This article is about using RFID technology and Active tags in particular to help us fully automate the paper records management process.

However, the first thing any organization should do is to remove as much paper as possible from the equation and in doing so reduce its RFID costs as far as possible. This is usually done by a proven process known as ‘Document Imaging’. In regular parlance we ‘scan’ as much of our incoming paper as is practicable and legally allowed and convert it to an electronic record usually in the form of an indexed PDF.

What this means is that the first result of the document scanning process is a TIFF file, a bit image (where no text is extractable or searchable). In order to add more value to our document and improve search results we should next process our TIFF image with an OCR (Optical Character Recognition) tool to extract the text from the bit image. We then combine the bit image and the extracted text into a single document by converting both of them to a single PDF document. The PDF now contains both image and text so it is full text searchable and the text can be extracted if required.

I will cover how we fully automate the management of all electronic records in the next chapter.

Now to the core paper problem; paper is generally either sent to us or produced in-house as the result of some business process. If we are smart we produce as little paper as possible in-house and manage all of our in-house documents as electronic documents (e.g., Word, Excel, PowerPoint, etc.). As mentioned above, we also try to capture as much of the incoming paper as we can by scanning it and converting it to electronic records (indexable PDFs). It is the paper we can’t or are prevented from converting to an electronic record we now have to deal with.

Our paper records are usually managed as ‘loose’ paper (i.e., as single documents) or file folders (loose paper collections) or archive boxes (file folder and loose paper collections). We have to decide which of these classes of paper records to apply RFID tags to (i.e., loose papers, file folders and archive boxes) and what data to encode on the tags.

I think the usual decision would be to apply tags to a small number of loose documents (e.g., important documents that travel ‘loose’ throughout an organization) and then to all file folders and archive boxes. There may be an argument that says we don’t need to tag archive boxes if the contents are tagged but that decision will depend upon each organization’s specific needs and tools.

As to what data to encode that is totally dependent upon each client’s needs and the various classes of documents and containers it manages as well as its compliance needs. It is probably worthwhile mentioning here that Active RFID tags can also differentiated by the way they store data. There are three types of data-storage (and again I generalise):

1. Read-only
2. Write-once, Read-many (WORM); and
3. Read-Write

Read-only tags are the least useful because they can’t be changed or added to once they have been created. WORM tags can’t be overwritten but can have data added at a later time. Read-Write tags are the most useful because their data can be modified at any time. Each type of tag has its most appropriate uses depending upon your needs and budget.

For my fully-automatic solution I will again choose the most expensive option, the Read-Write tag because it provides us with the most features and most flexibility and again let me say, “Computers are cheap, people-time is expensive!”

In my system, our Active RFID tagged documents now communicate directly with the records management system and database. Our paper documents are now active, not passive and are able to both receive and pass on information as they are processed.

They can tell us where they are, they can tell us that a process is due or overdue and they can tell us when they are due to be archived or destroyed.  They will tell us if they are moved and they will signal an alarm is they are moved away from a predetermined area or building. If we have also RFID tagged our employees (either chipped them – probably not practicable but certainly possible, or more reasonably, RFID enabled their employee badges) then our active documents will tell us who has them and who is reading them and at what time. They will also maintain a complete history or ‘Audit Trail’ either on the RFID tag itself or within our database.

It may sound like science fiction but it is not as all the technology and tools required are current and proven and ‘affordable’ and in use in other applications like container and freight and asset tracking. Our task is to simply adapt the most appropriate elements of RFID technology to paper records management.

Our now fully-automatic new system obviously requires a records management computer system and database plus RFID reader/writers at strategic locations in our enterprise. We also need a suitably appropriate and well-designed Taxonomy or classification system and a predetermined set of rules (workflow) with which to manage our now RFID-enabled paper. The Taxonomy and Workflows should be combined into what we usually call a ‘File Plan’ such that we can encode all of the information required to manage the full life cycle of the record (the Continuum model) into an RFID tag. As each document is captured with an RFID tag its information is automatically entered into our records management system’s database and the management process immediately begins with the document and computer system in constant contact thereafter.

