Page 3 of 5 FirstFirst 12345 LastLast
Results 21 to 30 of 41

Thread: 2008 Brings Smart Home Utopia?

  1. #21
    Moderator Kevin's Avatar
    Join Date
    Jan 2004
    Location
    West Yorkshire
    Posts
    554

    Default

    Quote Originally Posted by The Masked Installer View Post
    When we started I really thought that xAP or xPL was going to be really useful in what we did. We never really got it and by the time we were ready to get it we'd already got into AMX and Crestron and that was that. Our market had changed and what we wanted to do had changed but I still thought it looked interesting.

    In that context having the ability to create your own glue is invaluable. Ours is lovely but rather expensive. Where's the xAP and xPL glue these days
    xAP and xPL are exactly that 'glue'. They are in most aspects near identical in purpose and implementation and both offer a great solution to the X challenge. Bear in mind that xAP/xPL are protocols and not products , in exactly the same way that TCP/IP is a protocol. As such both xAP and xPL are here, now and offer a solution, it just needs people to use it.A comparison to Crestron or AMX is inappropriate as they are different things although I do run xAP on both. Perhaps the better question therefore is where is the widespread adoption of xAP/xPL that we would like.

    xAP/xPL do not have the resource, money and commercial qualities to give them enough prominence. Even within the original common goal UKHA brief for a simple X protocol there was still a project fork based on personal differences of view and even subsequent forks after that. So I see no possibility that an X protocol could ever exist in the tougher commercial world. The fork we suffered is likely to have placed insurmountable obstacles in front of both protocols and certainly doubled the work required and halved the resource available to both.

    Commercial companies need protocols that have resource behind them, both to ensure the technicalities of the protocol, to ongoingly support and add features they might need, and to ensure legal and commercial aspects are protected including allowing their products some degree of commercial isolation . The commercial vs technical is an old chestnut as in VHS v BetaMax , DVD R/W v RAM , BlueRay v HD.

    UPnP is very similar in purpose to xAP and xPL albeit far more capable but orders of magnitude more complicated and thus requiring bigger/faster hardware to run on. As such it often excludes itself from the custom integration market. It has millions of $ being poured into it and thousands of people working on schema definitions. It is backed commercially by huge names, Intel, Microsoft etc.... and yet even UPnP is struggling and slow to emerge. It has appeared to an extent in AV media devices as a standard and yet even within those areas the commercial aspects have broken the 'standard' and products fail to work together. Sonos is a classic example where the UPnP implementation has been intentionally adapted to commercially isolate (protect) the product. Even the wireless networking was changed although here there were justifiable technical reasons. UPnP could prove to be what's needed here at a commercial level albeit not end user accessible.

    xAP and xPL offer you an open and capable solution should you be prepared to invest in the time required to understand their purpose and then to support the protocol.

    Speaking for the xAP side, it is beyond our means as a small group of unpaid individuals with other full time jobs and families to develop boatloads of applications , document, handhold all the 'I'm not at all technical' support questions, canvas and achieve support in commercial products and address all the 'can you create support for the ABC brand mp3 player Father Xmas gave me' requests. Our hope is that the project snowballs and becomes self supporting but it is tough and likely unachievable. The xAP people in-particular have invested shed loads of time in releasing example applications, writing SDK's and getting xAP support included in leading commercial applications.

    I also have to mention that there is a general malaise on the UKHA forum that expects everything to be 'free' or for near nothing, which is in areas such as this, self defeating. There is really little reward for the investment and so I think we now see a scenario where the protocol is being adopted by those 'in the know' for their own purposes, and jolly good it is for that, but the evangelical 'its for everyone' approach is being tempered.

    I know of loads of xAP applications that have been written where the author just isn't prepared to release it publicly. It is probably 10% of the work to get something that works absolutely perfectly for you , compared with something that is configurable, documented, feature complete, online supported, protected against all user error issues etc, and for what reward ? There are even commercial applications - for example Charmed Quark that use xAP internally in their event model but don't expose it to the users.

