My celebration a few days ago, thinking I had solved the home monitoring technology challenge, was premature. I wasn't able to get client computers to find Willing Webcam's output. Maybe this was finger trouble but, as these are my fingers we're talking about, I blame the software, which was also somewhat fiddly to set up - tolerable as long as it worked, but not ideal.

I tried yet another webcam package (Webcam Tracker), but eventually asked my local computer expert what he would recommend. WebCamXP was his suggestion. I tried it. It worked with every camera I threw at it, it put the video stream on the net with no fuss, and it has all the features I am after. You get what you pay for, I suppose, as it is also the most expensive of all the options I've looked at, but at $39.95 it is hardly going to cause financial hardship.

There are still some question marks: I can access WebCamXP from only some other computers, and only when it is set up to work with specific types of clients (Flash and Java seem OK, Javascript is doubtful). Why all this is so, I do not know, and a few hours of Internet research has unearthed all sorts of related problems and solutions but nothing specific enough to help.

However, given that WebCamXP also works with the Belkin VideoBus II (my video to USB adaptor), I am now free to look at the full range of tiny spy cameras, knowing that as long as a camera has analogue video output, it ought to work with the rest of the system.

Here's the configuration I have in mind:

Infra-Red Wide-Angle Miniature Video Camera(s)
connected to
2.4 GHz Radio Transmitter(s)
transmitting to
            Radio Receiver(s)          
 connected to
Belkin VideoBus(es)
connected to
Dad's Windows XP PC
running
WebCamXP Server v. 2
through firewall to
Internet
through firewall to
any browser on another computer

The bits in red are the only remaining untested links in the chain, and I am confident about this section because I've had it working before in research systems - using standard commercial products. Moreover, the conversion to radio signals and back to video is quite independent of the digital part of the system - whatever successes or failures I have here do not affect the choice of software.

The bits above the red line present a number of alternative solutions:
  • Cheap
    Reuse a camera, transmitter and receiver I already own - which means I must solder some connectors to a transmitter I am not sure still works, and perhaps compromise on quality and angle of view
  • Flexible
    Buy any combination of compatible camera, transmitter and receiver and be able to custom-tailor the system to tiny size, infra-red sensitivity, and wide angle lens
  • Easy
    Buy a packaged kit of camera, transmitter and receiver - such as those made by Swann, which force a compromise between tiny size and infra-red sensitivity (since the infra-red cameras are relatively large) and, perhaps, reduced angle of view - something standard like 45-55 degrees, where 90 degrees would be more like it.
The system now (partially) working requires simply that clients enter the IP address of dad's computer and the number of a port into the URL window of their browser. They will be asked to enter a username and password; I added this as an extra safeguard of my parents' privacy. Clients will then be taken to a simple web page that contains streaming video - in colour, 320 x 240 pixels, and up to 5 frames per second, perhaps more, depending on the bandwidth of their internet connection. Pretty nifty!