 |
|
|
|
|
|
|
||||||||||||||
 |
||||||||||||||||||||
 |
||||||||||||||||||||
|
NEW! - Your 24/7 Network TM |
||||||||||||||||||||
 |
||||||||||||||||||||
|
||||||||||||||||||||
|
||||||||||||||||||||
|
|
|
|
|
|
 |
|||||
|
Information Management in the Enterprise Finally an Information Model that is Profitable by Nature A Zybic Inc. White Paper Abstract Following this abstract are an enormous number of words bringing readers, who have the energy to expend trudging through them, a deeper understanding of what should already be obvious to most of those who deal with our complex information infrastructure. And that is…
What the reader may not have known is…
Now for those of you with a life that is short on free time, but long on responsibilities, we suggest that you stop here. However, before you put this away, you owe it to yourself to do two more little things. These are:
While this abstract is certainly presented “Tongue and cheek”, we are very serious. There are an enormous number of words in this paper. Somebody somewhere could benefit by reading them… Unlike many purely technical Information Management presentations currently in the public domain, this paper is a high level abstraction treatise. It contains a novel analysis, a unique paradigm as well as an introduction to their manifestation as a software application that is well synchronized to fundamental human attributes and expectations.
Time advances. We dream. We analyze. We plan. We act. We push forward. We have been getting more and more done. And apparently because of the way we humans are “wired”, we by our very nature need to collect, use, store and pass along the information gained. But it has never been easy. Fortunately humanity seems to be unflappable. We engage. We push back. We don’t give up. When meeting resistance, we typically step away, muster ourselves then find another way to push through; collectively vowing to be smarter about it the next time we find ourselves focused on a new challenge. This attribute appears to be part of what makes us - us. Since the first individual scratched some symbol on a cave wall, our species has devoted significant resources to build and refine our information infrastructure to better leverage any immediate and potential benefit to us and our progeny. So generation by generation, many important information techniques and technologies have been developed and applied to help us gain higher value. In the beginning, crude labor intensive solutions were the only things we could put in place. However, as our capabilities improved and our experience built, we developed more efficient and appropriate solutions. We discovered that while still requiring significant participation, effort and intervention these innovations consistently released us from untold hours of drudgery, allowing us to devote ourselves to far more productive and creative pursuits. And the results of these efforts surround us. Just as Newton and Einstein credited the “shoulders of giants”, we too can point to a myriad of examples where we are benefiting from the contributions of those who have gone before us. Like our predecessors, we greatly value and continually invest in our collective information infrastructure. And now, our accomplishments are piling up at an astonishing rate. In fact, never in mankind’s long history have we been able to accomplish so much, so quickly – and with so little effort. It is clear that accumulating and sharing information is a tremendous positive force in our existence. We want to do it. More to the point, we need to do it. However, there is still a lot of work involved. While information has the potential to benefit us, expenditures of resources for the mechanics of handling it are all overhead and waste. Information just doesn’t accumulate, package, store, protect and retrieve itself. Just how do we maximize the value? Well let’s use the top line definition of value equals benefit/cost. We can generally influence costs, but benefits offer much greater resistance to our manipulation. So humanity’s opportunity to manage value is routinely found in the cost category. Since well before the industrial revolution, we have reduced our information costs by offloading or transferring duties to lower cost providers. Later, starting in the 1950s, computers quickly established themselves as the preferred target. Ever since, we have proactively knitted them deeply into our infrastructure. The results have been remarkable. Simply, as we can do more, we do more. However, our systems and tools brought with them all manner of inherent limitations, weaknesses and shortcomings. People were always required to provide backfill efforts to make up for this. We willingly did so as it was very clear that the value returned from this cooperative relationship was still far superior to any other available alternative. However, just as our ancestors, we continued to push through and improve things for ourselves and our children. Today our skills and technological prowess have matured to the point where we are at the beginning of a new age - one where our technology produces far more while requiring far less of us. Specifically, we have arrived to the time where technology can serve us without demanding that we make tedious and non-value adding accommodations to it. Now, for the first time in our history, our technology is at the point where it has demonstrated its ability to isolate itself from us. It is able to support our expectation/ideal that it should do all of the work while unselfishly delivering all of the benefits to us. This is the preferred relationship where technology can be thought of as fully ‘accepting’ and supplying all interface and service cost burdens. By requiring less and less of our attention and support, it increasingly extends our ability to influence our own life’s paths and contributions, as well as favorably impacting the ‘cost’ factor of the ‘value’ relationship. Further, as we become more experienced in this new arrangement, we are beginning to appreciate that the highest value and best solutions:
