It has been a far from quiet week or two in the world of library RFID lately…
What exactly IS going on with attempts to improve integration of ILS/LMS systems with third party solutions? To try and unravel the mystery we may need to first remind ourselves of the dramatis personae – and the options they represent.
First on stage is of course 3M. Since long before RFID was installed in the first library 3M have supported the integration of self-service devices with ILS/LMS systems through SIP (Standard Interchange Protocol). Initially the company sought to licence it to systems vendors (I was one of them at the time) but were eventually persuaded that the benefits that might accrue to libraries by allowing other vendors to use it outweighed any narrow commercial gains that the Minnesota-based giant might make.
So SIP was released into the wild and over the intervening decades has been pressed into service to facilitate a wide range of services that weren’t even conceived of when it was first designed. Unsurprisingly therefore it has had to develop and grow along the way in an effort to accommodate the demands put upon it by a changing library landscape.
By 2006 SIP was in its second major revision – version 2.0 – and had grown considerably in both its complexity and scope. The decision to use “extensions” to enable application developers to store and exchange values outside of those required to support circulation proved to be both a blessing and a curse. The “blessing” was its ability to accommodate new functionality, the “curse” the absence of any standards-based structure or control over the use of the extensions.
As RFID began to find its feet – and suppliers realised the opportunity to develop the technology into areas other than circulation – the shortcomings of SIP became more and more apparent. In London, as BIC/CILIP’s joint committee strove to create a national data model for RFID (to run under the subsequently published ISO 28560-2 standard) the most common complaint heard around the table – from RFID and LMS providers alike – was that SIP had become too limited for RFID (and other application) needs.
So once ISO 28560 was published attention shifted to addressing the problems posed by SIP.
Enter the next character in our play – Book Industry Communication (BIC).
BIC chaired the joint committee that produced the UK data model so it was a natural progression for it to turn its attention to the vexed question of SIP. The make-up of the committee – LMS and RFID solution providers, servicing companies and – most importantly – librarians made it the ideal forum to debate the issues, despite the realisation that such discussion rather exceeded the brief of matters RFID. After one exploratory meeting it was agreed that it should explore the possibility of working with 3M to improve SIP and extend its scope beyond circulation.
3M’s response to these overtures was instead to announce their intent to develop SIP themselves and SIP 3.0 was announced – albeit rather hurriedly – on the US RFID list in response to an email I had previously sent – to the same list – asking for clarification of their intentions. I, and several other non-3M personnel, were invited to join in the deliberations. I attended two conference calls but found only 3M personnel present – and with an agenda was firmly fixed on circulation rather than the wider issues being explored in London. I elected not to sit up late at night to attend any more and regretfully withdrew.
BIC now faced a dilemma. Should it continue to work on the integration problem or let 3M finish SIP 3.0 and hope it would eventually come to address our wider concerns? In the end the answer to that question was prompted more by Innovative Interfaces than by the RFID market. Along with other US ILS companies III had opted to largely ignore SIP and instead develop a set of web services to better integrate third party solutions – including RFID. (Still a popular approach in the US where Ex Libris and Bibliotheca recently announced their intention to “work closely”.)
It was becoming evident that this approach was becoming increasingly popular with other ILS/LMS providers and that, pretty soon, all integration of this kind would be provided by bilateral, entirely proprietary solutions that would “lock in” libraries even more securely than the proprietary data models we had only just managed to replace.
The decision was made to go ahead with our own deliberations and to invite 3M – and NISO – to join us. BIC was very fortunate in having its own communications/integration expert in Francis Cave who had previously tackled similar issues for the UK book trade and it was he who first mooted the BIC Library Communications Framework (BLCF).
The concept was very simple. With so many methodologies being developed to drive integration between ILS/LMS and 3rd party solution providers it would obviously be short-sighted of BIC to develop a protocol that dictated the means by which data should be exchanged. That was the difficulty facing SIP – as a protocol it prescribed the means by which data was exchanged as well as the actual values to be used. BLCF did something different – it defined the data elements and values but left the means of communicating them to the providers. It’s a pragmatic approach that was felt to offer a better chance of it being adopted since it allowed providers to determine the means that best suited their development environment, whilst allowing libraries to document the values used to provide functionality in a way that could be passed onto another agency if needed.
The third player in our drama – NISO – were invited to join in as they had for some time been working on a version of NCIP (NISO’s Circulation Interface Protocol) for use with RFID self-service. BIC had hoped that the flexibility of its approach would be attractive to NISO and enable it to look at the greater potential of RFID to manage more than self-service circulation. In the event neither NISO nor 3M responded to BIC’s invitation.
…and so to the second Act in this little play – The Publications.
BLCF was published in 2011 and announced at the London Library RFID conference in November. By February 2012 a project was underway to develop better integration of stock management routines using the BLCF framework and a web service based approach to data communication. Capita provided the LMS and the servers, Bibliotheca tabled the workplan and most of those present committed to work on the project with a view to delivering a product by November 2012.
Shortly afterward 3M announced both the release of SIP 3.0 and their intention of persuading ILS providers to implement the new protocol as quickly as possible.
It is at this point that the story becomes somewhat difficult for me to tell as some key players – including senior personnel at US and UK RFID providers, members of NISO and others – have offered me various insights as to how events now unfolded.
So what follows is pure speculation on my part – but perhaps quite well informed speculation.
