I'll make one public comment on this.
OpenHAB is the exact opposite of what I am thinking of. I would describe
OpenHAB as "You will write in Java, you will use our class structure,
you will limit yourself to the interactions that we think are important
and you will like it. We do." You can certainly build something this way
and get to the feature adding stage sooner, but you also create all
sorts of walls to run into. I have too many dents on my head from trying
to get these kinds of systems to dance anything other than the tune they
came with.
My house is way to complex for this kind of model, at both the
abstraction and interaction level. My house is designed to allow a
control system to run the system in as close to steady state as
possible, with variable source and distribution units. So "turn it on"
is a less common action than increase this flow or output to 42%.
I also have independent heating and cooling systems with independent and
overlapping zoning. Then again, zone is something a system implementer
thinks about, occupants really don't know or want to know about these
things. I'm here and I'm too hot is their level.
Now imagine trying to predict the solar heating rate from weather
reports/measurements and preheat/precool areas to at least reduce
massive errors. With radiant heat, these decisions are made in time
scale of fractions of an hour in advance.
The interactions need to be defined by the components in the particular
system and the service agreements between messaging agents that present
those components. My goal is to be able to have agents that implement
control algorithms / architectures and allow those to interact with real
world systems.
jerry
On 02/23/2016 11:16 AM, Matthias Urlichs wrote:
> Arley Carter <arcart...@bellsouth.net> writes:
>
>> Have you looked at the work done by the OPC Consortium?
> Not yet.
>
> [three hours later] O … K … Skip ahead if you don't want to read me ranting.
>
> If you want something that's *way* overspecified (I do not want to implement a
> structured version of SQL queries – among other interesting things – nor do I
> see any need to do so) and impossible to implement without (a) access to the
> reference implementation and its test cases and (b) at least a man year for
> the basics (the Java server example is 430 lines -- more than 70(!) import
> statements, way too few comments, and a heap of empty handlers most people
> would find to be somewhat essential), be my guest.
>
> This eerily reminds me of the X.400 train wreck of the 1980s.
>
> In any case, this is the OWFS list. I can guarantee that nobody in their right
> mind would ever connect an 1820 temperature sensor to such a system.
>
>> They are rather
>> far down the track in the direction you are headed.
> They also want $3000 for access to their C source code. Redistribution is not
> permitted, sorry, so any open source implementation will have to be done from
> scratch.
>
>> Why replow fields and reinvent wheels?
> Look at how much FHEM or OpenHAB can do. Ultimately I would like to build [the
> foundation of] something that's as capable, but somewhat more reliable (no
> single point of failure) and accessible ("if X, do Y" should be three lines of
> debuggable Python/Perl/whatever, or three Node:Red clicks).
>
> In any case I am no re-inventing any wheels. The wheels are out there, I can
> learn how they're built, and decide to use them in a vehicle that is easier to
> driver than what's already out there and that'll stay on track if one of the
> wheels gets loose.
>
>> Site24x7 APM Insight: [yeah, right]
> Adding commercials to emails is not a particularly good idea IMHO.
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> Owfs-developers mailing list
> Owfs-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/owfs-developers
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