This means that as a race, we are no longer forced to contribute to (or hammer ourselves into) technology’s rigorous adaptations, frameworks or translations. For us that work is pure waste and we will no longer sanction our part in it. To that end, we will continue to evolve our infrastructure such that it will be capable of making all efforts and adjustments that are required to interact with us in more palatable, natural and valuable ways. So for the first time in human experience, we can and do now insist that infrastructural solutions conform to our unique wiring, deliver value and stay out of our way. It is the position of Zybic that this is the obvious, logical, expected and desired outcome of an actively evolving information infrastructure built solely to benefit humanity.
History of Information The road to our ideal goal of having our information infrastructure do all of the work while unselfishly delivering all of the benefits to us is a long and challenging one. Sometime in our distant past we began this journey when we first learned how to communicate. Since that moment, all humans have struggled to improve the ways they collect, store, deliver and share information. First interacting with small and isolated communities, these early attempts were very difficult and inefficient for all involved. However the benefits must have outweighed the effort as our ancestors have actively and aggressively pursued improvements ever since. In time they discovered that people could augment and improve their existing capabilities by combining standardized approaches with technology such as the printing press, radio, telegraph, telephone and computer thus expanding the reach, impact and benefit of the information shared between them and across the generations. With each successive generation’s willing participation, this trend is ever accelerating. The value returned to mankind continues to grow at an exponential rate.
Data, Information and Knowledge Data, information and knowledge are very different things. Further, while we can never have enough knowledge, we are awash in data. There are literally trillions and trillions of bits representing this or that – piling up and patiently waiting to be usefully assembled and processed. A quantity versus relevancy hierarchy well illustrates this relationship. As the following graphic depicts, the higher in the “pyramid” one travels, the more useful the foundational data can become.
Data to knowledge However, this journey from data to knowledge is all uphill - it takes real work. It just doesn’t happen by itself. Further, we believe that "data" imparts value when it becomes "structured and navigable information" that is consumable by the human mind. Once information is consumable, it can be instantiated in the mind as knowledge. While people or systems can cull information out of data, only human beings can consume and render information. Then and only then does it become knowledge - embodied in the mind of a person. So data is translated into information by decoders and dictionaries. Information is translated into knowledge by the presentation and processing of information in the context of relational, spatial and temporal paradigms - things that all humans experience.
For instance, it might take the reading of a whole book on the design of a car engine to “understand” the operation of it. This book contains an enormous amount of data - the characters of text in the book. The “information” contained is provided by words, sentences and syntax that map out to real world dictionaries. Any “knowledge” (and semantics) delivered is not in the book at all. Rather, it is “materialized” in the mind that reads the book.
Information Exposed As soon as mankind recognized the value of accumulating and sharing data, we collectively began to amass great quantities of it. Soon thereafter we creatively crafted and implemented countless constructs - approaches, systems and tools to support this activity. Today we call these things components of our “Information Infrastructure”. So when we consider aspects of our modern information infrastructure, just what are we referring to? In the abstract, we are lumping together all man-made and machine/technology constructs that touch or influence data. Some facets dictate structure, others develop or promote standards, still others provide essential mechanics related to such things as the accumulation, parsing, translation, production, storage, filtering or the delivery of it. The list goes on and it is a long one. So to help ourselves, we have studied our information infrastructure and applied some structure to help us sort out, understand and manage it. Here follows some of the more important classifications, definitions and constructs in use today.