    To take a specific example I have xAP implemented for HomeVision, AMX and Crestron. I have released the HomeVision product as it's a UKHA favourite even though I myself don't base my system around that controller . Myself I use both AMX and Crestron and because of their extra processing capabilities have implemented xAP natively on them (no external hardware) , it's great ... but there is no incentive for me to release this and indeed there are difficult commercial issues with regard to both companies - particularly Crestron and IMHO their ludicrous software policy.

    As ever with me - a simple post has turned into a novel...But if you need a simple, expandable glue protocol that is really easy to support then consider xAP/xPL - it works for me..... but you will need to invest a little time getting to grips with it's purpose and implementation. The best thing about xAP/xPL though it is that as a coder it's about as near to 'easy' as your ever going to get.

    Kevin

    PS These views are all my personal views and not an official xAP positioning, and I certainly can't represent xPL although I believe we face similar issues.
    Last edited by Kevin; 29th December 2007 at 02:43 PM. Reason: typos

  2. #22

    Default

    Good post Kevin, and I agree with a lot of what you say. The Fork was a major blow that certainly confused the issue and reduced the impact. I was hopeful (for a while) that it could be worked round. Alas not.

    I of course had the UKHA_D born protocols in mind when I was writing the article the "x" in system "X"ap?

    As an aside I also feel breaking the protocols own mailing lists away from UKHA_D lead to a reduced audience for the projects, as well as a reduction in the quality of posts to UKHA_D). I notice several other breakaway lists are all but dead. Perhaps its time to bring some back onboard.

    You mentioned "xAP support included in leading commercial applications" - Can you update us on what applications/hardware currently support xAP?

    Is there any mileage in moving xAP/xPL to sourceforge or Google Code? How do we increase exposure? How do we move it to the next level? I'd imagine releasing your AMX and Crestron integration would make some installers look at it again!

    Thanks

    M.

  3. #23
    Moderator Kevin's Avatar
    Join Date
    Jan 2004
    Location
    West Yorkshire
    Posts
    554

    Default

    Quote Originally Posted by Otto-Mate View Post
    I of course had the UKHA_D born protocols in mind when I was writing the article the "x" in system "X"ap?
    Yes, I'd got there although I was worried you might have OSX fever to
    As an aside I also feel breaking the protocols own mailing lists away from UKHA_D lead to a reduced audience for the projects, as well as a reduction in the quality of posts to UKHA_D). I notice several other breakaway lists are all but dead. Perhaps its time to bring some back onboard.
    Understandably some people just don't want xAP rammed down their throat, even if was created as a UKHA project, and the xAP/xPL divide makes things worse.

    The other difficulty is that in order to support the various different applications listed below we have had to create xAP lists/forums on many other boards which has further fragmented the contributions. We have maybe 10 such boards now, most fairly quiet but some very busy eg the mcs xAP HomeSeer ones.
    You mentioned "xAP support included in leading commercial applications" - Can you update us on what applications/hardware currently support xAP?
    HA Applications:
    (Only listing HA control / User interface applications)

    HomeSeer: - 3 different xAP plugins available - all free
    HouseBot: - xAP Plugin (beta) released
    Charmed Quark: - xAP used internally and xAP server module promised but still n/a, script support however.
    XLobby: - xAP Plugin available
    MisterHouse: - xAP inbuilt
    Girder: Inbuilt xAP
    MainLobby: - xAP currently via HomeSeer conduit. xAP plugin coded but surpressed by Cinemar :-(
    xAP Desktop: xAP inbuilt
    xAP Floorplan: xAP inbuilt

    Hardware with embedded xAP
    (standalone devices - no PC required)

    Phaedrus xAP Netiom: Digital,Analog and Serial I/O , counters latches etc
    OPNone: 1-wire network interface with web server etc
    OPNX10: X10 interface
    OPN232: Serial interface (beta)
    OPNMax: Complete embedded xAP controller with scheduler/logic engine, webserver, scripting, SQL inbuilt database and PHP . Runs on NSLU2 (Slug)
    C-Bus Gateway: Direct attach to C-Bus (doesnt use C-Gate)
    HomeVision Gateway:
    IRTrans: Networked Infra Red distribution
    Meteor Interface: Caller ID device
    Quasar Interface: Temperature reporting
    (more to come .... in alpha currently)
    AMX/Crestron - achievable but not released

    Is there any mileage in moving xAP/xPL to sourceforge or Google Code? How do we increase exposure? How do we move it to the next level? I'd imagine releasing your AMX and Crestron integration would make some installers look at it again!
    xAP as a protocol definition doesn't need any code development (as there isn't any code) so these don't really suit. We found reaching agreement on the protocol really hard even in a fairly small committed group, and even then we had forks. However some bigger xAP applications or development tools might benefit from this , although it's not something that would suit me personally.

