EnvisionWare Renaissance End to End RFID Using Open Standards
BACKGROUND
There is much talk in our industry about standards such as Z39.50, NCIP, and a host of TLA’s (three letter acronyms) that will make the average person’s head hurt. So how do we define interoperability and open standards in a second generation industry that speaks of ISO 15693 and 18000-3? What is the definition of a ‘standard’ tag and why isn’t everyone interoperable if they adopt these same standards?
ANALOGY
Consider a software application. The application lies at the topmost layer of architecture. If you’re talking about an ILS, then you may next look at the standard database format. Perhaps it’s Oracle , MS SQL, or MySQL. If you know that the database is MySQL, you may not know the next layer which is the operating system. This is because MySQL runs on Windows, Solaris, Linux, and other Operating Systems. Thus knowing that the application uses MySQL does not define the OS. However, if the system is built on MS SQL, you can infer that the database runs on Windows. Knowing the hardware spec and OS spec isn’t enough information to ensure that you have open access to your data. What’s important is the database and the DATABASE STRUCTURE. Simply knowing the name of the database engine doesn’t define the names of the fields or the tables containing those fields nor does it define the table relationships. As such, if you want to SHARE your database information with another application then you need to know the database schema or structure so that you can map your other application to the appropriate tables and fields. Just knowing that it’s a certain database isn’t enough if you don’t know the schema.
HOW IT RELATES TO LIBRARY RFID
This is much the same in the RFID arena. Credible vendors speak of ISO tag standards and profess that their system supports a standard. That’s essentially like saying that each provides a common database engine but it doesn’t define or expose the schema or mapping of the fields. In fact, how the data is organized on those “standard” tags has been a mystery.
THE RENAISSANCE OPEN STANDARD
EnvisionWare has spent more than three years investigating the RFID market and looking for ways that a solution from the world’s largest end to end self-service library provider could offer something unique and beneficial to the library community. What we found is a very large community of RFID users that were collaborating and using an open standard — in Europe.
THE DANISH DATA MODEL AND THE EMERGING ISO STANDARDS
The Danish Data Model is an open encoding 'standard'. Anyone can use the Model to create products that interoperate with other vendors that support the idea of a true open standard solution. The Danish Data Model is used in several countries around the world. EnvisionWare strongly believes that the Danish Data Model is currently the best open "standard" solution for interoperability. (Quite frankly, it is presently more of a unified standard than the proposed ISO standard at the moment.) By adopting the Danish Data Model we eliminate all of the chatter about various ISO tags because the DDM also defines the layers below it, much like an application written to use MS SQL defines that a Microsoft Windows operating system will be supporting the system. The fact that EnvisionWare is delivering a solution using the DDM has been largely unique outside North America where proprietary solutions have dominated an early adopter market. The same is true in Australia where EnvisionWare Renaissance libraries are using the Danish Model. We believe it sets us apart as the company that has driven the industry toward open publication of encoding models. The DDM is not a global ratified ISO standard yet but it represents the most widely adopted open model for libraries today. This means that we are delivering a system that complies with an open standard and that our customers will continue to enjoy the use of a system that will operate with other vendors when they abandon proprietary encoding and introduce products using the DDM or successor standards.
A GLOBAL STANDARD?
The move toward standards doesn't end with the Danish Data Model. The NISO working group in the US has issued a recommendation which is under review by the ISO 28560 group. The original goal of all libraries was to have a uniform, global standard for RFID tag encoding. Unfortunately, the opinions of various national groups differ, and the result is a plan to create two ISO standards.
EnvisionWare is committed to the belief that customers have the right of choice. Renaissance tags are ISO Standards-compliant at the air interface and as a phyiscal tag device. Thus we have a universal tag standard that is widely adopted around the globe. The ISO 15693 / 18000-3 tags can work with a variety of data models.
Renaissance hardware is designed for adaptation. As new technology emerges, our RFID hardware can adapt to those changes. We support rewrite, which means that we can read two models at the same time and automatically rewrite a tag from one model to another. Should our libraries choose to adopt one of the two ISO encoding models under review, we are committed to support those models plus the Danish Model using the existing hardware we are providing to our customers today. This leaves CHOICE in the hands of our customers.
WHAT ABOUT PROPRIETARY SYSTEMS?
Many ask, what about my current proprietary system? Once a vendor like EnvisionWare introduces an open and public model as widely marketed as we have done, it's only a matter of time before others must do likewise. Many vendor-specific implementations have been published in recent months, making it possible for EnvisionWare to offer interoperabilty and migration paths to libraries that are using ISO standard tags. 3M, ITG, and TechLogic have agreed to make their vendor-specific encoding models available. Whether you choose to remain with your current proprietary model on Renaissance systems, support your current and a new model, or dynamically rewrite your tags, we have the technology to support your decision. We support multi-vendor interoperability and model rewrite so that seamless migrations to new models can be accommodated for our customers.
It is important to note that ISO data model standards may take some time to formalize. Our Company is a reviewer of the recommendations and the emerging standards. As with the open Danish Model, when the ISO 28560-1/2/3 standards are ratified, we will be one of the first to support the models. We will give our customers the choice they deserve, consistent with the founding principle of our Company.