Origins Data of all types are used when people attempt to develop and derive information from it. However, common information materializes through only 3 basic steps. Individuals (with or without the help of their systems): Note that for a particular piece of information any number of iterations may occur in the create/review/publish cycle. Lifecycle Further each type of information has a specific temporal component which impacts how much maintenance its data requires to allow it to continue to fit into the context it was and should remain coupled to. We can explore this attribute by looking at examples of two different types of information: First let us consider a formal resolution produced by a Board of Directors. In this case, the document contains historical reference information. It is static and requires little attention since its content is essentially fixed in time. Further those who access it, understand that it is dated and is to be absorbed and analyzed in the context of its time of creation. On the other end of the spectrum we can use an employee directory. All directories are created to provide accurate and complete ad-hoc support to those who wish to look-up and make contact with someone. However, since every human will spend a limited time performing any particular job, the data will naturally go stale compromising its value – unless there is periodic intervention to update it. This must be done as those who access it have the expectation that the data should be a complete and accurate representation of the staff roster, each and every time they might choose to look at it. After all that is why it was created. From these common examples we see that some types of information can be neglected for extended periods without compromising their usefulness while others are ignored at our peril. Therefore, information lifecycles must be effectively managed. Often data retention and other “reminder” policies are crafted and implemented to do just that. Sources and Sinks Data is manipulated and information is generated by people and their machines. These information sources feed hungry consumers across every type of enterprise. Organizations are run by people - but people must consume information in order to effectively perform their work. A person or system consuming information is known as an information sink. Information flows between various information sources and information sinks. In support of the value equation, it follows that a balanced state of equilibrium whereby sources and sinks are reasonably equalized is optimum – because there are costs and consequences associated with imbalanced flow. That is, too many information sources result in: Too many information sinks result in the surfacing of systemic inefficiencies such as: Misinformation Accurate, germane information that is quickly and readily available is what every individual, business, social or political entity needs to operate smoothly and productively. Struggling with information that is out-of-date, inaccurate or misplaced dilutes any value returned. It can mislead us to the point where it: Reliance on quality information is essential to every successful (profitable) information infrastructure transaction. Noise Anything unwanted that a person has to wade through when searching for information they seek is referred to as “noise”. In this use, the word should be considered to be a derogatory term. This is because noise often gets in our way, wastes our time, slows us down, distracts us, leads us off track and otherwise inhibits our quick absorption of the information we are looking for. For example, internet search tool returns are routinely peppered with hits that are probably not useful or relevant to the needs of the user. Thus the term “surfing the web” has become embedded in our lexicon, as it is a colorful and suitable metaphor - perhaps suggesting that users skim over search results until something catches their interest. Typically, users must surf through many results to find the information they seek. This is a waste of their time. The Search Problem The challenge of finding information to help one perform his or her job can be frustrating. In order to speak precisely about finding such information, we need to understand where we are looking. The term “information space” is appropriate as it represents all possible searchable locations for data and information. Consider a new hire that is grappling with many unfamiliar terms, definitions, business policies, etc. This information is generally difficult to locate, access and cross reference - especially when it is really needed. Also, the information usually exists in different formats and is not linked or related to similar information in the enterprise. So casual navigation across the information space can be very challenging. The root problem is that there are many: Therefore, each user/employee must master different tools while being constrained by the limits of their associated paradigms. So, the crux of the problem is that it is hard to find information in the enterprise. One reason why is because of disparate data stores. A more fundamental barrier is the lack of a common information model or schema allowing consistently productive navigation through the information space. Delivery and Flow Let’s remain focused on information infrastructure associated with the enterprise. It is agreed that the exchange of information among employees, suppliers, vendors, customers is the key to an efficient enterprise operation. And people consume, produce or exchange information in many ways. For example, employees might: Corporate