    AMX and Crestron are more complex to release in terms of the legal constraints that AMX and Crestron put on access to their programming tools. (particulalry Crestron ARRRGHH). Again these modules work for me but it's 10 times as much work again to make them resilient, universally useful tools. This would need the modules to be chargeable, which of course AMX/Crestron dealers like, and this incurs support obligations that need consideration. It's not out of the question though.
    Crestron/AMX demand for modules is generally driven by a need to integrate a specific device ... - a sort of pull rather than push situation which is not ideal for xAP. However having Sonos, UPnP, C-Bus, 1-wire or rich data (weather, TV listings, stock prices, news stories, email) etc available via xAP could stimulate this...

    Personally xAP has shown me the methodology - achievable within my own limited programming skills . I know how I should approach networking the devices and so I now have many many todo's keeping me busy on my own system for months to come.... I feel comfortable too that because everything is distributed I'm not painting myself into a corner. I've reached a situation where the hardware now meets all my needs and all the constituent parts are available - so it can be done, it's just time and effort.

    Kevin
    Last edited by Kevin; 29th December 2007 at 03:49 PM.

  4. #24
    Moderator toscal's Avatar
    Join Date
    Oct 2005
    Location
    Near Alicante Spain
    Posts
    2,010

    Default System xXx the next level

    Why not get manufacturers to use some form of serial port, that uses standard Ascii text strings. For example a washing machine could transmit " Program Start : Temp 40 : Spin 800 : Running Time 73m :" Or you could send a command like "# Start Program Cotton : Temp 40 : Spin 1000 : Start Now#", to start the machine.
    I just need a robot to unload the washing machine and put it in the tumble dryer.
    Most Home Automation controllers have a serial port or USB ports, USb to Serial adapters are freely available. Even my Plasma TV has a Serial port for remote control of the TV. The actual cost to implement this would be minimal as it could probably be built onto one chip. Cheaper units would just transmit their status and the more expensive ones would except in coming serial data as well.
    The main problem would be when manufacturers offer enhanced serial ports that only respond to their own equipment. Sony are good at at doing this.
    On another note, I was talking to one of the guys on the Siemens stand at a Home Automation fair and he was telling me that they are working on a project with a Japanese company to develop a fridge freezer that knows when you are running low on food. Say you are down to your last portion of frozen peas. The LCD TV on the front of the fridge would start showing you adverts for frozen peas or remind you that you need to buy some more and add it to your weekly shopping list. Then either print it out or send it to your local supermarket where they would then deliver your shopping to your door, or you go and pick up the weekly shop.
    Big Brother will soon be watching you.

  5. #25
    Moderator Kevin's Avatar
    Join Date
    Jan 2004
    Location
    West Yorkshire
    Posts
    554

    Default

    Quote Originally Posted by toscal View Post
    Why not get manufacturers to use some form of serial port, that uses standard Ascii text strings. For example a washing machine could transmit " Program Start : Temp 40 : Spin 800 : Running Time 73m :" Or you could send a command like "# Start Program Cotton : Temp 40 : Spin 1000 : Start Now#", to start the machine.
    Simple isn't it - and that's almost exactly what xAP and xPL do now Both can work over a serial port (or almost anything) and contain parameter keys readable by humans Temp=40 etc. We then create a 'schema' which defines the keynames - in this case it might be a 'washing machine' schema . Then if every machine supported this schema they all become controllable and can report their progress. Even better you can create an automation system around your ACME washer and then when you swap it later for the ABCWhiz - it still works. Ideal world It's creating the schema and getting them widely adopted that's difficult .

