Last week there was a lot of talk about Facebook's new setup, which would allow for tighter integration and sharing of everything that people do, with music being a key example. Whether or not that's a good idea, I have no idea. To be honest, I think that it could make sense long term -- but the way that it's been implemented seems like a disaster to me, as I discovered when I logged into Spotify today. Apparently, Spotify is pissing off a ton of people by requiring a Facebook login to use the service now. I have less of an issue with that than I do with the fact that Spotify popped up a box telling me I had to connect to Facebook, but not making it at all clear what that meant. It notes that Spotify can share the details of what I'm listening to with others, but does not explain what that means. Will it share everything I play with everyone automatically? Will it give me the option of what to share? Will it give me the option of who I'm sharing it with? That's not clear at all. Even worse, nowhere is there any explanation of how or where I can find out more. Instead, Spotify just opens as normal.

It turns out that Spotify just starts sharing everything you play on Facebook, without even making it clear to the user that it's doing that. I couldn't find that info on my own profile. It was only after I asked a question and a Facebook friend told me what I was listening to that I knew the info was being shared. Even worse, how to turn it off is not clear at all either. Thankfully Lifehacker explains how to stop spamming everyone with what you're playing on Spotify. You can do so by unchecking the following box, which makes no sense at all: If you can't see it, it says: "Get personal recommendations by sending music you play to Facebook's Open Graph." But, what does "personal recommendations" have to do with anything? Why can't Spotify just be upfront and honest and say, "spam all your friends on Facebook with what you're playing"? Again, I recognize that some people want to do this, and I have no problem with people choosing to do it. My problem is with the way that Facebook and Spotify implemented this, where it's not even remotely clear what you're doing. Given Facebook's similar problems in the past (hello, Beacon) you would think that the company would recognize the importance of being clear and totally upfront about what info is being shared with whom and how to control it. Instead, it seems like the exact opposite.

Again, I'd have no problem sharing some of what I listen to if I have control over it. But, really, are any of my friends really going to want to know when I play the "lullaby playlist" I put together for my son? There are some friends with whom I have no problem sharing what I'm listening to, but plenty of others where it's just not something I'd share with them at all. And perhaps there are hidden controls buried in the preferences somewhere, but it's not at all clear, which leads me to now totally distrust Spotify and Facebook. Facebook I was already on the fence about, but I liked Spotify (and pay for a subscription). If these companies can't even get the basics right concerning how I can share my info, I'm going to have to look elsewhere. It's amazing how quickly a company can destroy a ton of goodwill.

As Google and Yahoo negotiate a possible search advertising deal, the New York Times wonders if it would put Google in hot water over antitrust concerns. With Google and Yahoo as the number one and number two players in the search marketplace, respectively, their combined market share would be in excess of 80 percent—close to Microsoft's Windows market share in the early 1990s. And that will inevitably invite heightened scrutiny from government regulators. The article ponders what Google would need to do to avoid antitrust problems, and concludes that basically, nobody really knows. We know that Google has to avoid "predatory behavior," but the experts the Times spoke with differed on whether a search deal would be anti-competitive. One scholar thought it would be, while others thought it could be structured so as to pass muster under antitrust law.

It seems to me that this is a serious problem. Well-designed law is transparent and predictable. Eric Schmidt should be able to talk to any competent antitrust lawyer and get at least roughly the same advice about what the law allows. Of course, this ideal is never achieved perfectly, but in many areas of law, it comes pretty close. Property law, for example, operates under well-understood, predictable legal standards. This allows companies, with the advice of counsel, to make business plans without worrying too much about the risks of an unfavorable legal ruling. When the law is uncertain, as with patents on software, it creates all sorts of problems because companies make their best guess about what the law requires, make business plans based on that, and then often they wind up in court and find out the judge has a different idea of what they should have done.

Too often, I think this is the case with antitrust law. Ed Felten (who will be my PhD advisor starting this fall) has a post on the lessons of the Microsoft antitrust case ten years after it was filed. Now, as a life-long Mac user and a fan of free software, I can certainly find plenty to criticize in Microsoft's actions, both legal and technical, during the 1990s. But the antitrust case still makes me uncomfortable because I don't think that if I had stood in Bill Gates' shoes between 1993 and 1998 that I would have been able to predict which business decisions would have ultimately been found to be illegal by the courts under antitrust law. Felten uses the words "kinder," "gentler," and "more moderate" to describe the new, post-antitrust Microsoft, but "kind" and "gentle" are hardly legal standards. Ordinarily, we expect—indeed, shareholders demand—aggressive pursuit of profits by publicly traded companies. Steve Jobs is a brilliant business man, but he's neither kind nor gentle. If the law is going to demand that Microsoft forego the aggressive pursuit of profit in favor of "kindness" and "gentleness," it had better have some pretty clear rules about how much moderation is required.

This is a particular reason for concern when we remember that antitrust law is often invoked for political purposes. A cynic might suggest that the reason Microsoft has received less antitrust scrutiny in this decade than the last is not because it is "kinder" and "gentler" but because it has hired better lobbying and PR firms to plead its case in Washington. Certainly, the commenters on Felten's blog have plenty of complaints about Microsoft's recent business decisions. And indeed, this is one area where Google's approach has been strikingly different from Microsoft's behavior in its early years. Microsoft mostly ignored the political arena in the 1980s and 1990s. In contrast, Google has been proactively building a lobbying empire to ensure that it will have plenty of ammo to use in future policy battles, antitrust or otherwise. That means that if people do start asking questions about monopolistic behavior on Google's part, Google will be able to dispatch their army of lobbyists to head off any moves toward litigation.

This is a savvy strategy for Google, but I think it's problematic for democracy and the rule of law. The law should be based on clear and objective standards, not who has the most influential lobbyists. If the New York Times article above is any indication, the law's requirements are anything but clear. And that's a serious problem.