process engineers continue to design their business practices and daily routines around these established mechanisms of information handling and transfer. These tools have benefited commerce greatly, but have also set a precedent that is difficult to change since people now tend to think in terms of these toolsets. Thus, humanity has become entrenched in this traditional paradigm – the one that resists change. In order to simplify our understanding of information delivery and for us to move beyond the old paradigms, it is useful to generalize the important models of information exchange. Two main techniques are employed: Publish/Subscribe and Request/Response. For example, one may subscribe to a category topic (on a portal) and receive email notifications when new information is published based on that category. This Publish-Subscribe model is very powerful and provides a many-to-many communication mechanism for information exchange. Another popular way to acquire information is by pulling it. That is, explicitly retrieving information from a specific information source (e.g. running a report). This method is often referred as a Request/Response interaction. An information model supporting these approaches enables the delivery of powerful end user toolsets. In the enterprise setting “Information at your fingertips” used to be a popular mantra. Now with the right information delivery models in place this capability becomes real. Structured, Unstructured and Semi-structured Information is considered “structured” when a database schema imposes a structure (with relationships defined between the entities) on it. On other words, structured information is information that is broken into fields and stored in a database. This information can be browsed and queried using traditional information technology methods (e.g. SQL). Further, extraction of the information can be exploited through queries generated by common reporting tools. Unstructured information contains a variety of information, with no explicit relations with other documents or databases. Indexing engines are often used to index the content making them at least searchable (“full text searching”), but these documents stand alone in isolation (the “silo” problem). One of the most challenging information infrastructure problems today is the proliferation of unstructured documents that are not indexed or stored in document management systems. There is no way these documents can be related or “joined with” information in databases. Even if one has a document management system, the information contained within is not easily categorized or searchable along with information in databases. In contrast with formally structured information and unstructured documents, “semi-structured” information has started to make an appearance. This hybrid approach allows for a combination of rich text content and structured fielded data. Transactional and Non-Transactional Information comes in many forms. We generally are able to assign all types into one of two categories: Transactional and Non-Transactional. Transactional information is often thought of as data. Transactional information is what many information infrastructures ultimately process for our collective benefit. Examples include: Non-Transactional information is other types of less tangible, but equally valuable information. Meta-data: knowledge about how the business manages business. This information, embodied in the minds of every employee - “intellectual capital”, drives the execution of the business. Collectively, this information is the lifeblood of the corporation. And it is the information that is most often neglected. Common solutions attempting to capture and manage this information are characterized in an alphabet soup of acronyms: Historical approaches to mining non-transactional information included using multiple toolsets that were based on different and/or totally divergent paradigms. These “classical” tools had no common foundation and no relationship amongst their data sources. Entities & Relationships Relationships are the glue that bind information elements together and allow for navigation through the information space. They provide a form of “contract” between entities. In the world of documents, relationships to external documents/data sources are not easily accessible in the content of the document. Traditionally, it is difficult or impossible to navigate these relationships to mine or discover new information. For instance, objects referred to in corporate documents may not be defined at all. Furthermore, there may be no simple way to navigate to the related objects. The most one can typically hope for is that a “related documents” section will be available - perhaps containing hyperlinks to live external documents. Categories We humans either just like to categorize things, or simply do it to maintain our sanity. When we dive into in a sea of deep complexity we feel compelled to define and apply categories. It’s a good thing that we do too, since the act of categorizing allows others to browse for information by category. Classifying information in multiple categories provides a powerful means of finding information in n-dimensional space. Today, browsing and searching are the primary means used to locate desired information. We browse a table of contents, look something up in an index, enter a search phrase on a web search engine… So we willingly expend effort indexing and categorizing information in our information infrastructure so that future instances of searching become much more efficient. Taxonomies Since categorization is a “natural