Re: Is xAP Dying

[Kevin Hawkins <yahoogroupskh@xxxxxxxxxxxxxx>]

I replied on the xAP Auto list but essentially the published
specification are fine to develop against and are fully forward
compatible with the next xAP v1.3 release.

There are many suggestions for inclusion in future xAP protocol releases
and some are quite involved . One of the beauties of xAP is it's
simplicity and that it isn't a moving target. We don't want to issue new
specifications every month - or even every year and we must ensure
forward/backward compatibility.   Many feature suggestions are actually
simply higher protocol layers implemented using schema definitions
sitting on the existing protocol and so can be proposed /added as
desired anyway.

The introduction of sub addressing in xAP v1.2 and in particular it's
use within the xAP BSC schema raised a limitation issue where no more
than 253 endpoints could be used within one device.  We recognised the
need to increase this and hence we have extensively promoted that UID
format chnage that will be in xAP v1.3 on this list, and it has been
adopted by many developers who hit this limitation.

I agree however that not formally publishing an update to the xAP v1.2
spec is lamentable and I think we should now do so, even at the expense
of not including some features that are not yet formalised.

BSC v1.3 is current and works really well.  There are no intentions at
this time to release a xAP v1.4 (see adjacent post).

Also bear in mind that xAP is a created and maintained by enthusiasts in
their spare time, and these people release free software and tools for
others to benefit from.  xAP is a great protocol for providing the glue
between different devices and applications and has been absolutely vital
as the 'enabler' in my own system..  However conceptually it does take a
bit of getting to know and some time and effort invested in same.  What
I personally have realised is that most users just want their system to
be totally integrated but don't have the wherewithall to do this
themselves and so depend on others.   I personally don't have the time
to support this user base and so I have now taken a step back to focus
on my own system and needs alogside the hundreds of other commitments I
have in my life just now....

I welcome contact from anyone who wishes to get more involved as a xAP
evangelist and help to maintain our website, specifications, email and
developer contacts as well as people releasing their own xAP
applications.   Neil - perhaps you have some apps / embedded device
designs you can post too ?

xAP works really really well for me, it provided the methodology to
make everything in my home interact - and believe me I have a hugely
disparate range of devices, technologies and applications. But that has
created so many projects now sitting on my ToDo list that keeping my
head a little lower is the only way I'm going to make progress on my own
system here....

K




> Hi,
>
> My father is into model trains and he wanted to take his train layout
to the next level. I.e. Automated control of some Input/Output hardware.
Train control etc.
>
> I came across a great open source PC based model train interface
"JMRI"  http://jmri.sourceforge.net/ and
asked the guru's there if they could interface to the xAP protocol for my
I/O hardware.
> When I was passing them links to the xAP site, I noticed dates on the
main xAP specifications of 1st Dec 2002, BSC spec dates of July 2004 and
felt a bit embarrassed that the interface I was suggesting had stagnated
and died.
>
> Or have I got it all wrong and you just can't improve on perfection?
>
> Even small things such as updating of the vendor ID's does not seem to
be happening.
>
> The JMRI group, principally use Java but the link on the xAP
Frameworks page does not work http://patrick.lidstone.net/haweb/java.htm
> Regards,
>
> Neil Wrightson.
> N.W.Electronics
> ABN 76 768 513 867
> Embedded Controllers and Home Automation Products
> Skype : Neil_Wrightson
> Web     : www.nwe.net.au
>
>
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>
>



------------------------------------