originating from the same
process.
> > Queries originating from other processes on the same machine would
still be
> > processed.
> >
> my understanding is that a single process could host both an on-boarding
tool and an Iotivity client/server.
>
> so I guess my question is whether the proposed change would require more
extensive changes in order to support such a scenario. (fwiw, I'm still
struggling to understand the structure of ownership and security
provisioning.)
>
(thinking out loud, in search of enlightenment) isn't "process" kind of a
loaded term? insofar as it depends on OS,etc., I mean. what if there is
no OS? even a single-processor, non-preemptive system can do cooperative
coroutines. which could be thought of as coprocesses, and could do
networking separately. no? maybe not practically relevant, but fun to
think about.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.iotivity.org/pipermail/iotivity-dev/attachments/20160929/393868bb/attachment.html>
