RE: [Embedlets-dev] Global Light Blinker Project proposal

[James C. <ca...@vi...>]

Well Ted,

Why didn't you SAY SO!!

Well done.. It's a very good point that you don't always hit the target
you are aiming for.. I am impressed that you have got so far as to have
a manufacturer and working model which is driving your confidence.

I must admit I am not paying as much attention as I should be. So this
idea is to allow Home automation meaning people can connect to their
homes webbrowser and turn lights on and off?

BTW - for the dummies.. What is NAT?

James Caska
www.muvium.com
uVM - 'Java Bred for Embedded'



-----Original Message-----
From: emb...@li...
[mailto:emb...@li...] On Behalf Of
Ted Kosan
Sent: Friday, 4 July 2003 10:19 PM
To: emb...@li...
Subject: RE: [Embedlets-dev] Global Light Blinker Project proposal


Topic tags:[ARCH][JAPL][WIRING][DOCS][MGMT][STRATEGY][NEWBIE]
_______________________________________________

James wrote:

> This project does get me wondering though.. Wondering
> about what happened to outpost. I don't think the graphic wiring tool 
> was the thing that slowed it down.

I think that Outpost is alive and well but the problem is that my
version of a Outpost marketing strategy has the following dependencies:

1) Graphic Wiring tool.
2) Deployment/Configuration mechanism.
3) Solid JAPL specification and implementations.
4) Standard way to connect an Outpost to a backend enterprise system.

We don't have any of these pieces yet and it looks like it is going to
take us a while to build them.  Beyond this we do not have any users
using the code we already do have which makes it difficult to move
forward.



> I wonder if we are going to make the effort to build something like 
> this that it shouldn't be more along the lines of an outpost demo that

> might have a commercial impact rather than a hobbyst project.

In the process of building something commercial, what happens if a great
sub-commercial opportunity is discovered that directly helps build
towards commercial-quality code?  

The funny thing about Java is that it is a very inaccurate (albeit
powerful) 'weapon'.  Sun initially aimed Java at the Embedded Market and
it endedded up hitting the Enterprise market.  Here we initially aimed
Java at the Enterprise Market (with Enterprise Outpost) and I think
there is a very good chance that we are going to end up hitting the Home
Embedded market.

If we go back to the main reason that the Outpost idea was conceived in
the first place, it was to find a large market for Embedded Java.  This
overall goal supersedes all lesser goals.  The Embedlets/Outpost project
was designed to be a vehicle to use to help with this discovery.

From my perspective the project is meeting its purpose excellently
because I know that I have a much better understanding of Embedded Java
dynamics than I did when we started this journey back in October.  I now
think that I have a fairly good grasp of what a good Embedded Java
opportunity looks like and what the Embedded Java pitfalls look like.


We have all put a heck of a lot of work into Embedlets/Embedded Java
over the past year and now I think it is time for some of that effort to
pay off.  

In my opinion, the Global Light Blinker (GLB) project opens the door to
the greatest Embedded Java opportunity to come along since the concept
of Embedded Java was first developed and the Embedded Java community is
currently the only group on the planet which is in a position to pursue
it.


Here are my reasons for thinking this:

1) The GLB project shows a large number of people what Embedded Java is
capable of without having to explain anything to them (the project's
visibility potential is enormous!).  As soon as they see their lamps
turning on and off remotely through the internet they will instantly
'get it'.

2) As soon as people's 'lights go on', how long do you think it is going
to take for them to come up with dozens of ideas for things to control
inside of the their homes that they can actually use?  How many of these
people are business people that will instantly see how they can sell
devices like this to other people?

3) At this moment in time, the main opportunity for building these
devices, and showing others how to build them, is being offered to the
Embedded Java Community (as defined by
http://embedlets.org/images/flounder.jpg). The Embedded Java developers
represent the primary Community in the world that has the correct mix of
Hardware/Software skills needed to build these devices.

4) The great thing for us is that the 'remote control of in-home
devices' opportunity is in its very early stages because NAT has been an
effective roadblock which has prevented it from materializing.  JXTA 2.0
smashed this roadblock a few months ago and now, very suddenly, the
Embedded Java Community has *secure* access into the home.  To get
started, we don't need commercial-quality hardware or a finished and
commercial-quality Embedlets implementation.  As incomplete as it is,
the Embedlets implementation that we currently have is more than ready
for starting to pursue this opportunity.


To summarize, the Enterprise Outpost idea is a great idea, and we are
slowly getting there. In the process of building the Enterprise Outpost
the Home Device Outpost was discovered, and it has a large market that
we can pursue right now.



> The other thing we need to be careful of is not to dilute our efforts 
> to such an extent that progress stalls.

I would say that progress on Embedlets is about to make a great leap
forward. 
When viewed from the right perspective, a number of seemingly
disconnected efforts can be seen to actually be part of a
well-coordinated effort.  Some people's style when building a system is
to work sequentially and to completely finish the current component
before moving on to the next step.  

A style that contrasts this is to work on all of the components in
parallel so that the quality of the interfaces between the components
can be maximized and also so that the system can be better tuned to fit
its overall purpose.

I am definitely a parallel-development-style person so perhaps this
knowledge might ease your concerns a bit.



Anyway, for the past couple of weeks I have been working 'like a dog' to
get my mind around JXTA 2.0 with the goal of remotely blinking a house
lamp which is sitting behind NAT in someone's home.  I am so close now
that I can 'almost taste it' and with any luck the first LightBlinker
should go live sometime this weekend.

I have also already found a company that is willing to make the beta
hardware and I even convinced them to hire our University students to do
the assembly. 
The hardware specification is going to be open so that anyone can
implement them but it is my guess that most people would rather just buy
a reasonably priced pre-assembled box.


But actually, the LightBlinker project is just a proof of concept, the
real opportunity is in meeting the demand for all of the other things
that people want to monitor and control in their homes.


So, is anyone here interested in this idea enough to discuss it further?


Ted

__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100006ave/direct;at.asp_061203_01/01
_______________________________________________
Embedlets-developer mailing list
Emb...@li...
https://lists.sourceforge.net/lists/listinfo/embedlets-developer