Protocol Design

[Will Pearson]

I'm going to suggest we go with a json style protocol. That is key
value pairs in curly braces. So a message to a buzzer specifying a new
program for it to run might look like.

{
deviceid: "Dev001"
newprogram: "OUT:LED1:1; WAITM; IF M=1; OUT:LED2:0; ELSE; OUT:LED2:1; ENDIF"
}

This also illustrates the programming language I was thinking would be
needed for the buzzer. Some of the commands

WAITM: Waits for a message from the teacher. Gets stored in the variable M
WAITB: Waits for a local button press. Gets stored in the variable B

I'm going to institute the 48 rule on this!

Will

[Charles Yarnold]

Again, is this for the physical buzzer system? This seems to high for the microcontroller system, but possibly ok for a deamon attached to a serial data in/out from the electronics hardware.

Sol

[Will Pearson]

On 4 April 2011 00:17, Charles Yarnold <charles...@gmail.com> wrote:
> Again, is this for the physical buzzer system? This seems to high for the
> microcontroller system, but possibly ok for a deamon attached to a serial
> data in/out from the electronics hardware.
> Sol

It was for the physical system (or at least sending to the teacher
hub) that then decides how to pass it on to the buzzers themselves.

Are you envisaging simply sending messages like buzzer X turn on light
Y? This would increase the wireless traffic a lot. Where should we be
putting the brains of the buzzer system?

I'm not really bothered about the eventual answer, but I would like to
know as it affects the design of the system.

Will

[Charles Yarnold]

Humm, may it be worth having a chat on Tuesday evening then transcribing to the mailing list?

Sol

[Will Pearson]

Can do.

On 4 Apr 2011 16:29, "Charles Yarnold" <charles...@gmail.com> wrote:

On 4 April 2011 11:02, Will Pearson <wil.p...@gmail.com> wrote:
>

> On 4 April 2011 00:17, Charle...