The Read-Write tags chosen provide us with the flexibility required by for example allowing us to reprogram the workflow or retention as and when required. This feature is essential if we are to meet the real world requirements of the records management profession. We can do this via our records management system in a controlled and audited and secure way.

This article is just a skeletal view of what is possible and how it can be done. A complete description would entail writing a wordy tome inappropriate to this medium. However, the above is all possible; it is not science fiction.

Records are evidence of any business transaction and as such include all types of documents including paper, electronic and email. To meet any records management standard or compliance standard you must be capturing and managing records of all types. If you are not, then your practises are seriously deficient and I predict that at some time in the foreseeable future you will be discussing these same deficiencies with a judge or government bureaucrat. I can also predict that you will not enjoy the discussion or the financial consequences.

When looking at how to implement our new fully-automatic paradigm we need to apply tools and technologies appropriate to each type of ‘record’. In this chapter I will discuss the best technologies and tools to use to automate the management of paper records.

I started giving presentations on the coming paperless office in 1984 but gave up some 5 years later because it patently wasn’t going to happen any time soon. I made the point then that until government outlawed laser printers and copiers that paper usage was bound to increase, not decrease and so it has been over the intervening years. So, managing paper in all its forms (e.g., loose paper, file folders, archive boxes) is still a major challenge and a major cost for all enterprises.

The most common technology currently applied to paper is barcoding. Barcodes are usually placed on file folders, archive boxes and shelf or storage locations and then human beings point barcode readers at these same containers and locations to register and track them. It works fine when used properly and barcodes can be super labour-saving devices but it is still a very labour intensive and decidedly non-automatic process.

Barcodes are proven and reliable and in wide-use (e.g., UPC barcodes on all the products you buy in a supermarket) but there are better technologies now available.

A quick review of current and proven technologies provides us with a surprisingly short list and comprises:

1. Geographic Information Systems (GIS)
2. Global Positioning System (GPS)
3. Radio Frequency Identification (RFID)
4. Wireless Local Area Network (WLAN)

Of the four I think that GPS and RFID are the most appropriate for paper. However at this time I have rejected GPS because of technical difficulties (e.g., finding a satellite from the basement) and cost. This leaves me with RFID as the sole choice from proven technologies for our fully-automated paper management paradigm.

RFID has progressed a long way since its inception in the 1970’s both in functionality and cost-effectiveness (RFID tags now cost just cents) and it is now an entirely appropriate technology to utilize to automate the management of paper.  The question is, “how best to utilize RFID technology?” A second question may well be, “why aren’t more organizations today already utilizing RFID technology to automate paper processing?”

Let me answer the second question first. Historically the rollout of RFID has suffered because of high costs and a lack of standards. There are industry standards for barcodes (e.g., Code3of9) but historically each RFID vendor came up with their own standards both for the tags and for the computer interface required for other systems to ‘talk’ to them.

Whereas there are still issues with RFID standards the situation today is light years better than it was say ten or even five years ago to the degree that  the lack of standards is no longer a reason not to use RFID technology.

Before answering the first question let’s talk a little about RFID technologies. I will generalise and say that there are three types of RFID tags for us to consider:

• Passive
• Semi-passive; and
• Active

Passive tags are the cheapest to buy (say from 7 cents up) but have the shortest maximum reading distance, usually no more than twenty feet (six metres).  Semi-passive and Active tags are more expensive because they have more circuitry and batteries but have a far better maximum reading distance of up to one-hundred feet (thirty metres). And finally, Semi-passive tags still rely on the reader for power to transmit but Active tags can transmit under their own power.

For my fully-automatic solution I have chosen Active tags despite their higher cost because as I keep telling my customers, “computers are cheap and people time is expensive!”

