Agree, RFID isn’t ’sexy’as a discussion topic because its pretty simple, and if you do it right it ‘just works’ (though the ’soft; side of change management for the organisation is not quite as simple, but there’s a whole other debate..)

The potential of the tech is enormous if we can get our heads around what we want it do in the future (other than self issue), but you can’t map out a future if we’re all starting from different places – hence the basic ISO standard is *vital*.

I’m a sceptic in terms of what data goes on the tag (as long as its in a standard we can *all* read), and can’t see benefit in more than the unique id and other standard stuff – let your LMS do the hard work instead – but there’s lots of people far smarter than I out there who can & will see possibilities so we need to give them a level playing field to work up from.

The direction we take now will have pretty long term consequences, so engage in the debate – blog it, twit it, talk to people, bore your colleagues to death over lunch about it.. Lets make sure we get a consensus and get this right!

To my mind, the answer to your first question is a no-brainer : Yes, we do want tags to be usable across RFID systems; to be able to change RFID suppliers in future without re-tagging; to be able to move stock across institutions, handle institution mergers and establish consortia, secure in the knowledge that our tags are as universally readable as a barcode. To store item data on a tag which cannot be read by any system is a retrograde step. It’s actually rather shocking that RFID has come as far as it has – in modern libraries of all places! – without a standard data format.

The answer to the second question is also yes. We should always be exploring the potential of new technology to improve services, and as far as possible we should be implementing new technology in ways which leave the door open to potential new uses. This isn’t always easy when you don’t yet know what the potential new uses are, but some things are obvious door-closers. Leaving off the potential for storing more data on tags is a door-closer (although at the moment, I’m not convinced of the need for more than “licence-plate” data). Proprietory data formats (or almost anything proprietory!) can be door-closers. In the long term, what is the advantage of not having a common standard?

I think you are right that as a minimum ‘any’ agreement for standardisation is better than none.

Having spent many years in the bowels of LMS development I can attest to the fact that supporting a multiplicity of barcode standards, let alone RFID ones, does not scale and the industry has lumbered itself with a load of bagage that it has to continue to support (although some – not me – might see that as useful customer/vendor lock-in). We (libraries/LMS vendors/RFID syppliers) as you point out seem to be in the process of reinventing that particular wobbly wheel.

What I am sceptical of, as you know, are the percieved wonderous future benifits of throwing bibliographic and other data on to the chip.

If the ‘we’ I refer to above can at least agree a minimal very useful set of basic data to put on the chip (such as unique local ID, global ID of the type of thing (ISBN, etc.?), and which library it is from) and how it should be encoded. We would achieve two things – firstly avoiding the basic anarchey you predict, plus it will provide a stable platform for those who predict a wonderous future to prove me wrong from.

The only concern is that there are are approching the point where there are suffcient, proprietory, pragmatic implementations to cause concerns especially in the current economic climate, from suppliers about the costs of change to an ‘agreed’ way forward.

Following on to this, and your comment on Richard’s blog, having worked ‘both sides of the fence’, Public Libraries are in more danger, and its partly down to deskilling (though there are some very good people out there still), and partially its down to the pressure those libraries are under.. They are having to make snap decisions under rigid procurement guidelines & timelines to improve efficiency, and the long term issues tend to get put aside. There is a much more measured approach and greater awareness of the technical debate on the Academic side, well, generally there is.

It’s an education issue (for librarians, LMS vendors and RFID vendors), so I think all we can do is keep banging the drum, keep posing and answering the questions, and above all, get the message and information out there *and* properly signposted… I guess this comes back to seeing some advocacy at a National level.. has there been any feedback from CILIP to the recent Gazeete article?

My comment that I’d “try to find time” was not meant to be dismissive – rather an honest comment.

Clearly there is an issue about understanding of RFID and the relevant standards – I’m all in favour of this, and I hope no disagreements about what exactly we are aiming distract from this basic issue. I have to admit that I’m not completely clear what actions you would like to see CILIP or others take?

Anyway, on to the discussion

I’m not currently involved in an RFID implementation, although I have been in the past, and if I was, my recommendation would be not to put bibliographic data on the chip. When I was involved in an RFID implementation all 4 vendors we saw recommended minimum data on the tag – and it was in this context I commented on Richard’s blog post – obviously I accept from Mick’s comments that this is not the situation now.

Although I’m in no way against the idea of getting standards in place – of course it makes sense to agree standards where necessary, I do feel that if there is a message to implementing libraries at the moment it ought to be that it is not good practice to store data beyond the identifier on the chip. I understand the RFID chips can store more than this, but I’m yet to be convinced that there is real benefit to doing this – which I think is the point Richard made in his post when he said “RFID 2.0 … looks like a great solution searching for a problem to solve.” – I think (he can correct me if he wants) he was referring to the idea of storing the extra information rather than the idea of a standard which governed this.

Similarly it would be possible to issue RFID user cards that contain more information than just an identifier – but I really think this is a bad idea, and would prefer to discourage it than define a standard which governs (for example) the storage of name and date of birth – which gets us into all kinds of privacy issues I’d suggest are much better avoided.

You pose two questions:
Do we want to explore the potential for developing better services by using the full potential of the technology?

Generally I’d say yes to this. RFID is a sensitive area for ‘full potential’ – I’d be a bit careful what we are asking for. However, I do think (as I mentioned in my comments in Richard’s post) that there is considerable potential for the technology to be put to use in the technical services side of library activity.

The first question you pose is “Do we want to be able to use our expensive tags with any RFID system?”

I’d say, yes of course, but I need some more convincing that this would deliver substantial benefit. Again, I’m not against standardisation at all, just if this is going to be a priority then I need to understand why this is of real benefit – perhaps some scenarios where interoperability is key and there is clear cost/benefit.

Mick, I am glad that you quoted Alan Butters. As you know, but others might not, Alan and I are co-project editors of ISO 28560-2. We have never argued that all the data elements in ISO 28560-2 need to be encoded. The power of the encoding scheme is that:
 It allows any of the optional data elements to be selected
 It allows those data elements that are selected to be arranged in the most appropriate sequence
 It allows different data elements for different loan items
 It supports variable length data
 It supports selective locking within the rules of the tag architecture
 It uses an ISO standard encoding rule in ISO/IEC 15962, used in other sectors and which will increasing be used providing a wider choice of equipment and interoperability

None of the features are supported in the fixed memory structures as in the Danish model, nor in the proprietary encoding schemes of present RFID vendors simply using RFID as an electronic bar code.

I am very sympathetic to the arguments that the RFID tag should not replicate the LMS. I am working on other RFID projects associated with safety critical parts – which result in headline news when they fail – and the same argument exists in that arena in what should be encoded on the tag. The computer record is the authoritative record, but there are situations where having data on the tag with the item can. Read what Alan Butters at the top of the page.

Remember that the typical RFID used in libraries has more memory capacity than just for the bar code. Remember that none of the countries that used fixed models with more than the bar code want to go back to the single data element.

Try some encoding. We were asked by the relevant ISO committee to develop a demonstrator to show the encoding using the modelling tool that our software operation developed.
http://www.convergent-software.co.uk/libraries.htm