Em quinta-feira, 2 de mar?o de 2017, ?s 09:58:54 PST, Christian Gran escreveu:
> Hi,
> 
> did we consider that Arduino support got broken, because the stack with
> security (which is mandatory in my mind) is just to large now for that
> device?

Yes, we knew it was happening and moved on anyway. Security is not optional. 
So if the device couldn't support it, the device got dropped.

Later, IoTivity-constrained came about and showed that it is able to support a 
full OCF stack with security in less RAM and Flash than what IoTivity-full's 
CSDK could. But no one has ressurrected Arduino support in IoTivity-
constrained.

> How can we, ignoring that, just move on and try to use other small
> devices? Isn?t this a proof of concept - which now fails?

We cannot. It's been proven that devices that small won't work with IoTivity-
full. That's why we created IoTivity-constrained.

> Why did it break?
> Are we sure the stack can still fit?

IoTivity-constrained can. The whole point of this discussion is whether we 
should even try to get the full one to fit.

My reasoning is that it's a wasted effort to try. Just go with Constrained 
intead.

> I think it is easy to say - it is broken - ok - lets remove that one and
> look instead for other small devices (thin ice in my mind)  ,-)

With the right tools.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center

Reply via email to