Active tags provide far more options for us when trying to solve our problem so in the end they are the better choice and I believe, the most cost-effective choice because of their inherent benefits.

In the next chapter in this series I will answer the first question posed above and explain how I would best utilize Active RFID technology to fully automate the management of all paper records.

In the first Blog on this topic I asked the question “is the solution fully-automatic, self-registering, self-classifying, self-managing records? Is it possible? Do we have the technology?”

Since I invented the concept of fully-automatic, self-registering, self-classifying and self-managed records I would have to say, with a degree of bias, that it certainly is the solution to what appears to be an impossible to manage problem using current tools and methods; what I earlier called “this burdensome task.”

As to whether we have the technology I think we do but the solution may involve a synthesis of technologies from other industries to achieve our aims.

I have visited with hundreds of our customers all over the world plus hundreds of prospective customers and I can confidently say that not one manages its records one-hundred percent and that most don’t even come close. In a few cases it is because they don’t have the right tools but in most cases the tools have a super-set of the functionality required, they are just not being utilized effectively.

Certainly, not one organization I have had contact with meets all of the compliance legislation it is supposed to meet. Even when they have the tools required the job is just too big, too complex and, that word again, too burdensome. The biggest problem, apart from the huge scale of the problem, is getting the co-operation and ‘buy-in’ from all staff.

Many systems rely on each and every staff member becoming an expert records and electronic document and email manager and on each staff member always behaving in an entirely consistent and reliable manner. This basic assumption is of course fatuous and ignores both the diversity of human beings and human nature. We may all be born equal but we don’t end up equal in intelligence, experience, expertise and attitude. I also don’t know anyone who exhibits one-hundred percent consistent behaviour day after day after day. Au contraire, in my observation we are all human and we all have failings and we are all sometimes better and sometimes worse. Let’s call it the human condition.

The effect of the human condition means that any system that relies on each and every staff member behaving in an entirely consistent and reliable manner is seriously flawed and will never work. Let me repeat that, “will never work!” In fact, any system that ignores human nature is a bad design and will inevitably fail.

So in my mind, if I can’t rely on human beings to be perfectly consistent in the application of a given set of rules and processes then we need to take human beings out of the equation and replace them with something that is perfectly consistent in the way it analyses, captures and classifies records. Ergo, my fully-automatic model bereft of human frailties.

Because technology isn’t quite there yet I am not proposing to use robots or artificial humans (i.e., Androids) a la Blade Runner. I am however proposing to use rules-driven systems and where possible, artificial intelligence. I am also proposing to utilize technologies currently available but not yet generally applied within what AIIM calls the content management industry.

In the next chapter I will discuss the tools and technologies I would use and explain how my method will address most of the problems of this burdensome task.

Back in 1995 I wrote a paper called ”Document Management, Records Management, Image Management Workflow Management…What? – The I.D.E.A”

IDEA stood for “It Does Everything Application” and I predicted the merging of what were then disparate applications like records management, document management, imaging and workflow into a single ‘it does everything’ application. It made sense back then and it makes even more sense now because most content management vendors now have applications that encompass all of the aforementioned applications plus many more.

We have of course also followed the same path and the RecFind today is a far cry from the RecFind of 1995. Today’s version, RecFind 6, is in fact the 7th total redesign and rewrite of our iconic product. Just like the grandfather’s axe story; “the handle has been changed 6 times and the blade 4 times but it is the same axe my grandfather used  60 years ago.” It is the same RecFind just improved somewhat over time.

We have in fact moved way past the original concept of an IDEA and the current RecFind is now able to be configured by our customers to be almost any application. This is possible because of a new architecture which allows RecFind to run multiple and different applications concurrently. We believe that this is the new model for all high end information management software going forward; a significant improvement over my original IDEA paradigm.

