Re: XTB - the Future of X10 has arrived!

[Larry Moss <moss@xxxxxxxxxxxx>]

On 2006-06-05, Robert Green <ROBERT_GREEN1963@xxxxxxxxx> wrote:
> The "installed user base" effect is probably the most important in the PC
> area.  If you've got a lot of users, people will write software for your OS.
> People will design hardware based on your OS. Documents are easier to
> exchange with common platforms.  Education of users is easier with a common
> platform.  No one ever enjoyed reformatting Wang documents for a PC or
> trying to move a Ventura Publisher doc created under Gem to a Windows
> system.

My turn to challenge your assumptions.  I was doing computer support in the
early '90s.  The common platform idea sometimes has merit, but when it came
down to it, I prefered to help people with the appropriate software for
their job.  With very little truly cross-platform software, that usually
meant multiple hardware platforms as well.  If all of your users are doing
the same job it's easier to have them all use the same stuff.  But even
putting them on the same platform doesn't standardize the software used by
the art department and accounting, so much of the benefit is lost while
possibly adding unintentional complications.

> X10's not like an OS.  There's a lot more room for competition in the HA
> arena that there is for a new OS to take hold.

But only if you're willing to try the competition.  Your recommendation to
newbies is still to stick with the installed base.  Then once they've
gained experience, they already have an investment in hardware that they'll
want to keep using.  So how will another product line take hold?

> I'm a fairly frugal person.
> I'm not likely to abandon something I've spent a lot of time and money on
> unless there's no choice.

I'm not suggesting you abandon an installed system if a fix exists (which
it seems to).  I'm just suggesting everyone stay open to alternatives for
new installations.  I suppose I'm strange in that I started with a system
that I planned to yank out.

> I also think
> it makes sense for a first time buyer with no HA experience.  They start up
> costs are small and there's an awful lot of support resources on the net.

The start up cost is small.  That's why I started with it.  But while the
resources exist, I found myself with a lot of unclear and sometimes
conflicting advice.  And it took a lot of time to put all the informational
pieces together.  Out of the box, X10 is unreliable.  A new,
non-technically inclined HAer has to learn that you need pieces from X10,
Inc, including filters or other third party products (like Jeff's) just to
make the system work as it was originally advertised. Someone walking into
Radio Shack to buy stuff off the shelf isn't handed any of the reference
material you're referring to.  They just see a display that says they can
automate their lights by plugging in these few things. You have to know
that you need that information first.  I know a few people that tried X10
before, didn't get the results they expected, and then gave up on HA
without doing any research.  Creating a working X10 system may be easy with
the right knowledge, but it's a confusing path for a newbie any way you
look at it.  Now if X10 wants to start selling Jeff's device as part of a
starter kit, things might be different.

One thing I have to say never confused me and I wish it existed across the
board is code wheels or some other way of programming things right on the
device itself.  You can set the device in your hand quickly and easily and
lookat it later and know how it's set.  Like many people in this group, my
greatest concern about Insteon is the current lack of an easy way to set up
the network.  (I haven't look into any of the software that's been
mentioned here the last few days to solve this.)

> In reality, Larry, I *am* going with new technology:  Jeff's!  That's what I
> think is lost here.

But it's a patch on an old system.  Apparently a good patch, one that's
needed by a lot of people, and something I believe he deserves lots of
credit and money for.  But until it's integrated into starter kits (or the
X10 devices themselves), it's still going to be applied as a bandaid.  As
someone that has been following changes in the HA market for a little
while, that has read many of the online references, and has already
invested money in X10, it may well be the thing I choose to use.  I think
it's going to be a hard sell to a newbie.

> Worst case scenario.  "XsteoPB4WaveBee" or whomever goes out of business.
> Your house takes a hit but not all the equipment is ruined.  Just the
> controller.  Well, if there's no replacement controller, you've got to pull
> the rest of the switches - probably not covered by insurance because they
> are still OK - and get something else.  With X-10 that wouldn't happen.
> Loads of controllers from lots of different makers.  Probably a dozen NIB
> replacements available within days from a number of sources.

