The Adolescence of D-Star
This blog post falls squarely outside the norm for me, but, this is my forum, so, deal with it.
I recently acquired an Icom IC-92AD handheld transceiver (hereafter, "HT"). In addition to being capable of transmitting on a couple amateur radio bands, the Icom radio also has wide receive coverage (provides entertainment value by being able to listen to police chit-chat), and support for a nifty (and relatively new) digital voice (and data) protocol called D-Star.
D-Star has enormous potential and some pretty cool capabilities available today - like, I can use my new HT to talk to the repeater on the Empire State Building and I can specify that I want my voice to be carried to, say, the repeater in Perth, Australia (over the Internet) and broadcast from there for any D-Star user in range of that repeater, to hear (and reply to).
What the D-Star system has in vision it lacks in maturity. Let's contrast D-Star with the Internet. If you haven't yet taken the time to glance over the Internet Wikipedia article, you should do so - it'll help impress upon you the complexity of a system that scales to support billions of users (with arguable success). And then, hopefully, you'll marvel at the complexity (that you didn't even realize existed), because all you have to do is type "google.com" into the url bar, and everything magically works. In short: D-Star lacks that maturity.
In the Perth, Australia example above, I would need to specify four things:
1. KC2SUS <-- who I am
2. K2DIG <-- the local repeater that I want to talk to
3. xxx <-- the gateway repeater that the local repeater should use *
4. VK6RWN <-- the destination repeater (in Perth)
* In this particular case, K2DIG is also a gateway repeater, so I wouldn't have to specify an additional gateway repeater.
What I should have to specify is:
KC2SUS <-- who I am
VK6RWN <-- the destination repeater
What do I care how my voice gets there? I just want it to get there. Do you care whether your request for google.com gets routed across Level 3, Sprint or AT&T? I doubt it. What D-Star requires is that its users understand the network topology. If you travel cross-country or around the world, you shouldn't have to memorize how all of the repeaters are connected ... just store the frequency of the repeater you're going to use and you're set. Or you should be, rather.
When designing a system, a designer should first look at similar systems, understand how they work, and learn from its designers. So, to be clear: I'm not expecting D-Star to instantly be as mature and flexible as the Internet, but D-Star is a significantly less complex system, and they could have improved the D-Star design a great deal by employing a few more easily learned lessons from Internet routing ... BGP at the very least.