human tendency”, we rely on this instinct and ability to make it easier to find information at a future point in time. Furthermore, by introducing hierarchy into categories, a powerful classification system is realized that allows unprecedented searching and reporting capability. This hierarchical category management is called taxonomy. Taxonomy is essentially a hierarchical classification system. By classifying information through categories, we gain the power to call up or query the information at a later date using category information. The more well thought out and detailed the categories, the more effective and granular our queries into our information space can be. Knowing and navigating the taxonomy (hierarchical categories), allows us to more efficiently browse our information space. Further we can find information related to general or specific categories without dilution of the returns due to noise. For instance, consider a category called Finance. Within this category one might have sub-categories “Accounts Payable”, “Accounts Receivable”, “Collections” and “Contracts”. Under the “Contracts” category you might find sub-categories for “Vendor Contracts” or “Purchasing Contracts”. As you can see, categories are not unlike folders in a filing system - they are used to find information quickly. Powerful hierarchical categories allow us to quickly find relevant information either by “rolling-up” queries into more abstract ones or by “drilling-down” into them. Enterprise Organizations There are many ways to find information in an enterprise. Categories/taxonomies and searches based on keyword are common methods. Another method used to discover information is to browse by organizational department. For instance, if someone knows they are looking for Finance-related documents, the Finance department could be browsed or searched on the corporate portal. In essence, the enterprise organization becomes a primary filter for that information. Further, since the information is typically authored by people who are members of their own organizations, access to the information is often restricted and exposed only to those with proper qualifications. Authorship of information is maintained along with organizational relationships thereby producing a powerful filtering/security capability. Value Added The phrase “value added” is used to describe a modification or increase in value of a “product”. A value added activity is one that somehow modifies and increases the value of a product. In this context, a product can be literally anything from an intellectual construct to a physical item. Further, 'increased value' is typically evaluated from the beneficiary’s perspective. For example, when data is converted into information, value has been added to the data. Value Added Cost Variance “Value Added Cost Variance” is the difference between the ideal and the actual cost to produce a desired value added. No matter how well understood or controlled, each and every instance of value added has a unique cost associated with it. If individuals and/or systems are consistently knowledgeable, experienced, properly supplied and supported perhaps the variation will be small. However, this is often not the case. Unless all factors are consistently “on the mark”, costs (waste) accumulate quickly. Impedance Impedance is a term of art typically used to refer to the ability of one system to efficiently accommodate the output (energy, information, etc.) of another. It is used and measured in many ways, but in general the most efficient exchanges between different systems happen when their impedances are closely matched. The “Impedance Mismatch” Problem Often the attributes of each system’s interface are cast in the laws of physics and are immutable. We must accept them and do our best to work around them. Other system interfaces while perhaps appearing rigid may in fact be adjustable. So these types of systems can usually be modified allowing them to directly couple with others (so avoiding linking them through some sort of less efficient conversion device or process). Here are some common examples where the effects of impedance are evident: Mechanical – The Automobile Transmission An automobile transmission’s sole purpose is to better match the engine’s output to the changing demands of the road. It is a necessary accommodation we make that addresses the variable torque output of the internal combustion engine to be suitable to the challenges of accelerating a vehicle. To match the system impedances, the driver or the transmission itself adjusts the gear ratio to deliver the best overall efficiency and performance under constantly changing conditions. Interesting, in the early days of the automobile, steam powered cars were very popular. They did not have or require transmissions as the steam engine impedance was a much better natural fit to the application. Electrical – The Transmitter Antenna A radio transmitter must use a tuned antenna to enable it to efficiently radiate its output into the surrounding airspace. Without the right length of antenna for the operating frequency, most of the energy will not make it out. So to be efficient the antenna’s properties must be adjusted to match the impedance of the aether. Human/Human Communication - Language If you only speak and understand English and you wish to communicate with someone who only speaks and understands French there will be little information exchanged unless you find a way to match impedances. Your only solution is to employ the services of a translator to convert and bridge your communications. While you have not been