The RFID blog is back, now managed by Margaret Hazel of Eugene Public Library. I didn’t remove the old links from my aggregator, hoping against hope this blog would return.
RFID came up in the audience-question portion of the Top Tech Trends discussion at ALA Annual. My response–and I think that of others–emphasized two concerns: return on investment, and interoperability.
I think our profession’s RFID privacy questions, while well-intended, often got overblown, and even the latest version of the RFID document from PLA/IFC has some of that spittle-flecked hysteria that so cooled me to the Electronic Frontier Foundation, which had something on the order of a vendetta about Berkeley Public Library’s RFID installation. (Anyone who needs Peter Warfield as a front man is scraping the barrel.) For several years I had served on the Pioneer Awards committee, but this whole mishagosh pushed me away.
I pretty much kept quiet on RFID for a long time because I felt the RFID installation was being used as a wedge issue by a library determined to stay in 1987. It was an uncomfortable spot to be in (not helped when various interested parties tried to put words in my mouth). Well, the last-century crowd won, and they can stay in 1987, for all I care. But I feel I can talk about RFID a little more these days.
A big-library director approached me after TTT and said if the BLD had to do it over again, the BLD would not have installed RFID at that point. The ROI just wasn’t there, in part because manual checkout was still required for AV material, and because library RFID is so vendor-specific, an RFID implementation undercuts a half-century of work in library resource sharing… locking a library into a technology that ain’t easy or cheap to implement or undo. That, and the fact that it can’t be used for AV materials–which can be much of a library’s circulation–and that for self-check, as grocery stores know, barcodes are a reasonable system for now… there are a lot of arguments for not going to RFID, even for opening-day collections.
What frustrates me most about public libraries and RFID is that had a few libraries held out for a uniform standard early on, their sheer fiscal clout would have been important. But most of the discussion (and even some seemingly authoritative “white papers”) focused on whether or not RFID affected user privacy.
It’s not too late. After the TTT discussion, a librarian named Leif from the Danish national library approached me to describe how in Denmark they had agreed to push vendors to a standard, and recommended against installations until the standards became real.
Standards. Standards. Standards. Repeat after me. Standards. Standards. Standards.
Oh, and my other technology philosophy (which quite intentionally holds two opposing thoughts in tension): Pioneers get the best territory, but the second mouse gets the cheese.
Posted on this day, other years:
- To be cool is to be young and male? - 2007
- Atonal Bell - 2005
- Urban Legends Top Searches Feed - 2004
- Bloglines Improvements - 2004
For obvious reasons I’m not going to comment on the bulk of this entry.
However, at least one vendor has found a solution for the AV problem. It works and at the library I am most familiar with the circulation for all items, including AV is accomplished with 70% self-check by patrons. I suspect that in the near future (the AV component just completed its beta testing at its first library site) that will rise much higher.
There is no reason that RFID has to cut into resource sharing. Many of the libraries in the Bay area that are using RFID are members of LINK+ along with libraries that are not using RFID. Barcodes haven’t disappeared off RFID items not just because of resource sharing but because of patron self-renewal etc.
Because barcode systems have not found a reasonable way to do most AV self-check and maintain security they are not an adequate alternative to a system that does allow for AV self-check. Now that at least one vendor has found a sensible way to provide that service I suspect other vendors will quickly follow.
ROI on technology implementations can be hard to measure in early days. We heard high levels of satisfaction from patrons on the service. Lines were shorter, privacy increased as you could truly check out your items with no staff intervention. Short term, repetitive stress injuries in staff have greatly decreased. The time that staff was spending to check-in items has decreased by 60%. We staffed one less position at the circ desk. While we still had two staff at the circ desk at all times we frequently placed one in front of the desk to assist patrons learning the system. These are measurable ROI’s and, as I said, its early days since the end of beta testing on the AV system.
ROI is always going to depend on staffing costs. I talked to a director in North Carolina where the annual cost for a library assistant was in the low $20,000 (including benefits). At Berkeley, the annual cost for a library assistant, including benefits is $67,000. A system that cost $650,000 to purchase and $500,000 to implement, will quickly pay off if you simply don’t have to add staff as circulation increases there while the North Carolina library might never be able to justify it.
First, a small question, why are barcodes necessary for patron self-renewal? I log in, I see the item, I renew…?
Second, on resource sharing, you base your rationale on the presence of a barcode. If everyone migrates to RFID, and libraries are using competing systems, eventually someone’s going to ask why they are barcoding AND tagging, and why the barcodes are required for resource sharing. The answer can’t really be because library systems are using competing and noninteroperable systems.
I agree that RFID should lead to LESS staffing, but then so should barcode self-check.