Nearly 20% of the libraries we serve represent a migration from another vendor. We have in-house expertise to develop and manage painless migrations for our customers.
The open standard used by EnvisionWare and introduced widely to the U.S. in the first end to end, turnkey solution for full self service is the Danish Data Model. To view or download the Danish Data Model (DDM), please click here.
Standards supported:
Tag
ISO 15693 / 18000-3
Data Model
ISO 3166-1 Country of Owner Library
ISO 15511 International Standard Identifier for Libraries and Related Organizations
UTF-8 Unicode encoding
Danish Data Model
NISO has submitted a recommendation to ISO in support of an ISO Data Model. EnvisionWare will support multiple open standards, re-encoding, and dual-standard detection.
EnvisionWare may provide interoperability/migration for some of the current US propretary models in order to provide upgrade paths to libraries using proprietary, legacy systems.
Communications
SIP2
III Item Status API
This section is used to describe the typical application of the Danish Data Model as used by EnvisionWare Renaissance customers in North America and Asia Pacific. There is great value in storing information beyond a simple item ID. The model defines a structured method of describing materials which facilitates inventory, ILL, and enhanced sorting capabilities. Detection of complete multi-item media sets is inherent in this model. Encoding is international, using UTF-8
The information is programmed dynamically and statically. Using the EnvisionWare Conversion Software, the library would enter static information such as country, library name, AFI values, etc. The item ID is dynamic, read from the barcode scanner and programmed automatically. Programming books simply requires application of the tag and scanning of the item barcode. An additional step that requires a keystroke or mouse click is used for media. The library staff would select the media type and encode the number of items in a set as well as the ordinal number (1 of 3). This permits other applications to verify that all items are present in a set. It further permits ease of data collection such as at gates. Since gates are not connected to the ILS, having media type in the gate permits the library to record statistics about the types of items and quantities of each type that cause alarms.
The power of the tag to describe the item to sorters, gates, self-check, inventory, and other applications adds to the value of RFID.
• AFI (security)
• Check method
• Standard version (version of the Danish Data Model being used)
• Type of usage (1:circulation, 2:not for circulation [cannot be checked out], 7:discarded, 8:patron card, etc.)
• Primary item ID (item 'barcode' up to 16 characters. You may use the alternate item ID field for longer item IDs)
• Number of parts in item (3 items make a set of DVDs)
• Ordinal part number (item 1 of 3)
• OPTIONAL: Alternative item ID
• Country of Owner library (the library's country as defined by ISO 3166-1 AU, CA, MX, NZ, US, etc )
• Owner library (name of the library that owns the material as defined by ISIL, ISO 15511. Supports standardized and non-standardized) eg. City of Onkaparinga Libraries=SNRL
• OPTIONAL: Extended owner library (used if longer than defined character length)
• Media format (1:book, 2:CD/DVD, 3:magnetic tape, 4:other, 5:careful handling, 6:small item [used with sorting systems]
• Supplier ID
• Item identification
• Order number
• Invoice number
• The Danish Data Model supports the ability to add customized information.
During the pre-planning meeting with your project manager, he or she will discuss the value of each element and help you to determine the items you wish to record in the tag. We configure the system with your static items, and you make the media decisions dynamically. The system automatically handles item IDs by simply reading the item barcode.
At present, the ISO 28560 standard is actually developed in three (3) parts. One part defines the common elements of the standard. The second two parts discuss two different approaches to the standard, which results in two different encoding models.
ISO/CD 28560-1 Information and documentation -- Data model for use of radio frequency identifier (RFID) in librairies -- Part 1: General requirements and data elements
ISO/CD 28560-2 Information and documentation -- Data model for use of radio frequency identifier (RFID) in librairies -- Part 2: Encoding based on ISO/IEC 15962
ISO/CD 28560-3 Information and documentation -- Data model for use of radio frequency identifier (RFID) in librairies -- Part 3: Fixed length encoding
One of the biggest concerns we have is the absence of any requirement or structure for information other than item ID. The proposed standard suggests that the item ID will be written in two different models but that the item ID is the only required element.
Contrasting with the Danish Data Model, for example, the Model defines a specific location for owner library, country, media type, type of usage (circulation, non-circulating) and multi-item set information. When considering off-line circulation or automated materials handling (self-service returns), this additional data is invaluable. A model that provides information about media allows you to set rules at your self-service station and when off-line you can still track whether all items were in the multi-item set. It could permit continued sorting even when the ILS is out of service. The added information could be used to determine whether to sort an item (why sort if it's not your material?). It could refuse to check out an item that is reference material during an off-line transaction. Since SIP can affect the latency of any automated return solution, using tag data increases efficiency and speed.
The proposed standards permit extra data, but the structure of the data is not standard and the presence or absence of any data other than an item ID is up to the individual library. Further, that fact that the standard is actually two standards can lead to frustration or even confusion in the market. "Are you compliant with the ISO 28560 encoding standard?" A vendor could say yes, yet you would be unable to read those tags.
We applaud the concept and desire to create a universal, ratified standard. We will absolutely support both of the standards in the Renaissance system. But we want customers to know what you may be sacrificing when making a decision to use ONE of the TWO ISO standards when they are issued.