My experience is that it depends a great deal on the types of the goals and how they're presented. The article mentions vaguely-defined, general self-improvement goals, like taking up a new hobby. I can see a public commitment being counterproductive there. But I'd bet that it works differently for more quantifiable goals, especially ones with a specific starting time, such as completing Nanowrimo, or telling your boss that you're going find a fix for bug XXX before Monday.

Quitting smoking? A hell of a lot easier to change your mind about if you haven't told anyone.

Of course, we're just going on the article. I don't have access to the original paper, but just reading the abstract already suggests (big surprise here) that the article is probably not a faithful summary of the study in question.