    Realistically a capable wireless protocol is the most consumer friendly, even if it costs a bit more, it is likely more cost effective to integrate the device into a system (cosmetics/wiring/labour costs etc) . Plus of course its 'networked' then and not point to point. For some applications and devices wired will be preferred though for dependability and security.

    xap-header
    {
    v=12
    hop=1
    UID=FF223330
    class=washing.event
    Source=ACME.ClothesWasher.Kitchen
    }
    current.mode
    {
    State=on
    Temp=40
    Spin=800
    Running Time= 73 minutes
    }
    Last edited by Kevin; 29th December 2007 at 05:50 PM.

  6. #26
    Automated Home Legend Paul_B's Avatar
    Join Date
    Jul 2006
    Location
    Eastbourne, UK
    Posts
    603

    Default

    I've seen a few German appliance manufacturers are starting to offer HA in the high end products. Siemens DECT have the Gigaset Home Control, Liebherr have Net@Home for fridges and freezers. Although no common interface.

    Is the HA environment polarising between the DIY enthusiast and the Professional Installer? This polaristaion requires different solutions surely? The high-end professionally installed system is for a customer who doesn't want to tinker and a system that is rock solid. They may not even do their own washing ;o) Finally, this customer can pay the high cost for installations.

    The second group is probably more represented by this forum and will install their own system where it is almost a hobby. They want to tinker, expand and a system that encompasses disparit systems. However, cost is a major issue.

    Can a single system work for both groups? Would the X system not work better with the latter group? Why has x10 been out for so long but does not seem to have moved forward? Is it now possible to bring all the splinter technologies and companies together to define a universal system?
    Last edited by Paul_B; 29th December 2007 at 05:45 PM.

  7. #27
    Moderator Kevin's Avatar
    Join Date
    Jan 2004
    Location
    West Yorkshire
    Posts
    554

    Default

    Paul,

    I think that is very true and a development paradox. People want a glue that allows seamless and transparent integration between devices - not a hint of configuration or anything technical in sight and yet others want very low level control. One mans' simple room thermostat would be anothers' nightmare as it didn't offer optimised start and support multiple sensors and dew point avoidance. Generally the simpler and more reliable you make it appear the less open/customisable it becomes . A user interface might be able to expose just the bit that appeals by sitting on to of a capable protocol but not easily.

    In a way Crestron implements this even within their own development environment - which has three tiers essentially. Preformed templates and wizards at the top, a graphical drag/drop programming interface with modular devices for average systems and a programming language (SIMPL) at the lower more complex level . All this supplemented by large libraries of drop in modules for specific devices, and IR/serial control databases often user contributed and free . This is where the investment has been made by Crestron (AMX similarly) and why the costs are still high and market entry by others now difficult.

    I think that is maybe why when we positioned xAP as a simple solution that potentially could make it all work... then we couldn't address the 'so make it work for me and my XYX mp3 player and BTW I know nothing about computers' scenario. We sit between the two levels touting a solution but not having the resource to move either way.

    My view now is that as a HA user/enthusiast you either need to develop your own skills inline with your HA interest and aspirations or use a company that can provide it transparently for you. Yes, the latter can be expensive (AMX/Crestron) but that must be moving to more affordable. The actual electronics in a controller or touchscreen from either of these companies is a fractional part of the sale cost and already programming is separately (and expensively) charged. If the pyramid base is to be broadened these costs must reduce.

    Interestingly it's also the direction that, Charmed Quark, MainLobby and even XLobby are all now heading, targeting their UI software at the custom installers, and increasing the costs to satisfy their need (the installers) to make money...

    xAP is within the capabilities of people with fairly modest technical expertise - anyone with a bit of programming experience. Things like UPnP are not. Stick within a range of compatible products eg KNX and it's so much easier... just a little restrictive. If you wish to support many different devices/interfaces, and I think this is a typical scenario with legacy devices and limited funds, then be prepared to invest some time and get your hands dirty, or find someone to do it for you. Integrating differing devices and manufacturers without X is just always going to be awkward and X is still a way off ... if ever ... and mostly for non technical reasons (commercial politics).

    K
    Last edited by Kevin; 29th December 2007 at 06:51 PM.

