Mugr: from facebook to mugshotbook

Why should cops have all the fun? Mugr, which went into private beta yesterday, aims to bring face-recognition software to the masses. Its “face-based” search engine, deployed through its own social-networking site and offered to other social networks and applications through an API, connects images of people’s faces to information about their identities. So if you’re at a bar, say, and you see someone you’d like to find out more about, you can whip out your cameraphone, snap a picture, upload the shot to Mugr, and get a message back telling you who the person is.

Mugr notes that “face recognition, particularly face recognition that can be carried out over the web or with a cameraphone, invites immediate questions of privacy.” But the company has an immediate answer: “the technology that powers mugr.com is not so terribly different as that possessed by many governments and law enforcement agencies. As such, there is no reason that the public should not have the ability to do what it will with such technology. In the end, the technology at mugr.com is only frightening if its users make it so.” Kind of like assault rifles, I guess.

From a quick look at the site, Mugr has a long way to go before it’s prime-time-ready. But the concept seems inevitable. As cheap computing power, massive databases, and our seemingly unquenchable desire to publicize ourselves converge on the net, face- and voice-recognition programs will become routine tools for social networking. Mugshots wiil no longer be limited to criminals and celebrities. Everyone will get one.

3 thoughts on “Mugr: from facebook to mugshotbook

  1. Dan McClary

    Nicholas, after being alerted to your post, I felt compelled to say just a little something.

    First and foremost, I thank you for writing it– I’m amazed anyone even noticed our little launch.

    Second, I surely hope what we’re attempting isn’t as similar to assault rifles as one might initially suspect.

    When I wrote up the privacy statement for Mugr, I was faced with the challenge of condensing several months of personal debate into a reasonably small block of text. After all, I didn’t want to bore anyone with my own endless ramblings about the state of privacy in the modern world.

    In a sense, I believe the assault rifle-comparison is accurate. At least in the sense that no one can accurately anticipate what will *actually* be done with the device/service/technology. Assault rifles, hammers, even oranges– it’s impossible to say what any of these things will be used for when they are sold or distributed. Generally, assault rifles are designed to kill and maim, but they are occasionally used for peace-keeping. Oranges are typically eaten, but I’ve had the unpleasant experience of having one thrown at my head while on an evening jog.

    Is our technology potentially dangerous? Yes. Is it potentially useful? Yes. Do I know how people will use it? Not in the slightest.

    There were certain questions I focused on during my personal deliberations with the technology we’re attempting to develop. The largest of them were as follows:

    “Are we doing anything that a sufficiently motivated person couldn’t do, given time, resources, and patience?”

    Honestly, I don’t believe so. With the glut of visual information available on the web, a sufficiently motivated person could attempt to sort through a subset of it in pursuit of a person’s identity.

    “This technology might be interpreted as a form of surveillance. But is surveillance inherently different from other means or tools of oppression ?”

    This is the question that motivated the mention of face recognition in government/security sectors. In the end, I believe that surveillance is significantly different than rifles, or any other technologies that physically damage people and places. Surveillance, at its core, is effective because information that was thought to be private is, if not public, accessible by those for whom it is not intended. Thus, it loses its effectiveness by making some amount of information public.

    This is the reason that Mugr is an opt-in service. If you don’t post pictures, we *cannot* find you. If don’t register, we *cannot* find you. We have neither the resources, nor the desire to spy on the lives of others.

    Of course, the big question is this:

    “Does this sort of technology make society decidedly worse?”

    I don’t believe so. If nothing else, I think it raises an important point. Privacy is something we’re all entitled to. I, in fact, quite enjoy my privacy. However, there’s now such a tremendous amount of personal information floating about the internet that privacy may not be something we can *assume*. Rather, we may be entering an age in which privacy is something we must actively manage.

    If nothing else, I hope our little effort encourages people to ruminate a bit more on privacy and its management.

  2. Wayne

    When trying to determine how people will use something like this, remember that it will both be used for “good” and “bad” ways.

    Your system might offer an opt-in policy where I have to say it is okay to post my picture. I’m willing to bet that others will take an opt-out (if it is even offered) where your picture will be posted unless you ask otherwise.

  3. Bertil

    A friend of mine explained to me yesterday how he was made an account into Facebook by a well-intended and enthusiastic friend: a photo and his details were obviously available, and no need to confirm through a secure e-mail did the trick.

    How can you be positive that the opt-in is agreed to by the person, and not anyone using Google Image, or Facebook? E-mail confirmation?

    Are you considering massive photo imports, like Facebook does from e-mail friends lists?

Comments are closed.