Based on this thread and the lack of any technical reasons presented for not 
following the proposed plan,
the decision is to go again as planned.  I will ask CJ to remove Arduino from 
the regular jenkins CI builds.

And yes we will make sure marketing is such that the iotivity suite of stuff 
(iotivity & iotivity-constrained) does
support Arduino, via the iotivity-constrained project, for which Arduino 
*shall* be part of jenkins CI builds.

Dave Thaler
IoTivity Platform Support Maintainer

-----Original Message-----
From: Thiago Macieira [mailto:[email protected]] 
Sent: Wednesday, March 1, 2017 4:26 PM
To: iotivity-dev at lists.iotivity.org; Dave Thaler <dthaler at microsoft.com>
Cc: Ashok Babu Channa <ashok.channa at samsung.com>
Subject: Re: [dev] Should IoTivity Arduino support be dropped?

On domingo, 26 de fevereiro de 2017 22:34:12 PST Ashok Babu Channa wrote:
> Yes, I support the dropping of IoTivity for Arduino but not immediately.
> 
> "Iotivity does not support Arduino" sends the wrong message.
> n  Yes., until we have some constrained releases, better we keep it.

I'm of a different opinion: I don't think Arduino is a sensible platform for 
real devices. Everyone I've asked that is serious about making devices tells me 
that Arduino can't be taken past prototype stage. Instead, for real uses. 
they need real OSes with good WiFi, 6LoWPAN and/or Bluetooth, plus security 
functionality, and that's the kind of OS that IoTivity-constrained targets. 
Therefore, it makes little sense to even prototype with Arduino.

As for the lack of release, that's easily solved: we can make a 0.x release at 
any time. I feel that a 1.0 makes no sense so long as OCF certification 
requires keeping the radio on all the time -- no one will make constrained 
devices like that (this is a topic I will bring up inside OCF next week).

> We can find sub-maintainer till then if that is a major problem.

Point them to IoTivity-constrained. If there is sufficient interest and 
motivation, an Arduino port can be revived there. In fact, it makes more sense 
to be there anyway.

So my suggestion is to do as Dave proposed: drop Arduino support immediately 
from the 1.3 releases, remove the #ifdefs as time goes by. And going further: 
remove the #ifdefs that are in the way and clutter the code base ASAP (we have 
quite a few of those).

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

Reply via email to