Grand Rapids, RFID, and Open Source--An open letter of support

Posted on January 26th, 2010 by Jed Moffitt and tagged , , , .

Anyone familiar with my previous writing and thinking about RFID may find this post ironic, if not contradictory.  I'm hoping rather that it simply reflects keeping an open mind about the evolution of technology, and the value and energy that can be created by public libraries working together to create great new services through newfound control of their information systems.

In summary, this post is an open-letter of support for a grant proposal led by the Grand Rapids Public Library.  The proposal is for the development of a Radio Frequency Identification (RFID) sub-system and standard for interoperability with the Evergreen system.

The potential benefits for public libraries include better patron service and staff efficiency for material handling, self-checkout, and possibly a few security options to try to discourage disappearance of popular media materials.

Cutting right to the chase, the possible benefit that I'm really jazzed about for KCLS is the ability of this concept to make it so that our backroom staff don't have to open every single DVD or CD that returns to the library to verify that:

a) the disk is actually in the container and

b) that multiple disk sets actually have all the disks included

KCLS has alot of experience with developing Automated Material Handling and Self-Checkout systems-- based on optical scanning-- that are really effective and economical.  Take a look at this YouTube video of our central sorting operation:

See video

But when we were talking with the Grand Rapids/Calvin College crew via conference call yesterday, they captured our attention and imagination with visions of what they described as a "killer app".  That application being the inexpensive, fast, and reliable reading and recognition of individual and combination disks while still in their casing, via RFID.

This kind of functionality has been promised by RFID vendors for some time now, but there have heretofore always been issues with metal, and collision, and general inaccuracy of this kind of read, combined with extremely high prices for disk media RFID tags (often over a dollar per tag).

Grand Rapids/Calvin are looking to explore possibilities here that would make the tags roughly a tenth of the cost, and make the reads and recognition of the disks accurate and usable.  If so, that is indeed a "killer app" and KCLS is interested in supporting the effort.

 

Share this
0
Your rating: None

Comments

Jed, The piece that caught my

Jed,

The piece that caught my eye (from a KCLS centric perspective) was the ability to read destination locations on a tag during central sort operations.  Imagine a world where we didn't need to make a SIP connection anymore to know how to sort our material.  Since the tags are writable, it could be as fine a sort as we like:  this library has all their multimedia returns separated from their multimedia holds from their book returns and their book holds from their kids everything!  Could be very exciting!

Comment by Lori Bowen Ayre on Feb 5th, 2010 at 1:22 pm

The current practice with

The current practice with library RFID is to only write the item barcode into the tag and rely on communication back to the ILS, typically using SIP/SIP2 protocol, to get real-time details as to the item's status, collection location or check-in the item.  As Lori Ayre notes above, the RFID tag on the item could hold much more information such as the owning library name, zip code, etc.  The benefit to having more information contained in the tag is for central sorting in a regional hub where many ILL are passing through.  The sorting at this level would need to be fairly rough, using conveyors to have items land in appropriate delivery routes and bins.  In this instance the number of library ILS that would need to be queried real-time via SIP/SIP2 might be prohibitive for rapid sorting, so having more information contained in the RFID tag would assist towards central sorting in a state-wide delivery system.  All that would be needed is the ability to read the tag and know what library owns the item and determine what delivery truck/route the items needs to go towards.

To this end, I think several things must be considered:

  • Uniform tag standard, such as the ISO 15693 and 18000.
  • Tag writing practice should be standardized for libraries, preferably at the North American level, so that software writing the tag will use a particular sector for item 14 digit barcode, another sector for ownership, etc.  This is detailed in section 2.5 of the NISO document "RFID in US Libraries
  • Tags should be written and left unlocked so that future advances in RFID technology or practice can take advantage of the read/write/modify abilities of an RFID tag.  Why tag a collection, locking each tag, only to doom innovation later?
  • Adoption of RFID should be "incentivized" within library materials, either through the acquisitions process or at a state library level
  • Bulk purchase of RFID tags would significantly drive down the price per tag.  Library consortia or consortia of RFID libraries should consider ways of pooling purchases so that the number of tags in an order would be in the millions, rather than the thousands.

My experience at the Oak Park Public Library as an early adopter of RFID was that the proprietary market of RFID systems back in 2002 lead to some regrettable outcomes.  The market is much improved today, particularly with the ISO standards of library RFID.

 

Comment by Librarian Aaron on Jun 4th, 2010 at 12:36 pm