Within hours after the Boston Marathon explosions, Google and the Red Cross had uploaded databases helping people locate loved ones. Shortly after, Internet communities including Reddit and 4chan launched their own online search mission: Find the perpetrators. Sift through millions of megapixels, through a thousand individual Zapruder films, and identify the person(s) who planted the devices that killed three people and wounded nearly 200 more.

It’s the natural impulse of 2013. The world is taped and tagged, uploaded and user-generated. The world is so thoroughly surveilled, it seemed inconceivable that the villain wouldn’t have inadvertently been captured wandering through the background of someone’s freshly posted Vine clip.

So online users set to work, playing a version of Where’s Waldo in which nobody knows who Waldo is, what he is wearing, if there is one Waldo or four.

On a 4chan Imgur.com site, commenters used red circles to mark up anything they found suspicious on user-submitted photographs: A man who ran too fast from the explosion; another man who did not run fast enough. A man who did not seem to be paying close enough attention to the race, another man paying too much attention. On various amateur detective sites, special focus was given to a navy fleece-adorned male dubbed Blue Robe Guy because of the unique way he appeared to be holding his backpack.

Participants weren’t profiling so much as they were sketching — using screen-grabbed sniplets of human behavior to draw caricatured conclusions. On Wednesday, what users focused on depended on which rumor had most recently circulated in the news: Was the suspect a dark-skinned man? A white male in a baseball cap and gray hoodie? The Reddit board filled with fuzzy close-ups of private citizens and debates over whether strangers’ sweatshirts were gray or black.

Over the course of the past two days, an irony has emerged: The danger to our privacy might not actually be the long-feared Big Brother, but rather hundreds of thousands of Little Brothers — poking, prodding, desperate to be included, acting with enough enthusiasm to ruin people’s lives.

More than once, temperate-minded participants rang in with words of caution: “Does anybody remember Richard Jewell?” one such poster wrote, reminding the Reddit community of the Atlanta man falsely accused of planting a bomb at the 1996 Olympics. “You may get lucky and your pet suspect turns out to be the bomber. But I’ve seen at least 10 people singled out, and not all of you are correct.”

On the Atlantic magazine’s Web site, tech writer Alexis Madrigal put it more bluntly: “This is vigilantism, and it’s only the illusion that what we do online is not as significant as what we do offline that allows this to go on.”

There has been speculation that all of this aspirational sleuthing represents a pivotal moment in crowdsourcing: that this would be the instance in which the skill of the masses was tested against the skill of trained police, who have experience but not numbers, who care deeply but who are also bound by rules.

More than that, though, it represents a pivotal moment in crowd responsibility: one in which we road-test not our skills, but our abilities to restrain them. One in which we evaluate not the police, but how well we are able to police ourselves, tempering the freewheeling nature of Internet collaboration with the due-process, innocent-until-proven-guilty rules we attempt to obey offline, in real life.

The technology is advanced enough to aid in catching criminals. The more interesting question is whether we are advanced enough to use it well.

A final note: As of press time, police said that the camera of “special interest” to them wasn’t the one that produced any of the uploaded images being circulated online, but rather the security camera stationed at a nearby Lord & Taylor. The film from this camera had not been released to the public.