xAP Talk Transcript 2007 02 27

From XAP Automation
Revision as of 23:28, 27 February 2007 by Jamest (Talk | contribs)

(diff) ←Older revision | view current revision (diff) | Newer revision→ (diff)
Jump to: navigation, search

27/02/2007 14:49:55 * quit (timeout)
27/02/2007 14:51:26 ‹KevinH› .
27/02/2007 14:55:56 * James joins My room
27/02/2007 14:56:05 ‹James› .
27/02/2007 15:00:32 ‹James› .
27/02/2007 15:04:26 * Gregg joins My room
27/02/2007 15:04:38 ‹Gregg› Hi J/K
27/02/2007 15:05:04 ‹James› hi gregg
27/02/2007 15:05:45 ‹James› K's having food at the mo
27/02/2007 15:06:06 ‹James› i'm still liking this send a xAP, phone rings and it talks..
27/02/2007 15:06:39 ‹Gregg› Soon, it will talk more "politely" (or at least understandable)
27/02/2007 15:06:50 ‹James› :-)
27/02/2007 15:07:19 ‹Gregg› Maybe I can apply the same politeness filter to mailing lists ;)
27/02/2007 15:07:44 ‹James› no comment
27/02/2007 15:07:50 ‹KevinH› I'm still sat with a rumbling stomach - but will have to go shortly
27/02/2007 15:07:57 ‹KevinH› Hi G
27/02/2007 15:08:05 ‹Gregg› Hi K
27/02/2007 15:09:03 ‹Gregg› You appear to be quite busy w/ the opnode event manager
27/02/2007 15:11:41 ‹KevinH› I am quite impressed with it really
27/02/2007 15:12:29 ‹Gregg› Yes--I hope to soon quit "playing/testing" and begin integrating it into my setup
27/02/2007 15:12:32 ‹KevinH› I've just been focussing a little on that side because I have very few 1-wire devices to really contribute much and little xPL interest
27/02/2007 15:13:11 ‹KevinH› Also I feel that the event manager will be as attractive to people as teh 1-wire stuff so if we can get them both in there then there is more apeal at the price
27/02/2007 15:13:57 ‹KevinH› I have moved a few thinsg over to embedded - end to end - now which is fun
27/02/2007 15:14:58 ‹Gregg› agreed; I'm thinking of how to integrate functionality for monitoring and controlling (essentially fail-safes) of my reef tank--which I've never yet allowed a PC to manage
27/02/2007 15:15:03 ‹James› did you ever get you meteor rabbited?
27/02/2007 15:15:14 ‹KevinH› It has surprised me though - how simple BSC is yet in actuality how many things crop up
27/02/2007 15:15:44 ‹KevinH› Yes I have my Meteor on a Rabbit :-)
27/02/2007 15:16:00 ‹KevinH› But not all the schema supported yet
27/02/2007 15:16:16 ‹KevinH› (just space) as it sits alongside C-Bus
27/02/2007 15:17:29 ‹KevinH› For G's benefit teh Meteor is a Uk CID device (RS232) and teh Rabbit is the embedded hardware platform I use
27/02/2007 15:17:35 ‹James› sapce and the rabbits. seems a never ending issue
27/02/2007 15:17:55 ‹Gregg› yep--actually knew that although I've never seen either.
27/02/2007 15:18:24 ‹Gregg› There's a netcallerid device (US) that seems to be comparable to the Meteor
27/02/2007 15:18:35 ‹KevinH› I do envy Daniel with all the space he has in his device - but I've got more serial ports and I/O !
27/02/2007 15:18:51 ‹KevinH› Yes - and there's a xAP conduit for it ...
27/02/2007 15:19:57 ‹KevinH› Written in VB.Net by Glenn Sullivan. I exchanged emails with him yesterday and he's happy for me to pass the source to Ian D I think who wanted a VB.Net xFx example
27/02/2007 15:20:36 ‹James› could i see a copy too?
27/02/2007 15:20:41 ‹Gregg› ahh... good to know. Mine is connected to mh and I export CID info for SB to process
27/02/2007 15:22:33 ‹KevinH› I'll send it J .. just looking 4 it
27/02/2007 15:22:45 ‹James› ta
27/02/2007 15:23:29 ‹KevinH› It was his first VB.Net app i believe so not necessarily a reference way to do things but i believe it works
27/02/2007 15:23:59 ‹James› a start though
27/02/2007 15:24:02 ‹Gregg› back to the opnode discussion... I hope Daniel works on his "supernode" controller that would be scriptable
27/02/2007 15:27:47 ‹KevinH› I spoke with him re this and I keep steering him in that direction but he seems keen on opn232 next with X10 and then ADI Ocelot support - then maybe opnsms (which I assume is text messaging) and then maybe the controller
27/02/2007 15:28:55 ‹KevinH› I always feel I'm pushing him a bit in directions he maybe hadn't considered in the current beta discussions
27/02/2007 15:29:16 ‹KevinH› - and sort of away from 1-wire ... hmmm
27/02/2007 15:29:51 ‹KevinH› did those column classes work J ??
