They started counting the gunshots near the Johns Hopkins University on Nov. 20.
So far, 93 sound sensors have detected two noises loud enough to register on a new computer system designed to pinpoint gunfire by the explosion that forces a bullet from the barrel of a gun:
* Dec. 2 at 2600 N. Calvert St., at 12:34:05 in the morning.
* Dec. 3 at 4 W. University Parkway, at 8:38:32 in the morning.
Only two? That's great news for residents of Charles Village and parts of Homewood, Abell and Harwood - residential neighborhoods east of the Hopkins campus in North Baltimore. Even better is that the bang heard on Calvert Street was nothing more than firecrackers; the noise on University Parkway remains a mystery.
But it also raises a question of whether this new SECURES Gunshot Detection System is installed in the right place to test whether it works and whether it helps police respond more quickly to the reports of gunfire that, unfortunately, are a fairly common call for Baltimore police officers.
There certainly are neighborhoods where the sounds of gunshots seem as common as cars backfiring, and Charles Village and the Hopkins campus aren't on that list. But, in Hopkins' defense, the company gave them the system for free - allowing city police to evaluate it and maybe buy it if they like it, and the university to boast it is the one of the first college campuses in the country to count gunshots where its students live and play.
The sound sensors are wireless, but they have to be close to a permanent base unit, so the system can't be easily moved to, say, East or West Baltimore.
But university officials did broaden its reach east of the campus to include areas from Howard Street in the west to Barclay street in the east and University Parkway in the north down to East 25th Street in the south.
I have to admit, the system is fun to watch at the Hopkins security office in Remington, where dispatchers keep an eye on a bank of computer screens and live video streaming in from 155 surveillance cameras. A large screen on the wall shows a map of the area covered by the gunshot detection system.
When four sensors register a noise - they can differentiate a car backfire from a gunshot - a starburst appears on the map, which is superimposed on Google satellite imagery showing the location within 10 feet. The Hopkins dispatcher can then notify city police and get an officer there quickly.
The incident on Charles registered as a gunshot but was picked up by only three sensors because one was broken. Martin L. Beauchamp, the security systems manager for Hopkins and a retired Baltimore police major, said officers responded but found nothing.
He sent his security guards out the next morning "and we found a pack of firecrackers" near a tree 9 feet southwest of the back of a rowhouse - precisely the location the computer had spit out. Beauchamp said that had the fourth sensor been working, the system probably would have noted the sound but recorded it as unlikely to be gunfire.
The second hit came just a few hours later at West University Parkway and Canterbury Road, near Homewood Field, in the middle of the block and the middle of the street. Beauchamp said two security guards were nearby and also heard the noise. "They thought it was a gunshot or a car accident," he said. "We weren't able to figure out what it was."
Other cities, including Washington, use similar detection systems, though Baltimore officials have expressed skepticism about their value. City police will be at Hopkins on New Year's when celebratory gunfire sometimes drowns out fireworks, and perhaps that will be a good test.
"It's another layer of high technology," said Edmund G. Skrodzki, the director of Hopkins security and a retired U.S. Secret Service agent. So far, he said, he's impressed by the system, and it seems to me with all the reports of gunfire, city police might want to explore this.