More broadly, I’m absolutely sure that RFID or one of its successors is the future. But to get there, we have some work to do in the area of standards… overdue homework as it were.
You get up way too early, Karen.
You are assuming that all patrons who renew do so on the computer. Many patrons renew over the phone. The barcode is essential for that.
You are still going to assign a book a self-identifying number whether it is on the tag or elsewhere. Some libraries may choose to go without a visible barcode. I think the positives of leaving a barcode far outstrip the negatives. A barcode is a very tiny cost. Even the costs of having staff put them on are very tiny compared to the benefits that we get from still having a visible barcode (and yes, we did a cost study of that).
Barcode self-checking does decrease staffing costs. You only have to look at San Jose to see that in an environment where you are willing to force the user to self-check you can achieve great savings. However, I still see very long lines as SJ’s main library as patrons wait to pick up AV materials.
You have to solve that problem or you have to be willing to put your materials out on the floor with no security cases and damn the theft rate. Most libraries aren’t willing to do that.
I absolutely agree that standards are essential. I also think that until standards become a reality the most sensible time for most libraries to adopt RFID is when they are building an enlarged facility or expanding an existing facility and don’t have the money to increase staff-an increasingly common occurence. Then the benefits, in many cases, will outweigh the costs and risks of adopting a new technology before standards are implemented.
As always, your mileage may vary.
Jackie
That’s interesting on the telephone renewal. So if I have out a copy of the Da Vinci Code and I call in to renew, there’s no way for the clerk to know what book I’m referring to? How often do people have out multiple copies of the same item? In fact, thinking back over telephone renewals–which I have done a fair amount of in my own day, wearing my various hats–I don’t remember *ever* discussing a barcode with a patron. Maybe I did, but I surrrrre don’t recall it. It’s the sort of thing I’d avoid bringing up, like trying to actually explain what RSS is.
(Speaking of Palo Alto, from a usability point of view I really like how their main page prominently says “login/renew” so patrons know that’s an option; reminds me of the usability analysis we did at MPOW where we learned to our great surprise that we have to clearly and repeatedly say the newsletter is FREE.)
In any event, it is still hard to see where RFID has a compellingly strong ROI over moving to barcode self-check.
As for San Jose forcing anyone to self-check, well… that was interesting wording; Palo Alto has self-check and people don’t seem forced at all. It feels streamlined, and I feel reassured by the library staff who are always no less than five feet away, even if they are busy doing other stuff. Their smiles are a big help. (Unlike Albertson’s, where the self-check stations are far, far from the single-staffed help station, whose staffer is often off on some task.)
But back to standards. You say, “I also think that until standards become a reality the most sensible time for most libraries to adopt RFID is when they are building an enlarged facility or expanding an existing facility and don’t have the money to increase staff-an increasingly common occurence.” This is also the most sensible time for those library administrators to press the standards issue, for exactly the reasons you state: the library has a need, and has money to spend to meet it. All eyes are on it, and they can rally assistance far beyond their walls. They become a case study in the need for standards.
Furthermore, the bargaining position is excellent. It’s not like saying you aren’t going to have a catalog, sucky or otherwise. A library expanding its facility doesn’t absolutely need to go to RFID as the sole solution for coping with the stressors of too much facility for the staff; it’s in a great position to lean on vendors and rally help and attention from the library technology community. That library has the most powerful leverage of all: discretionary funding.
I also think RFID standards would have been a reality years ago if technology leaders (and I am one of them, in a way) had targeted their focus better. That’s not flagellation, just an observation. We librarians have to learn that the first word out of our mouths needs to be “standards.” Everything else, as the religious scholars say, is commentary.
We seem to be talking at cross purposes. My references to renewing items is to the automated system for telephone renewal in which a patron calls and punches in the barcode number. They are not talking to a staff person. With such a system (such as III libraries are now using) it is essential that a patron be able to access the barcode.
I believe that SJ itself would tell you that they have insisted, especially in their branches, that their patrons move to self-check. And, they’ve devoted staff to making sure that experience goes well. There is nothing perjorative there just what they reported to us when we went there to observe their system.
I understand that you don’t feel that RFID has a strong ROI. I continue to believe and know from my own experience and from the studies that Berkeley did on cost that certainly in some situations it does. And, that return is, in those situations, higher than the return on barcode systems. As I said before, that will certainly depend on the situation you find yourself in and your staffing costs.
Jackie
Ok on the telephone renewal. (Though that raised ROI issues itself, and some questions about how renewals are offered; if you were predicating a system on patron-initiated renewals, and balancing the ROI against dual-tagging, one low-cost phone response could be to automatically renew ALL items or guide the user to the Web. In fact, with telephone circ renewal in person, we were trained to offer the renew-all option anyway.)