27/02/2007 15:29:56 ‹Gregg› The additions to opnone, have (IMO) made opnone more useful. So, it's hard to say it's non-1-wire; just not protocol specific
27/02/2007 15:30:18 ‹KevinH› I think they will be good grounding for later stuff too...
27/02/2007 15:30:39 ‹KevinH› Sorry - got to go and eat - I'll be back later....
27/02/2007 15:33:32 ‹James› so gregg are there any more extras in the works for the axc?
27/02/2007 15:33:39 ‹James› (just curious)
27/02/2007 15:34:03 ‹Gregg› Maybe ;) I
27/02/2007 15:34:14 ‹Gregg› I'm not overly fond of the call-plan integration
27/02/2007 15:34:38 ‹James› as in teh IVR bit?
27/02/2007 15:34:41 ‹Gregg› I'm thinking of moving some of the call-plan extension logic into an AGI
27/02/2007 15:35:01 ‹Gregg› really--both--as they rely on some common voice, TTS stuff as well
27/02/2007 15:35:31 ‹Gregg› Also, I've been toying w/ composing the IVR menus in xml--which "should" be "simpler" to create menus
27/02/2007 15:36:15 ‹Gregg› In other words, what's there now is ok for "proof of principal" and just general "playing"; but, I'm not thinking it's as usefal/maleable as I would like to see
27/02/2007 15:36:18 ‹James› i don't know how it works but a freepbx module for configuring would be a real bonus ;-)
27/02/2007 15:36:52 ‹Gregg› Also, I want to make the unattended dialout more robust w/ failover extensions/channels/numbers
27/02/2007 15:37:28 ‹James› as teh basic answer is there are quite a few extras planned then!
27/02/2007 15:37:55 ‹Gregg› More of the "polish" variety rather than fundamentally new--which is what has been coming out recently
27/02/2007 15:38:23 ‹Gregg› I'm also not too sure what the "barriers to adoption" might be. Definitely, I need to improve my start at the wiki docs
27/02/2007 15:39:12 ‹James› I do think docs based on a trixbox install. maybe even an install script would be good
27/02/2007 15:39:38 ‹Gregg› I think I will need to rely on some other kind soul for the trixbox info ;)
27/02/2007 15:40:16 ‹James› there's always that nice VMplayer image available ;-)
27/02/2007 15:40:34 ‹James› but i can document asa i go
27/02/2007 15:41:04 ‹Gregg› yes--although my general rule of thumb is that I only try to support what I run in production and use on a routine basis
27/02/2007 15:41:57 ‹James› that's a pretty good thumb
27/02/2007 15:42:43 ‹Gregg› actually, I really need to start making progress on the insteon gateway as that is my more personal and immediate interest
27/02/2007 15:43:21 ‹James› so all your insteon is on the workbench/boxed
27/02/2007 15:43:59 ‹Gregg› All Insteon gear is "dual-mode" -- where the other mode is x10. So, no--it's in production, but the Insteon part is not being used
27/02/2007 15:44:37 ‹James› k, that quite a nice feature having the old x10 functionality there
27/02/2007 15:44:46 ‹Gregg› And, although the US x10 extended codes make fancy lighting work ok, it's still unreliable and slow x10
27/02/2007 15:45:12 ‹KevinH› I'm back ( as it appears the oven temp setting has defeated SWMBO)
27/02/2007 15:45:18 ‹Gregg› unfortunately, it allows for being slow to adopt as it does work
27/02/2007 15:45:34 ‹James› i've never understood why cm12u x10 commands take so much longer than an x10 switch.
27/02/2007 15:46:13 ‹KevinH› above reminds me on the lighting schema - which currently I implement as extensions to BSC
27/02/2007 15:46:40 ‹Gregg› yep--and my plan is to continue that w/ the insteon gateway where possible
27/02/2007 15:46:53 ‹Gregg› for one thing, mh already has plenty of BSC support built in
27/02/2007 15:47:23 ‹James› i wonder who put that in there
27/02/2007 15:48:16 ‹KevinH› Greg noted your comment on XLML for menu representation ... this could be applicable to a new 'mode' type device too
27/02/2007 15:48:23 ‹KevinH› XML
27/02/2007 15:48:46 ‹Gregg› So, do you have specific ideas on an XSD?
27/02/2007 15:48:48 ‹KevinH› A device that has a number of valid states base don its current one
27/02/2007 15:48:59 ‹Gregg› or, even non-specific, napkin-scrawlings
27/02/2007 15:49:00 ‹KevinH› Not yet...