Our customers can now leverage off a single investment (RecFind 6) to satisfy multiple application software requirements. These include Asset Management, Business Process Management, Case Management, Content Management, Contract Management, Customer Relationship Management (CRM), Digital Asset Management, Document Management, Document Scanning, Imaging, Enterprise Content Management, Electronic Document and Records Management (EDRMS), Email Archiving, Email Management, Help Desk, Human Resources Management, Knowledge Management (KMS), Library Management, Records Management (RMS) and Workflow.

The big difference and major improvement over past generation systems is that everything is done via a high level configuration tool, not by changing source code. This means every RecFind 6 customer in the world runs a ‘standard’ version of our software even though each customer has RecFind uniquely configured to its precise requirements. This means every customer can benefit from our regular updates and new releases and that no customer is impacted by a ‘special’ maintenance contract.

When I presented the concept of RecFind 6 to a bunch of other vendors during a round table discussion at a records management convention a few years ago they all scoffed at the idea and said it was impossible. One vendor even said they had investigated doing what we proposed but that it was impossible.

As we developers all know, nothing is impossible in IT; it just takes time and money. RecFind 6 is evidence that it can be done and I already see other vendors following our lead.

The concept of  generic application solution (which is one way of describing RecFind 6) is one that both empowers and liberates the long-suffering customer. The customer can now change anything, including the data model, reports, language and any business process without waiting on the vendor or sending the vendor money. The customer can leverage off an investment in a single product to create multiple products and realize significant savings. Surely this is the model the world wants and needs, especially in these difficult economic times we all live in?

The records management process has been a bothersome burden since primitive man first began cutting notches in sticks and tying knots in string to record things. You can just imagine the excuses for bad record-keeping a few hundred thousand years ago.

“The dog ate the string.”

“The stick caught fire.”

Unfortunately, modern man has millions more things to manage than his ancestor and no more time (I imagine our ancestors were pretty busy just staying alive and catching food). In fact our ancestors had a few advantages. It is pretty easy and very fast to tie a knot or cut a notch and no training is required. Additionally, notches and knots are naturally multi-lingual and multi-cultural and avoid all the language difficulties we are faced with today. Knots and notches are also pretty durable (if kept away from the fire and the dog) and lend themselves to long term preservation. We on the other hand struggle with the need for initial and ongoing training and overcomplicated software and processes and paper and cardboard that disintegrates after the first serious storm cause leaks to appear in the roof.

Basically, modern man has a much more difficult job to do with harder-to-use tools. Is this what we call progress?

So, is the solution fully-automatic, self-registering, self-classifying, self-managing records? Is it possible? Do we have the technology? Do we really want to rid ourselves of this burdensome task?

In my mind the answer to all of the above questions is yes but then I do have an active imagination. I really do envision a world where the records manage themselves and all we have to do is search and find what we need. In fact, the next phase should really include some really very clever ‘push’ algorithms such that we no longer even need to search, the system correctly anticipates what we need before we need it, “Bacon and eggs for breakfast sir with a little wholemeal toast and the files on the Clark-Deakin case?”

Unfortunately I am also a software designer and developer of records management systems so when I say silly things like the above I eventually have to follow rhetoric with product.

I already have most of the core components and technology and algorithms and most are already in use, in various stages of development, in our current product range. The ‘fully-automatic’ part of the process is already evident with products like RecCapture and GEM but market acceptance of the fully-automatic paradigm is still a major problem so we have been reluctant to get too far ahead of the curve. I am prone to the disease of designing products about five years before people are ready to use them because I often let my enthusiasm and imagination over-rule a more conservative view of what the market is actually ready to accept. This is probably because:

a)      I am the owner of the company; and

b)      Designing and developing application software is still fun and exciting and even more so if you use your imagination to its full.

In subsequent blogs I will further elaborate on how this abolition of the records management burden is not only possible but probable in the near term barring a total collapse of the world financial system because Greeks don’t pay their taxes and banks give money to anyone in good times all the time hoping that the government will bail them out if it all goes belly up.

Frank