In general, I see the potential value of having a visible item number on the item itself. Again, it’s an ROI issue more complex than saying yes/no, and may have answers other than dual-tagging.
But let me return to my main points. I think that the privacy issue of RFID was overblown and ultimately distracted two issues with wider implications: standards and ROI. In fact, standards affect both ROI (Return on Investment) and TCO (Total Cost of Ownership).
I have worked in areas where consortial sharing was much more signficant to library services than it is here in California, which with the exception of the not-so-well disseminated Link Plus is somewhat feudalistic (in fact, that pretty well describes the Pacific Northwest). But even here, the idea that libraries will have inherent incompatibility (or dual-tag, which is to say internally use one system and for resource-sharing use another), rubs me wrong.
On standards, the Danish national library has it right. Get the standards settled so that systems are interoperable. That’s not just about resource sharing. That’s partly about not locking yourself so deeply into a vendor that escape is expensive. A big part of technology planning involves thinking around blind curves, but some things we know, and one is that in the long run, particularly with something as massive as inventory management for a physical collection, it’s good to keep your options open. Vendors can fold, vendors can be disappointing, etc. Look at ILS’s. We get locked into an ILS and then the company fizzles–I’m thinking DRA, fill in your own blank–and for years we toggle between praying for the arrival of Taos and considering migration.
Finally, as an option, there is a place between moving to RFID and doing circ the way we always done it. Providing self-check for barcoded collections is an interesting interim technology that deserves to be costed out for both short and long-term benefits whenever RFID is considered–and balanced against the library’s many other technology needs and wants.
I have been trying to determine if my law firm is one of the first special libraries to plan an RFID implementation. We have approval and budgeting for this project which is part of the implementation of the ILS circulation system, which we purchased this year.
A prospective vendor is currently working with our ILS vendor to refine the bridging software.
Our primary motivation is three years of statistics on missing materials. The cost of our installation will be less than one year’s loss.
Here are some differences I perceive from the situations that I have seen described in many libraries:
With our attorney patrons we do not have privacy issues.
The facility is open to them 24/7 while my circulation staff is not available.
Two to three of my staff have spent a full day one weekend of every month searching attorney’s offices for missing items – while paid for substantial overtime. Scanning the offices is a lot faster than peering under stacks of books.
The new tags we will be using will have printed numbers which will be visible like barcodes. The readers will read both barcades and RFID tags.
Cost of the new Gen2 tags is very low and we can recyle them by using adhesive pockets – something most public libraries would not choose since the tags could be removed.
However it does mean being a technology pioneer. I am certainly open to advice and commiseration.
How many items? What’s your ILS?
Since you have a small, enclosed system, it could be fine. I’d still be concerned about buying anything that vendor-specific, but I see the point for what you’re doing.
My experience in working with lawyers, when I managed a law library (part of a gov library), is that if they aren’t too tech-savvy, but if they figure out taking out a tag will mean they can hang on to a book they think they need, they’ll do it. (No offense to any lawyers reading this blog…)
I expect to tag about 12,000 to 15,000 items, or rather a summer intern will tag them. We are weeding lots of reporters and many large sets that are easily accessible electronically.
Our ILS vendor is EOS, the web enterprise version. I didn’t choose it, but it turns out to be a pleasure not to rely on our already overworked in-house IS department for service and support. All upgrades etc. are performed on our remote hosted database usually while we are not using the system.
I agree with you about the tags and our patrons, but for the most part there is peer pressure to have the firm-owned materials available to all who need them.
In phase two we plan to add readers to all floors, about 15, so the items can be tracked around the firm.
It should be interesting once we get going.
I’m heartened by this discussion which I just stumbled across. The case for standards is a good one and that’s why I’m on a RFID data structure standards committee. The real question is why we librarians are paying big bucks for non-interoperable systems. In 2004 we opted for an off-grid system, i.e. a decoupled system composed of off-the-shelf self-check kiosks, simple add-on staff station components and an ILS embedded RFID functionality. This shaved 10’s of thousands of dollars off the cost of a single third party system and, its introduction, led to considerable industry changes. We need more of this. While our system is not fully interoperable, it is a major step forward and does provide the platform for future interoperability. My worry is that the data structure standards will not be enough to ensure systems are truly interoperable because of the proprietary nature of the security functionality and the need for content encryption. Interoperability will only come if libraries require it and refuse to pay the high cost of current proprietary systems. The same issue is happening in the manufacturing/retail worlds where proprietary RFID systems are not interoperable. Thanks for the good discussion. I learned some new things.