Anyone interested in new software / open source contributions

[Steve]

Brad,

Let me start by saying that I wholeheartedly support and respect your determination, talent and desire to do this. That being said, I think you are perhaps underestimating what it takes to create a product that is rock solid reliable and feature rich. By the time you build/test/debug what you have, the other packages out there will probably fill in whatever gaps there may be. IMHO, unless you have more free time than you know to do with and just want to do this to fill and the time and because you love it, you may be better off finding a package that does 90% of what you want now, and work with the manufacturer to perhaps fill in missing pieces.

I agree that I do no like Windows as a core backbone of my system, but I have to admit I am believing more and more that most of Windows problems are the user and everyday tinkering. I think a 'stripped' version of XP with nothing but the essentials and left alone will be pretty darn stable. And alot of that has to do with the software running on it. I know Dean has spent many many years building the infrastructure of CQC, it began way before CQC was even conceived. It's had many years of testing, tweaking, etc which is probably 1 reason it is super reliable. I can tell you I personally hate Java and have had more problems with it than anything else. I would never buy a Java based product. Elk is abandoning Java for their Virtual Keyboard app and in my IT department we have more compatibility issues with Java than anything else.

That world isn't perfect and neither is any 1 product. You can spend years of your life dedicated to building a product the way you like it but in the end I think you are better off using a currently available product and getting it improved or working around some of the limitations.

Whatever you decide, I wish you luck.

 

[purple_motion]

Take a look at xAP and xPL. I am looking at xPL and using the xPL perl modules.

I really like the idea. There is a well defined separation between the device drivers and the core (the event driven scripting engine). The problem with a lot of the HA apps is that they integrate all the device drivers. They do this to attract users and then keep them locked into their platform. In reality it means each HA app re-writes device drivers for all the popular devices. It would be so much better if the device driver was written once in xPL/xAP and then each HA app could make use of it. Bugs would be found quicker and not replicated in different HA apps.

I am sort of following the path of the xPL perl developer (check xpl-forum for his name & website).

He is using the web as the interface. You can get the look/feel of native applications using AJAX and DHTML. Super portable. No downloading of plugins/applications on the client side. Works on windows/linux/mac. All you need is a web browser.

MD

 

[shenandoah75]

steve - fair enough and i appreciate the candidness...

The problem with a lot of the HA apps is that they integrate all the device drivers. In reality it means each HA app re-writes device drivers for all the popular devices. It would be so much better if the device driver was written once in xPL/xAP and then each HA app could make use of it. Bugs would be found quicker and not replicated in different HA apps.

BINGO!!!!

 

[johnnynine]

I agree with Steve, and just to add one more thing... burnout. There usually seems to be a lot momentum at the beginning of these kinds of projects, but we all have other priorities that usually catch up with us in the end, and or we just get burned out on it.

 

[upstatemike]

Some things I would like to see included would be:

1- Serious good, extensive, elaborate, TTS and VR like you see in VCrib. I don't understand why more HA apps don't allow for easy switching between TTS voices etc.

2- A distributed speaker app similiar to Homeseer. There are lots of situations where it is good to have different things spoken in different locations at the same time which can't be done by just distributing sound from one sound card.

3- Integration with Slimserver that provides full multi-room support AND allows HA messages to be easily directed to Squeezebox displays. (Easily done now with scripting but it would be nice to have a gui for this).

4- Elk support that not only covers the full Elk feature set but also provides enhancements and missing features not provided by Elk.
Examples:
--Ability to adjust Tstats from Elk Security Keypads (regular manual adjustment - not trigger presets)
--Ability to kick off TTS or wav messages though phone system after Elk dials paging access
--Ability to set Elk touch screen to a particular camera in response to doorbell or motion sensor
--Ability to load weather station data from VWS into Elk variables for use in rules (If counter windgust is greater than 15 then...)

5- I'd like to see a good point and click rules engine like you have with ELKRP except with much deeper logic (or statements and nested if/thens) and many more functions like there is in WinEVM. I like to focus on the logic of getting things to work and NOT on the syntax of some scripting interface.

 

[CollinR]

People get open source and free all confused.

You could have an open source but not free software.

You can have closed source but not charge for it.

You can charge for distributing and serviceing "free" software.


In truely "Free" software the free comes from freedom of use not that it has no monitary cost.


I have no clue why people build plugins for SageTV, CQC, Girder on and on and then release them for no cost. It's like they are dreaming of having "Free" software but don't. You won't catch me falling into that cycle, if I start helping with a project it will be a Free one. Otherwise I'll pay my license fee and go on.


EventGhost, xAP, xPL and quite a few others are good starting points. Also obviously Vaughn here with Vcrib seems logical.


I like Java apps, SageTV is one and I have really enjoyed it. Really it is the frontend for all of my automation and I only continue to improve it.

 

[IVB]

I have no clue why people build plugins for SageTV, CQC, Girder on and on and then release them for no cost.

Heck, that's an easy one: Three reasons that i've heard, in descending order of importance. This is by no means a conclusive list, only what i've heard from folks. Others may not share 1 or more of these, and may have additional reasons of their own.

1)Karma/pay-it-forward.
Sure, today you may be writing something that many others will use, but tomorrow you may need something from them. Do you really feel like getting into a whole consumer-istic ballgame where you get nickel&dimed for every last little thing you want, or would you prefer a nicer world? Move on to ML or HS if you want the pay-per-driver model, SageTV/CQC/Girder are for those folks who tire of being bled dry at every turn.
I know personally i've benefited greatly from the work many folks have done, and wouldn't have nearly the system I do without their work. Names like ripper(graphics), ellisr63(graphics), beelzerob(sage), robertmee(concerto), NishanF(sage), jkmonroe(rfid/xm), jonathan/bradsjm(rfid/xm), jrlewis(recipes), helheim(web-concall-thingey),zaccari(VK011), and I can't remember who did the db logger, and that's just a partial list. Thing is, at least some of those folks wouldn't have joined CQC in the first place unless me/jkmonroe/zaccari/lpott did the monthly webinars. It's a win-win, I help you for free, you help me for free.

2) ROI/Tech Support.
Most of the CQC driver writers are already charging requestors for the upfront cost, having folks pay per-download would generate all the negative dimensions of entitlement that appropriately come with commercial plugins. How much could folks possibly make on recurring plugin revenue from the DIY sector, and how much work would they have to put in? Even if it reached $10K/year (and no one will make that from plugins) we'll all need to keep our day jobs to pay the rent. Better to just release it, and provide a "best-effort" type of support.

3) Increase odds of CQC/SageTV/Girder survival.
It's been proven true that there's typically only one winner per space, very occasionally two, with the other companies going out of business or being acquired for their customer base with the product itself being dissolved. How many folks use an auction engine other than eBay? Buy books & CDs without checking with Amazon? We've put so much blood/sweat/tears into the non-CQC portion of our setup, and optimized the CQC portion to leverage it, that it's in our best interest to make sure the company is always in business and we don't have to start over. For many folks, expanding the appeal is one way they can help.

As said above, this isn't the definitive list of reasons, just the ones i've personally heard. For me personally, #1 is the reason I do what little I do. The rest of life is me-me-me/$$ driven enough, i'm interested in being part of a nicer, gentler, more noble community where folks help each other without any expectation of anything in return. And Dean deserves much more credit for creating a community like that than any mega-HA application with 900K lines of code could ever attain standalone.