How would you design the browser ballot?
Posted April 16, 2010on:
I recently stumbled across an old post by Mitchell Baker of Mozilla. The post is essentially an offer to the EC to advise on what steps might be appropriate to implement a remedy for Microsoft’s illegal behaviour. As he says “There are separate questions of whether there is a good remedy, and what that remedy might be”. It’s one of those scenarios in which it’s interesting to imagine what you would do if the responsibility for a judgement suddenly (inexplicably) ended up in your lap.
The problem in need of a remedy he nicely sums up as Microsoft “promoted IE through activities that the US Department of Justice and the U.S. Courts determined to be illegal. As result, Internet Explorer ended up with well over 90% market share.”
Its interesting to think through what one might expect from such a remedy. When a wealthy drug dealer is busted their assets are seized and sold since they are the proceeds of crime. The proceeds of crime in this case would be Internet Explorer’s market share. In my head any remedy should be aimed at ensuring the perpetrator does not benefit from their crime, otherwise the incentive is there to commit another crime. The difficulty is that you can’t just confiscate market share.
The remedy that was eventually decided upon was the browser ballot. There were a lot of different ways it could have been implemented but in the end a web page was created that would display the major browsers to the user in random(ish) order. The user would decide what to install from there.
While think I might have offered a similar solution myself in my hypothetical world, I don’t think I agree with the implementation. Displaying a randomly ordered list of browsers gives them all an equal opportunity to be seen but the other browsers a starting at a deficit; to the user the are looking at a list, that has the program they have been using (possibly for years), or a bunch of stuff they have never heard of. An equal chance a being first on the list is not likely to overcome that.
It is a fact that for many users Internet Explorer IS the internet. For many more, even if they are aware of difference between the browser and the website displayed within it, Internet Explorer is the only browser they have ever used. The familiarity that comes with that is also the proceeds of the crime. While the browser ballot is definitely the way it should be, the problem lies in the fact that its the way it ALWAYS should have been.
With Microsoft’s illegal actions creating a bias in the mind of the user, I would be very curious to know what the actual outcome of the browser ballot is. Thinking about it leaves me with more questions than answers:
At what rate are users installing other browsers?
Is this rate sufficient to deprive Microsoft of the “proceeds of the crime”?
If it isn’t, should the ballot favour the other browsers and how much?
If it is, on what timeline?
Is that timeline short enough to feel like an ethical solution to people who have to live with the consequences of this illegal action?
Is this rate high enough to feel like a punishment rather than a market fluctuation?
If the goal is shifting market share, what is the ideal we are shifting it towards? How would you know when to stop?
While there are probably no “right” answers to most of these, just better answers and worse ones. For me the browser ballot falls somewhere in the middle.
What would your remedy look like?