One can only imagine the delight of ILS/LMS providers in learning that the industry has found not just one, but two ways in which it could prevent them from selling extremely lucrative proprietary solutions to meet a rapidly growing number of integration needs. Joy must have been unconfined as 3M announced that there was now a new protocol available to manage self-service circulation.
With its major competitors seeking to make inroads in 3M’s domination of the US market by extending the reach of RFID into other areas of acquisition, accession and management it is also just possible that 3M realised – rather late in the day – that they had somewhat limited their RFID options by focusing on SIP 3.0 as being simply a better self-service protocol.
At the same time NISO were still trying to produce a first viable implementation of NCIP for self-service. 3M may have seen an opportunity to get out from under the limitations of SIP and do NISO a favour at the same time. So earlier this year 3M may have begun the process of divesting itself of its SIP responsibilities/limitations and begun courting NISO to take it on.
NISO will not necessarily have recognised 3M’s overtures as the precursor to the major philanthropic gesture to the library world that 3M had intended. They may indeed have viewed it as being something of a poison chalice. Where would the acceptance of SIP leave NCIP’s self-service offer? Would US libraries be forced to adopt/subscribe to two protocols? Most important of all – who would commit to support and development of the new protocol?
It may be worth mentioning at this point that it is my understanding that ILS outnumber RFID providers 5 to 1 on the newly established NISO SIP committee. (Bear in mind my comments about the likely enthusiasm of ILS providers for SIP here.)
Nevertheless NISO agreed to take on SIP 3.0 and 3M announced the decision – and the good news this represented for the library world in a press release that I received on June 7th. The release also contained an invitation for anyone interested in the process to join in by emailing NISO (I did – but received no reply) or following a link to a workroom on the NISO site (which returned the message “Resource not found”).
In 3M’s press release Todd Carpenter, NISO’s Director was quoted as saying,
“The maturity of the SIP protocol and its implementation track record should allow it to move quickly through the NISO standardization process,” stated Todd Carpenter, NISO Managing Director. “We anticipate that version 3.0, as it currently stands or with very minor revisions, will be adopted as a standard following a brief period of review within a NISO Working Group.”
By last week Todd was defining this short period a little more precisely when in an exchange of messages on Twitter he told me,
TAC_NISO: Mike Dicus (ExLibris) and Rob Walsh are now talking about NCIP and SIP at#NISO update #ala12 10:42pm, Jun 24 from TweetDeck
mickfortune: @TAC_NISO Do they have a timetable for concluding deliberations ? #NISO#ala12 10:44pm, Jun 24 from Tweetbot for iOS
TAC_NISO: @mickfortune Obviously, time lines are guides, but it helps that SIP is mature and widely adopted as is #NISO #ala12 11:05pm, Jun 24 from TweetDeck
mickfortune: @TAC_NISO thanks for the update. Had heard that SIP and NCIP might be merged.11:23pm, Jun 24 from Tweetbot for iOS
TAC_NISO: @mickfortune Probably not likely they will be merged (any time soon at least) Could be clarification of when 1 vs other, addressing overlap 11:25pm, Jun 24 from TweetDeck
TAC_NISO: @mickfortune Rationalization is probably a better term than merger 11:27pm, Jun 24 from TweetDeck
(TAC_NISO is Todd’s Twitter handle)
So now it’s Q2 2013 – and not a merger but a rationalisation, and only after a gap analysis has been carried out.
I don’t have any figures to back up Todd’s assertion that SIP is widely adopted (I presume he means earlier versions) or what he refers to here as “NISO” (by which I presume he means NCIP – but possibly not the self-service version).
These are key points since if both NCIP’s self-service protocol and SIP 3.0 ARE widely adopted it will be very difficult to explain to the market why both are necessary – and possibly why NISO took on SIP at all. If they’re not already widely adopted NISO and 3M have just handed the ILS providers the best reason they could wish for to press ahead with proprietary solutions. By the time something emerges from NISO it will already be too late to be putting genies back into bottles.
So where does all this leave librarians?
Vulnerable is probably the word that sums it up best.
LMS/ILS providers may try to persuade them that SIP is dead (with some justification) – or at best delayed, and that the best way forward is to abandon the ‘old’ ideas anyway and embrace the wonderful world of APIs and web services – an area secure from competition and able to command whatever price can be afforded cash-strapped librarians.
NISO – as the single most influential standards agency in the world (because so much library systems development is still US oriented) – appears uncertain about SIP, NCIP and the whole RFID landscape so not much hope of salvation there.
Where to turn? Could BLCF help here?
Well yes, it could – but only if librarians demand it be used. Since it allows application developers to use whatever means they wish to integrate their applications all that is necessary is for them to use the values set out in the document – rather than inventing new ones. BIC has undertaken to make the framework freely available to anyone who wishes to use it, to maintain it in perpetuity and to incorporate new values as needed – subject to ratification/management by the BLCF steering group.
Even when SIP 3.0 and/or NCIP emerge from NISO’s deliberations they will still only be regulating circulation control. New products that utilise the increased capabilities of RFID tags – post ISO 28560 are already in development. The only hope of preventing a plethora of proprietary solutions from overwhelming the library market is for librarians to insist that suppliers – RFID, Smartphone or LMS/ILS develop their integrated solutions within a common data framework – and the only one we have is BLCF.
It’s your choice.