The Berkeley Public Library is continuing its unfortunate tradition of seriously misrepresenting the facts about radio frequency identification (RFID) as it seeks a waiver of the Nuclear Free Berkeley Act from the Peace and Justice Commission (P&J), and eventually from the City Council.
The library wants to sign a multi-year maintenance contract with 3M, a company that has taken over this function from the original RFID system seller, Checkpoint—and 3M has refused to sign a standard city form verifying that it does not, and will not for the life of the contract, do “work for nuclear weapons.”
In her Nov. 10 letter to the Peace and Justice Commission chair, seeking the waiver, Director of Library Services Donna Corbeil made no mention of the possible alternatives to obtaining maintenance from 3M, other than getting another RFID system from a different vendor. But the law requires any waiver consideration to evaluate alternatives and their costs. The most obvious alternative is to switch systems and install reliable, inexpensive bar code technology that continues to be used by 99 percent of libraries today.
The bar code option would include using existing or new bar code checkout equipment—and eliminating expensive and unreliable RFID self-service checkout machines. Bar code self-service checkout machines could be substituted. Alternatively, staff could be provided as necessary to replace these machines altogether. That could provide more personal service and avoid the high cost of purchasing self-service machines that cost about $20,000 each.
The library wants to keep its existing system, and its strategy appears to be to understate the ongoing costs, overstate the supposed benefits, and make any real alternatives look unthinkably painful and expensive. But the union previously recommended a return to bar code technology. And the Oakland Public Library did just that—it pulled the plug on RFID and went back to bar code technology several years ago at its Cesar Chavez branch.
Past benefit claims for RFID have typically been unsubstantiated—and were actually incorrect.
1. RFID was supposed to cut repetitive stress injuries (RSI) dramatically—but it did not.
Actually, RSIs had been declining and went to zero—that is, none—in fiscal year 2003-2004, the year before RFID was installed. In fact, there are serious questions as to whether bar code technology necessarily causes any RSIs. As Lee Tien and I wrote in our March 4, 2005 Daily Planet Commentary analyzing Workers Compensation at the library, “RFID Should be Canceled Immediately,” “There were no RSI claims in 1998, 2000, and 2004, and only one RSI claim worth $1,008 in 1999.” Bar codes were used in all of those years.
Our analysis of the library’s Workers Compensation history concluded: “Simply put, the documents we received from the library contradict the library’s claims that RSI is a major financial burden [when using bar code technology].”
Indeed, RFID appears to have actually increased injuries. A Daily Planet report on April 6, 2007 said that “because the [RFID] system does not consistently function properly” a library worker reported “repetitive stress injuries are up.” The reporter added that this was “something of which Library Director Donna Corbeil says she is unaware.”
2. Tag prices were supposed to go down—instead, they went up.
The Berkeley Daily Planet reported on May 18, 2007, in “Library Budget Raises RFID Questions,” that Donna Corbeil said media “donut” tags for CDs and DVDs cost $2.12 and regular tags cost 77 cents each. The story noted this was up from prices reported to the Board of Library Trustees (BOLT) 19 months earlier, when the cost was $1.15 for donut tags and 60 cents for regular tags. That represents an 84 percent price increase in donut tags and a 28 percent price increase in regular tags. The story quoted the technical services manager as saying in 2005, “There is general agreement that in the near future the costs of the RFID security tags should drop below their current 60 cents apiece.”
3. RFID was supposed to allow simultaneous checkout of multiple items, making the process easier and quicker for staff and public. It didn’t turn out that way. Checkouts typically occur one at a time, reportedly because reading multiple items is simply too unreliable.
Here are some recent misrepresentations.
The library sent a Dec. 5 letter responding to questions posed by two of three members of a Peace and Justice subcommittee which was appointed to do further research. The responses were quite unresponsive and highly misleading.
Example one: The answer to one question indicated that RSIs had gone down with installation of RFID, when in fact they went up.
The letter says, “Staff repetitive stress injuries fell significantly from six incidents in fiscal year 2002 and five in fiscal year 2003, to a single incident in fiscal year 2005 following the installation of self-checkout at BPL.”
But the library’s list of years left out a crucial year—2004. In that year, before RFID installation, RSI claims were zero. In other words, RSIs had dropped to zero before RFID was installed, and actually increased in the year following installation.
Example two: Question 4 asked, “How many libraries in the U.S. use RFID technology? The library’s reply was: “According to trade magazine ‘Computer in Libraries’ Feb-08 in a non-comprehensive survey there are 7,741 self-checkout units installed in U.S. libraries.” What the library did not make clear is that many self-service checkout units are not RFID—but typically use bar codes—such as Berkeley Public Library before installing its RFID system, and as currently used by San Francisco Public Library and others. The last figure for RFID installations that we have seen is about 300. There are more than 30,000 libraries in the U.S.
RFID in the Berkeley Public Library has been an expensive failure. It has not lived up to performance and cost promises, and there are hidden costs including privacy threats and potential health risks. Bar code operation is far cheaper and more reliable, with bar code stickers and security strips costing from one-quarter to one-tenth the RFID tags. A bar code system is an alternative that would allow the library to avoid signing a contract with 3M, a company that refuses to sign that it does not, and will not for the life of the contract, do work for nuclear weapons.
The Peace and Justice Commission will consider the waiver at its next meeting, Monday, Jan. 5, at 7 p.m. at the North Berkeley Senior Center, Hearst and Martin Luther King, Jr. Way. Please attend, and send letters to the commission, with copies to the City Council care of the city clerk, as the City Council has final say over whether a waiver is granted.
Peter Warfield is executive eirector of Library Users Association.