27/02/2007 15:49:42 ‹KevinH› MY XML experience really is justa couple of books I've read
27/02/2007 15:49:51 ‹Gregg› I may try to hunt for some; most of my XSD knowledge is in the are of elearning
27/02/2007 15:49:57 ‹KevinH› I have never used it as such
27/02/2007 15:50:10 ‹KevinH› beyond very basic stuff
27/02/2007 15:50:13 ‹Gregg› oh--well, the apps that I build are very XML/XSLT centric.
27/02/2007 15:50:56 ‹Gregg› It would be good if there was consistency in the structure though as possibly editors could be common to multiple uses
27/02/2007 15:51:11 ‹Gregg› I do have a question for you re: CTI.
27/02/2007 15:51:29 ‹KevinH› What we need is a new device type alongside BSC that reports its current state and advises what new states are valid (implemented as a BSC text device)
27/02/2007 15:51:44 ‹Gregg› I have a number of messages that axc relies on and another that I'm about to introduce that I've prefixed as "CTI2" so as not to collide w/ yours
27/02/2007 15:51:54 ‹Gregg› But that seems somewhat klunky
27/02/2007 15:52:33 ‹KevinH› Mine was only a first pass and it's only when I stopped and realised the scope of a CTI schema that I got scared :-[
27/02/2007 15:52:37 ‹Gregg› Oh--that's a possibility--except that implies the liklihood for a lot of text encoding to deal w/ the states
27/02/2007 15:53:20 ‹Gregg› Well, the question is whether you want for me to continue to use "CTI2" or to consider some form of merging. Clearly, the first step is for me to document
27/02/2007 15:53:31 ‹Gregg› and that's next on my wiki page list
27/02/2007 15:54:11 ‹KevinH› The schema should I believe allow for a whole menu structure to be described - complete with hierarchy and the enabled /disabled options at any one level - but maybe not teh dependencies that the enabling is based upon
27/02/2007 15:54:46 ‹KevinH› I am happy for you to take over and enhance the existing CTI spec ...
27/02/2007 15:55:07 ‹KevinH› ie merge / deprecate whatever
27/02/2007 15:55:26 ‹Gregg› wrt CTI spec... how do you want that to occur? updates to the existing page? a new page w/ links into the existing?
27/02/2007 15:56:04 ‹KevinH› One thing that has concerned me is that we dont report the revision level of a schema in use anywhere in xAP
27/02/2007 15:56:16 ‹KevinH› eg if we move BSC to v1.4 how do we identify that
27/02/2007 15:56:40 ‹KevinH› Sorry got to go again... probably char grilled now !
27/02/2007 15:56:46 ‹KevinH› back l8r
27/02/2007 15:56:57 ‹Gregg› that's a good point; in some cases, it's useful to know the diffs--are you considering that as well?
27/02/2007 15:57:07 ‹Gregg› ok--bye
27/02/2007 15:58:19 ‹Gregg› On a similar note.... I started thinking about application/code mirroring recently. I know the recent experience w/ trying to gain access to the xFX code and apps wasn't entirely smooth
27/02/2007 15:58:27 ‹James› well the wiki deals with the documentation diffs. for schema versions cti2 while not pretty is a good division
27/02/2007 15:58:47 ‹Gregg› Does it make sense to have some "agreements" to have code mirrored so that it doesn't just go vanish?
27/02/2007 15:59:40 ‹James› stuart can probably get ftp set up on xapautomation that way each developer can have a mirror
27/02/2007 16:00:02 ‹James› i have indended to send K a dvd with all my stuff jsut in case
27/02/2007 16:00:23 ‹Gregg› J--is it obvious easy to know how to do these wiki diffs specific to specific "marked" versions (like the way that version control uses "tagging";)?
27/02/2007 16:00:58 ‹Gregg› Ok--I just wonder if users may feel more comfortable if they know that the apps/code are being managed in a way that will persist
27/02/2007 16:01:27 ‹James› i'm sure they would be happier knowing there was a mirror
27/02/2007 16:01:53 ‹James› wiki diffs. just click history on a page
27/02/2007 16:02:43 ‹Gregg› wrt wiki mods, I tend to make frequent mods (kind of like frequent checkins); but only one of them should eventually be considered a viable "version"
27/02/2007 16:03:10 ‹Gregg› So, does the xapautomation ftp route seem like the best? or, is something else preferable?
27/02/2007 16:03:13 ‹James› there is a minor edit tick near the bottom
27/02/2007 16:03:41 ‹Gregg› Right--I try to remember to click that. So, the minor edits can be ignore then across the history?
27/02/2007 16:04:05 ‹James› i think it is a good plan don't know about how stuart would feel re. the overhead of the ftp accounts
27/02/2007 16:04:24 ‹James› could alwasys store in the wiki i guess
27/02/2007 16:05:20 ‹James› a big advantage is that all the sites are on different hosters now