RE: Re: xPL announcement/description protocol -- was xPLDiag

["Ian Lowe" <ian.lowe@xxxxxxxxxxxxxxxxxx>]

My thoughts exactly - what Gerry's post last night has confirmed is, I
feel, a need for clarification, and firming up what we mean by "the
xPL
environment".

I'm digging through a bunch of older emails at the moment to see what
other overhanging niggles there were, see if we can't perhaps do all
this as some sort of roundup document, and update the docs to reflect
the bigger picture rather than just the component parts.

I.


-----Original Message-----
From: ukha_xpl@xxxxxxx [mailto:ukha_xpl@xxxxxxx] On
Behalf Of Tom Van den Panhuyzen
Sent: 25 September 2005 11:07
To: ukha_xpl@xxxxxxx
Subject: Re: [ukha_xpl] Re: xPL announcement/description protocol -- was
xPLDiag

I 'll add my typical sloganesk reply ;-)

How many people are running an xPL network without brains like HAL or
the scripting in xPL4Java ?  I bet exactly 0.

Which xPL application/device needs a device discovery protocol ?  Only
the brains.

The brains are an essential but different part of the xPL network.  I
fail to see why this component would not be portable.  It talks xPL at
one end and XHCP at the other.  There is no OS dependency.
It shouldn't be talking xPL on both sides because the XHCP-talk doesn't
concern the apps listening on the xPL network.

If you accept that the brains keep a list of devices then the detection
latency is a non-issue because it is solved using rapid hbeats at
startup.

Cheers
Tom