Free as in Free Speech AND Free Beer (sorta)

I've always felt part of open source uniqueness is that it is so hard to imagine outside of the software industry. Few industries have the same dynamics as the software industry where people can do something voluntarily and have it product anything commercially viable. I've always said that the pressure people put on Microsoft to expose source and the subsequent derision for not doing so was a little unfair given you don't see it in other industtries. Or do you?

People are definitely looking for ways to riff on the open source movement outside of software. Korby Parnell brought an interesting story to my attention about how a university is trying to open source a beer recipe. I don't know if this is a joke or not, but the fact that someone would even think of this is entertaining on so many levels. Their goal is to become the "Linux of beers" (I kid you not). The recipe is protected under the Creative Comons License and you can feel free to extend and sell. If you make money selling their unique beer, you have to give them credit and publish any changes you make to the recipe under a similar license.

Meanwhile, singer-songwriter John Mayer has decided to go "open source" with his latest song. Here's the catch--there are lyrics, but no music. He's letting everyone write the song for him. Different riffs on the same kernal--hmm, could this be the start of a trend? I suppose we've seen this for years in the other direction where alternate lyrics are provided for the same song. Weird Al Yancovic has made a career on this with his parodies and Axl Rose did it to himself with "Don't Cry" and released both versions at the same time. But those are both instances where the entire songs were written and then someone changed the song. Mayer is taking an interesting approach--"I've got the start of a good idea and I want to see if someone wants to help me finish it". Many projects on SourceForge and GotDotNet are just that--people with great ideas and great intentions that are looking to get something done and looking for a little help to do it. About 10 years ago, the Smashing Pumpkins put out a CD-single for the song "Zero" (great song) and it had a 23-minute track called Pistachio Medley with at least a hundred five-second riffs of different songs that Billy Corgan essentially started by never finished--some really good and some not so great. I always thought that some enterprising musician would take one of those riffs and write his or her own song. Who knows--maybe if he caled the track "Open Source Medley"...

Another example of open-source type behavior is the coolest baseball site ever invented for a stats-geek baseball fan such as myself: I am almost ashamed of how much I love this site. I has every boxscore and player stat ever recorded through the 2004 season. No player or game is too obscure for Retrosheet. It has the boxscore of pretty much every game ever played, including the first Oriole game I saw at Memorial Stadium (in 1985) and the first game I ever saw at Camden Yards (in 1992). It also has the career details of all of my all-time favorite Orioles. I'm not just talking about Cal Ripken--that'd be too easy. I mean guys like Steve Stone, Al Bumbry, Mike Boddicker, Eric Bell, Craig Worthington, and Ben McDonald. Everyone is in there, no matter how obscure. Yet, for all the effort it takes to log every game ever played, these people do it strictly for the love of the game (message to all you baseball-haters: the sport does indeed inspire that kind of passion). In fact, there's no attempt to make money as Retrosheet users are free to make any desired use of the information, including selling it, giving it away, or producing a commercial product based upon the data. Retrosheet has one requirement for any such transfer of data or product development, which is that they get credit. It feels a little like open source statistics.

But where do you draw the line? I was thinking about the pharmaceutical industry, for example. Patents are very tightly held there and the free markets reign. Years of research go into finding drugs that manage high blood pressure, fight AIDS, and even deal with embarassing sexual dysfunction. Now we're talking about lives here. Technically, we pop these pills without knowing 100% of what's in them (or at least I do). The question is: are we better off with the secret ingredients unveiled and some publish-back mechanism? I don't know. On the one hand, you never know how many lives would be saved. There is the story about a drug company named Cipla doing this for third-world countries, violating patent laws, but providing drugs at affordable prices. They can be considered saints, but make no mistake--they are doing it for the money and the fact that they don't need to do the level of research that the big companies invest enables them to sell at the steep discount. Before we start sobbing for the pharmaceutical companies (please, no comments on their marketing budgets--I know that issue), I think I would be more concerned about the potential product in an open source market. I don't know if I'd trust what came out of the "collaboration". It's one thing to do it with a song (heck, I didn't like John Mayer anyway), but when things get more serious and my health is at stake, I trust the Pfizers of the world more than the random dude that messes with pharmaceuticals. To me, I think that is how I think of software (not that software is a matter of life-and-death, per se). I think there is some software that leans closer to the music and beer in that you can take it or leave it or at least manage your reliance upon it (tools like NAnt, NUNit, or NDoc, for example). But with the operating systems or mission-critical software, I prefer to leave it to the professionals. Obviously, this probably stand as a Microsoft-centric view (hey, I work here for a reason) and I see why some people still believe going the Linux route is where safer ("there's no way a buffer overrun gets by that many programmers!"). Plus, I must admit that the big companies are certainly not infallible (in the case of pharmaceuticals, see VIOXX). But in those cases, there is some level of indemnification (Merck is paying through the nose for VIOXX ) and knowing what these companies have at stake does make me feel more comfortable about how they stand behind their product. And, of course, there is always the FDA to regulate the industry while you don't have that piece of mind in software. Looking at the latest open source startups, that's what companies like SpikeSource are trying to do--validate open source stacks the way the FDA validates drugs. Is that enough? Only time will tell, but for now, I am still betting on Microsoft...

{Red Hot Chili Peppers - By The Way}