That worst case scenario is bad, but unlikely.  It's a gamble I think I
would be willing to take.  (I know you already made a distinction between
your own money and someone else's, so I already know your objection to
this.)  I wouldn't install a controller that's already off the market.  But
I'd be willing to take the chance that a controller that *may* die, *may*
be impossible to replace.  Using your own example of ebay, when this
happens, there will either be someone else in the same positionas you that
wants to sell the old stuff or that will buy your old stuff.

> The more complex you make a system, the less reliable it becomes.  That's
> just the way things are.  I would not want to mix protocols unless I
> absolutely had to.  I've had to help maintain Apple and IBM on the same
> network.  What absolute and utter misery THAT was.

I don't see the problem here.  In those old networks you needed an Apple to
talk to the IBM.  Here you just need a controller to send the right signal
to each device and receive signals from those devices.  The devices aren't
talking to each other.  Mixing protocols on the same branch of the network
might be an issue since you can end up with one device being a signal
sucker for another.  I have one server (think of it as the central
controller/bridge) in my house that understands NFS, SMB, and AFP.  None of
the non-server devices that use any of those protocols care about any of
the others and there are no conflicts.

> Well, that's a perfectly viable opinion to have when you're spending your
> own money.  I wouldn't recommend any of the new protocols to a newbie,
> still.  As Dave Rye said, and I paraphrase, these new technologies all claim
> immunity from the technical problems "plaguing" X-10 but in reality they are
> neither old enough or well-tested enough to know whether they've got
> problems all their own.

And to paraphrase Dave Houston, "the XTB isn't old enough or well-tested
enough to know if it has problems." :-) And what if Jeff stops selling his
device for any reason?  If you're dependent on his hardware to make your
whole system work, you're in the same boat as if you went with something
else.

> There's no
> other protocol that offers so many different types of controllers so
> cheaply.

This is true and disappointing.  If I could get Insteon motion sensors,
that might have changed my decision to start with X10.  (Actually, I don't
recall if Insteon was selling yet when I bought my first modules.)

> Why, exactly, would you choose another protocol, Larry?

Well, for one, I have the same distaste for X10 Inc as I have for MS.  The
spam and pop-up/under crap they pulled early on was more than just an
irritant.  I know that has nothing to do with protocol, but that distaste
has me wanting to find a decent alternative from a company that has more
user friendly business practices.  I can buy smarthome modules, but I
haven't had the best of luck with them (which does have me worried about
Insteon).  I didn't realize that some of the other stuff I bought was
indeed made by X10 until after I got it.

It's not really the protocol I'm deciding on anyway.  It's the overall
system. Can I get the pieces I want (a plus for X10)?  Is it reliable (not
just the communications, but the hardware itself)? Can I get the status
information I want (I like 2-way communication)? Will other advances in
technology potentially cause problems down the road? (I prefer wired to
wireless if I'm building new since wireless has a greater chance of outside
interferences.)

A keyless doorlock system is also important to me.  No way I'd trust
security stuff to X10.  But right now, I'm using X10 along with my own hard
wired keypad.  Hey there's an example of mixing protocols with no
interaction between them.

Aside:  One thing that really bugs me about what I have right now may be
due to X10, it may be due to MisterHouse, or it may be that I'm doing
something wrong. I'm open to any suggestions.  When I walk into a room with
a motion sensor, it may take several seconds before my lights turn on.  I
don't have the lights on the same code as the motion sensor. Instead, I
have MisterHouse respond to a message from the MS by turning on the
appropriate lights at the right times or otherwise logging activity.  Is
this delay of a few seconds normal with X10?  with other protocols? or do I
have something screwy I need to debug?

> Thanks for sharing your views, Larry and for giving me the opportunity to
> explain my reasoning in a little more (and maybe excruciating) detail!

Same here.  I appreciate the discussion.  I will somehow figure out what I
want to do when I build the new house.

--
Larry Moss, http://www.airigami.com
PO Box 23523, Rochester, NY 14692, (585) 359-8695
Airigami: The art of folding air in specially prepared latex containers.