The “proper” way to do this kind of multi-NIC setup involves a terrifying trip through proxies and subnets and the nuances of the IP protocol. The “better and works most of the time” way is to simply tell your computer, “hey, when you need to access something, try to get it through the iPhone first, and then if that doesn’t work try the corporate LAN.” This forces your web browsing traffic through the tether, yet still allows you to access internal IP addresses like network file shares, printers, etc.

The way you do it is by adjusting what’s called the “interface metric” on your network adapter. Every network adapter has an interface metric that tells your computer how expensive it is to get data from that network. Usually this metric is determined automatically, but if you want your computer to prefer your iPhone instead of your LAN interface, you punish the LAN – set the interface metric for the LAN to be much higher than for the iPhone tether. I set mine to 100, and it seems to work pretty well.

Of course, you have to first enable tethering on your iPhone. I’m too lazy to write up the details of that here but a quick google search will point you in the right direction. You’ll know you’re set up when you see your tether’s network interface show up in the network connections window (look for the one with a device name like “Apple Mobile Device Ethernet”).

Once you’ve found it, first thing you probably want to do is rename it to something more descriptive than “Local Area Connection 3.” Rename it “iPhone Tether”, so that things look similar to this screen:

Now, right click on the NON iPhone network (the corporate LAN, in the screenshot above, it’s the selected “Local Area Connection 2″), and say properties:

Next, select “Internet Protocol (TCP/IP)” and click Properties:

(Your screens here may look different depending on your network config). Click Advanced, and then uncheck the automatic metric option, and put in a metric of 100:

Click OK back through all the dialogs. Then, go into the properties for your iPhone tether and check the option to show it in the notification area when it’s connected. That way, you’ll see when it’s in use.

That’s about all there is to it, you may have to restart some apps but you should see data start to flow through the phone. When you pull the iphone from its dock, your computer will fall back to the corporate LAN, and when you plug your phone back in, your apps will stampede towards the iPhone’s “lower cost” connection. Anything on your internal network is still visible, but your computer will try to get to it through the iPhone first, so there may be a slight initial delay as it discovers that’s a dead end.

Enjoy, but remember, I make no warranties, and I’m not a network guru. And there may be some legitimate security concerns here, since you’re essentially opening your work PC to the internet at large, without your corporate firewall to protect you. So, be careful, and take the appropriate security precautions. It’s working well for me, hopefully it’ll work for you too. Browsing through the phone is of course slower than the LAN, but this might be a good thing, since it’s a slight disincentive to waste time at work. Overall, this setup is better for you and your company: you get more privacy, and unfiltered internets, and your employer doesn’t have to waste bandwidth streaming your stupid cat videos.

I’m tearing myself apart trying to decide which platform(s) to create it for. For starters, there’s the obvious: the iPhone. Tons of free engines and code samples make creating a game for the iPhone a no-brainer. Unfortunately, that means there are thousands upon thousands of them, and getting any sort of attention in that crowded market would require more work to market and publicize than it would take to actually create the game.

Then there’s the G1 – its hardware is on par with iPhone, yet it has a much smaller user base. This is a blessing and a curse – less users means fewer sales but less competition. Same deal with Blackberry – I don’t have the numbers to back this up but my hunch is that the number of iPhone users dwarves both the number of G1 and Blackberry users. At any rate, Don’t Dial iPhone is outselling Don’t Dial G1 by at least a 10:1 ratio (but, that’s with a very iPhone-centric marketing effort). Maybe with some G1 exclusive marketing I can get that to 5:1.

The Blackberry has in its court a much less crowded market, and a push for quality (RIM has explicitly said they’re not going to tolerate the crap apps that Apple does). Also, I’d be a fool to sell my game for anything above $1 on Apple’s app store, but I could sell for $3, maybe $5 on Blackberry and remain competitive. G1 is probably the same story, but muted slightly due to the number of free apps, as well as the G1’s open source spirit, which attracts a user base that’s a bit less comfortable with paying for apps.

The really discouraging thing is that technically, the three platforms couldn’t be farther from each other. Don’t Dial has taught me that getting any meaningful code sharing across all three platforms is difficult if not impossible. I’ve toyed with the idea of building a cross-platform metalanguage that would “compile” down to ObjC or Java code, but doing that is a lot of work, and then you have to abstract all the functionality of the APIs you need.

Someone, somewhere, is doing this, or will be soon, so there’s that too – should I develop my own abstraction, or wait until one gets released? The phonegap guys are making good progress, but they’re all Javascript, which makes sense for web apps, but for games is a harder sell.

Lots of questions, very little answers.

I’m missing the GDC. Every year before the GDC I think “yeah, too expensive” and then during GDC week I get all wistful and weepy about not being there for it. This has been occurring now for several years and is occurring yet again this year. And every year, I say this: maybe next year I’ll go back and say hey to all the gamedev.net guys and we can reminisce about how much better the conference was when it was in San Jose (or maybe not). Those halcyon days of the early ’00s!

And yeah, you can say I’m biased, but here’s truth: when it comes to GDC coverage, gamedev.net has consistently blown every other site completely out of the water. Even gamasutra. The gamedev team works their asses off, and I guess that’s my silver lining – all I have to do is sit in front of a browser and refresh obsessively.