Updated on September 19 at 3:15 p.m.

Just before 8 in the morning on Monday, cellphones chimed in unison across New York City. It wasn’t the sound of text messages: It was a dissonant siren, repeated six times, accompanied by a short note. “WANTED: Ahmad Khan Rahami, 28-yr-old male. See media for pic. Call 9-1-1 if seen.”

New Yorkers have received emergency alerts before. In extreme weather, like Jonas earlier this year, the jarring tones have instructed people to move indoors and stay there. AMBER alerts, which use the same system, occasionally ask people to look out for abducted children in their area.

But the city never has sent an alert like this. Typically used only to keep people safe, the Monday alert asked New Yorkers to join the police’s search for a wanted man.

Emergency alerts in New York City come from the Office of Emergency Management, one of tens of thousands of “alert originators” in the country that are authorized to push alerts to mobile subscribers in their jurisdictions. The emergency-management office coordinates with other agencies in city government to send alerts, which must either be official AMBER alerts, or involve “imminent threats to safety or life.” (Weather-related alerts generally come from the National Weather Service.)

At about 7:45 this morning, the New York Police Department reached out to the Office of Emergency Management to request that an alert be sent to all five boroughs in New York City about the primary suspect in a series of bombings in New York and New Jersey. Officials decided that the alert related to an imminent threat, and drafted and sent the alert within 15 minutes, a spokesperson for the NYC Office of Emergency Management said.

How did the push get sent so quickly? “We try to limit the approval process,” said Nancy Silvestri, the agency spokesperson. “We are empowered to make the decision locally.” That allows timely notifications to go out without being bogged down in red tape, she said.

Rahami was arrested in New Jersey after a shootout with police later Monday morning. It’s still unclear how police tracked him down.

The Monday alert was the second notification that had to do with the weekend’s bomb scare. When an explosion resounded in Manhattan’s Chelsea neighborhood on Saturday, New York City officials sent an alert only to cellphone users around that neighborhood, warning them to stay away from windows. That alert was similar to an emergency warning sent after the Boston Marathon bombing in 2013, which instructed people near the blast to shelter in place. (The Boston alert did not, however, identify a suspect.)

Officials are able to target cellphones in a specific area because of how the technology works: Once a local agency submits a wireless emergency alert and a specific area, it’s passed along to wireless carriers, which then send the messages to their customers from cell towers in the target zone. Most providers distribute the messages on a channel that’s separate from the usual ways cellphones receive voice, SMS, and other data from cell towers, in order to be able to deliver alerts even when voice and text channels are congested.

New York City’s Office of Emergency Management was the very first alert originator to adopt the system when it was first introduced in 2012. Four years later, the technology has some of the same limitations. It only allows messages to include 90 characters—less than a tweet’s worth—and it doesn’t support attachments, like photos. (That’s why this morning’s alert had to point people to the media for the suspect’s photo.) Messages also can’t include tappable URLs or phone numbers.

Last year, the Federal Communications Commission tried to overhaul the alert system, but wireless carriers rebelled, Motherboard reported last month. The commission’s recommendations included extending alert lengths, enabling web links, and improving their crude geotargeting technology. The changes haven’t yet been adopted.

Because of the length limitations, Monday’s alert contained very little information about the suspect NYPD was looking for. People who went searching online for more about Rahami would have found a photo, and the description police shared with the media: A 5’6” 200-pound male with brown hair, brown eyes, and brown facial hair.

Some criticized the city’s decision to push the alert to the entire city.

Publicly identifying a suspect-at-large can be risky. In Boston, a twisted string of events led news organizations to wrongly identify 22-year-old Sunil Tripathi as the marathon bomber. And after five police officers were killed in Dallas earlier this year, the Dallas police department tweeted the photo of a man with a gun slung over his shoulder with the caption, “Please help us find him!” That man, Mark Hughes, was not involved in the violence, and was complying with open-carry laws. His attorney said he received “thousands” of death threats after the tweet.

When I asked the Office of Emergency Management whether the lack of information could put some New Yorkers in danger of being profiled and harmed, the spokesperson said the city had decided that the importance of disseminating the information “outweighed” the limitations of the alert.

Peter Donald, the NYPD assistant commissioner for public information, said the first priority was to alert “every New Yorker” to the identity of the person being sought by authorities—anyone troubled by the lack of information in the alert could go online to find more. Since Rahami was caught a few hours after the alert went out, Donald said, the “proof is in the pudding.”