Some folks have come along, having heard that the bounty has been claimed, and been purturbed to discover that the solution that claimed the bounty will be commercial. I think it’s worth taking a quick look at this.

Not all Bounties are Created Equal

The most famous bounty I know of was the ~$14k bounty for being able to install and boot into Windows on an Intel Mac. Now $14k is real money and pays for real development time. I assume that the folks thst set up the bounty realized they were appealing to a large audience and could raise this kind of real money . Claiming the bounty like this is almost like a contract job for a developer – do work, get paid.

Now let’s look at my BlackBerry bounty. Going into it, I knew two things:

1. the audience I could appeal to was much smaller
2. it was unlikely to be a simple solution¹.

Because of this, I did not place any kind of “the solution must be free/open source, etc.” conditions on the bounty. In fact, I did the opposite:

If the solution is not made available for free (ie. is packaged as a commercial product), anyone who contributed to the Bounty must be given a free license.

This was a considered choice, not an oversight.

Because I thought it was unlikely I’d be able to raise enough coin to actually pay² for the development of a solution, I used the bounty as a proof of market – showing would-be solution developers that folks out there want this capability, and are willing to pay for it. My belief was that by demonstrating this market, someone would be willing to take the time to develop a solution to meet the market demand.

It’s worth noting that the also-rans in the competition were also planning commercial solutions.

Now that we have that as the premise, let’s look at the actual solution that was developed.

I first heard from Daniel in September, and he was the first one to get anywhere close to a working proof of concept. Since that initial contact, we’ve been in touch via IM and e-mail several times a day and I was fully aware of the progress being made and the challenges being faced.

I know the solution has been rebuilt several times in dramatically different ways as various roadblocks proved to be insurmountable. I also know that he’s put a huge amount of time into it – basically full-time work for weeks now.

Not to mention, it isn’t completely done yet. I expect Daniel will have to invest at least as much time as he already has to get a salable product. By the time it is all said and done, this will be several months worth of effort.

Like me, Daniel runs a small development shop so I highly doubt he could have devoted this much time to this project had commercializing the solution been forbidden.

Commercial v. Free is a problem that solves itself

Now let’s look at the solution itself. I know it sounds kind of strange to hear that this was such a big undertaking. After all, the BlackBerry 870* already works as a modem under Windows. However, if it had been a simple solution to get the BlackBerry to work as a modem for OS X, the solution being commercial or not is a completely moot point.

If claiming the bounty had proven to be easy and the winner had not chosen to release the solution for free, someone else would come along and decide it wasn’t worth paying for and build their own solution. Eventually one of these would be available for free. The development community is generally very pragmatic and great at policing itself this way.

The solution that Daniel has built is not simple. First he created software for the BlackBerry that gives the BlackBerry a Bluetooth Dial-Up Networking (DUN) profile, something it did not have before this. Then he had to translate data in 3 stages:

1. Laptop to BlackBerry
2. BlackBerry to Proxy Server
3. Proxy Server to Internet

and vice versa. From the stuff Daniel has commented to me about what he’s been working on (debugging at the protocol level, etc.), I know that this work was way over my head. 🙂

Bottom Line

I believe I was very pragmatic in regards to the bounty terms, and I am quite pleased with how it has played out. The Mac/BlackBerry community will be getting a solution to allow them to use their BlackBerrys as modems, and as a bonus the solution will also work for Windows and Linux users as well. And since it will be a commercial product with a great deal invested in it, we know that Daniel will be motivated to support it.

Sounds like a win-win to me. 😎

1. As confirmed by Grant, the creator of the Bluetooth modem script for the BlackBerry Pearl (8100). [back]
2. Meaning: give a developer their normal hourly rate for the time they spent on the solution. [back]