  8. #28
    Automated Home Legend Paul_B's Avatar
    Join Date
    Jul 2006
    Location
    Eastbourne, UK
    Posts
    603

    Default

    Kevin,

    That is very interesting about the higher end market.

    I am not convinced even an IT literate person could programme in xAP. I've been in IT for 10 years and it is only the last three I became a scripter using VBScript and now latterly .Net, Powershell, Javascript, VB.Net, etc.

    Do we not have a proprietary level at the hardware / module level and even the next stage at the protocol layer? It as this point xAP / xPL can come into play providing a unified interface and abstract individual HA companies. But then we still need a dynamic GUI interface to allow basic wizards and also granular control and scripting rather than programming. But this is a massive task and needs quite a few dedicated developers. Realistically doesn't this need someone like a Google project?

    I remember something similar with print drivers in Windows 3.1 and it wasn't until Windows 95 when Microsoft provided a unified framework and manufactures just had to provide mini-drivers up to this layer.

  9. #29
    Automated Home Legend TimH's Avatar
    Join Date
    Feb 2004
    Location
    UK
    Posts
    503

    Default

    Quote Originally Posted by toscal View Post
    Why not get manufacturers to use some form of serial port, that uses standard Ascii text strings. For example a washing machine could transmit " Program Start : Temp 40 : Spin 800 : Running Time 73m :" Or you could send a command like "# Start Program Cotton : Temp 40 : Spin 1000 : Start Now#", to start the machine.
    Picking up on Wintermute's earlier points about Ethernet, I guess ideally the washing machine or whatever, would pick up its info from Ethernet over Powerline networking technology. This appears to have matured sufficiently, if not completely and, like wireless, is essentially invisible. There's also no additional cable for the consumer to "forget" to connect...

    Cheers,

    Tim.

  10. #30
    Automated Home Legend TimH's Avatar
    Join Date
    Feb 2004
    Location
    UK
    Posts
    503

    Default

    Quote Originally Posted by wheelbarrow View Post
    Hi Mark,

    Haven't you posted this 96 days early.
    This was my first thought when I read Mark's initial post.
    Soon after that I got to the "isn't this what xAP/xPL are supposed to be doing?" point...

    Subsequent posts have confirmed that yes, this is xAP/xPL territory but where the gap lies is getting these protocols incorporated into mass-market electronics. Technically I can see no major challenge to incorporating "X" into any of the products on the "Automated Home" home page. Politics yes, but not technical issues.

    The list of xAP-capable products is very impressive considering that it is a part-time, unpaid undertaking. But to reach Mark's dream of bringing a new toy home and having it work out of the box is going to take some miraculous efforts, which is sad. (note: no disrespect intended to any of the xAP/xPL guys, but I'm just trying to face-up to the mountain ahead).

    From the stuff I use, Skype appears to have been the fastest-growing phenomenon in the past few years. Tonight it tells me there are nearly 8million users online. Their website used to say how many downloads they'd had and that was over 20million.
    Getting to my point...for commerical manufacturers to take notice and to start incorporating "X" into commerical products, like the next Philips photo frame, is going to take some doing. I don't know how we can do it but we need to change the current almost word-of-mouth method into something more dramatic.

    Automation at home is slowly gathering momentum. Programmes like the Gadget Show and Click are bringing "digital lifestyles" into people's living rooms. The population as a whole is becoming more tech-savvy, hell, even my Gran has a mobile phone!

    But, in the next five years, to be able to buy a washing machine that will text you when its done, or even a photo frame that will show you a picture of who's at the door is, well, optimistic at best. Sure there will be people for whom this will happen - it's happening for some right now I'm certain, but for mass market integration?

    Perhaps its getting Philips (or whoever) to market a web (door) cam and photo frame bundle which does this. And caller ID. And local weather. And local traffic reports. Perhaps like the Shell Home Genie stuff was suposed to work, just a bit s3xier than home security...

    That all sounds a bit gloomy as an outlook, but it needent be. We need a Max Clifford, or a Don King - someone that can evangelise to the world about how great this automation is, and why people need it in their lives. But I also think we need products on the shelves so people can try it out.

    I've rambled enough. Let me know what you think

    Tim.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •