Reading this blogpost left me wondering about some companies/people that do not understand the notion of "active check-in" on Foursquare (or now Facebook Places). See for yourself: "The active checkin requirement is one thing holding back location-based social networks (also called “geosocial” networks) from widespread adoption. (According to Forrester, only 4% of Internet users have ever used them.)". It reminds of the opinion about Foursquare stated by this analyst: "It seems like the marketplace has taken a step back 5 years. All of a sudden people seem to be convinced that this kind of technology -- where you have to actually remember to tell people where you are -- is the best thing since sliced bread. (...) The crucial flaw with FourSquare et al., is that it's based around manual push notifications." For this kind of analyst, an explicit interaction (doing a check-in) is perceived as backward and lame. In engineering circles, this sort of argument is highly common and I would refer to it as the "automation bias", i.e. the firm belief in automating whatever human activity that can be transferred to computers/machines (which is grounded in strong positivist ideas about progress obviously). The comments I quoted above do not acknowledge the reason why interaction designers have chosen this solution over, say... CellID triangulation or a nearly magical GPS signal detection. Readers here have certainly read my opinion about this topic here, there (or in French). But I think it's worth repeating the claims here:

Of course, decreasing users' burden is an important adoption factor, I fully acknowledge it. However, automating this can be perceived as a threat by people who feel that they will loose control of their personal data. It can also be problematic for some individuals because this automatic feature will make explicit situations they don't want to make public. Technologies should be "conservative of face" as described by Adam Greenfield some time ago: wherever possible they not unnecessarily embarrass, humiliate, or shame their users. See for example this comment in the original Mashable blogpost: "I go places that I am not always proud of (think Waffle House at 2:30am) and at that point (think less than sober) I can see myself forgetting to turn the auto check-in off. (...) there has to be a better way for it not to be obtrusive, but still controlled.". Letting people doing manual check-in is more respectful of people's habits and, above all, it enable people to lie (which has always been a good adoption factor). This is why the proposition to have an intermediary solution is interesting: " the app would have some sort of pop up/notification that lets you know you are in a check-in-able location"
What is showed in my research: self-reporting one's location has a value in itself. Declaring your whereabouts is not just a piece of information, there is also an intention attached to it. Say I'm in a Bar and the name of this place is sent to my colleague, it's both a statement about where I am and an act of communication that tells others that they can act upon this information (to draw inferences about my availability or my willingness to interact socially for example).

Having said that, the problem is not about the manual check-in but instead, it's about the extent to which people use this feature. I know "checkin fatigue" is important... but doing it manually means that the place where people check-in are more meaningful to others... since Foursquare removed the leaderboards (and hence the incentive to gain as many points as you can), users I have interviewed said that they stopped checking-in everywhere (supermarkets...) and only made their position available when they wanted to meet others or access to certain information. I am curious about this and we are currently launching a user study of Foursquare to understand this kind of issue.