prevented from communicating, this approach is much less efficient and much more expensive than communicating in a common language. Human/Machine Communication - Language All computer based systems are built up in layers of function. At the foundation is binary machine code. However, people do not easily or naturally communicate in machine code. To deal with this impedance mismatch, we have transferred much of the translation effort to the machines. Layers of machine software continue to be developed to make the appropriate interface translations more natural and efficient for us. This slows the machines down, but we are happy with the arrangement. As pointed out by these everyday examples, we see that direct matched couplings are always better (more efficient, less costly, etc.) than relying upon some intermediary conversion/linking device or technique. However, if accommodations must be made, it is better for us when our technology makes them. Impedance - Human and Otherwise So when we speak of impedance mismatch during a discussion of our information infrastructure we are referring to how efficiently and naturally coupled our systems are to each other as well as to us. Both people and computers can be flexible and adapted to the requirements of some challenge before them, but here we are interested in pointing out how each one is intrinsically wired or set-up. Computer Impedance Digital computer based systems run on binary code. Information represented as bits set to represent either one or zero operates on untold numbers of simple logical elements to produce complex and useful work. However, since many families of microprocessors have been developed and deployed there are several lines of incompatible low level syntaxes and instructions sets. (Think of these as incompatible grammar and vocabulary.) Fortunately, many layers of code have been developed and applied to allow low level compatibility issues to be masked through translation. While native, common binary instruction sets are the most efficient, the power of today’s systems is so great that translation and interface demanded between incompatible systems is of little consequence. Therefore, for all intents and purposes any microprocessor based system can be made to emulate or behave as any other. So they can efficiently communicate among themselves in essentially their natural languages. Human Impedance Human based systems run on who knows what. But it certainly isn’t binary code. Human beings are sentient and operate via associative memory. Thoughts are triggered and cascaded by association. This is why many of us “go off on a tangent” when engaged in a conversation - one thought is linked via an association with something else, perhaps totally disconnected and unrelated. To us words and phrases mean things, triggering us to evaluate them in context. To computers, words are no more than data to be manipulated – data without meaning to be acted upon without interpretation. Further, while admitting to our fair share of incompatible syntaxes, grammars and vocabularies, as a community, we do enjoy a fluency in far richer language structures. Unlike computers. humans by nature: Could these two groups be any more different? Not really. Admitting that we truly need our machines and are not about to willingly give them up, we continue to resolutely address the inherent impedance mismatch issues that separate us. Impedance Matching - Who (or What) Should Make the Effort In the early days of our relationship with our machines, we flipped toggle switches or punched paper cards attempting to communicate with them in binary or some other very low level variant. We had little choice in the matter as it was far simpler to adapt ourselves to their requirements than force them to be supportive of ours. Back then the impedance mismatch between us was as pure as it was large. So as a community we devoted significant resources building layers of abstraction and interface capabilities that now allow these systems to interact with us in more human or human-like ways. This is an ongoing effort. Where is suits us, we design our systems to become more and more fluent with all manner of unnatural languages.
Modern Components of our Information Infrastructure As our skills, experience and technology allow, our species has crafted and implemented manifold tools and approaches supporting our efforts to better manage and benefit from all things informational. Over time, these have been widely applied and now touch almost every human activity and transaction. Empowering mankind with such systems has produced great productivity gains. Typical systems included in this mix are:
Follows a more detailed description of selected information infrastructure functional elements currently in widespread use:
E-mail and Messaging While their role is limited to essentially one of transmitting information between entities, these “store-and-forward” computer based technologies are widely embraced being used for everything from ad-hoc, frivolous personal banter to formalized structured data transmittal. Humans interact with them in very intuitive and natural ways. The information content in e-mail is typically unstructured and narrative in form. Collaboration/Groupware These heavily formalized tools allow groups to collaborate by facilitating the sharing and exchange of ideas, documents and other information. Online “notes” databases, shared documents etc. have been popular for some years, but require the user and support community to conform and support unnatural frameworks. They can be characterized as awkward to use and therefore remain unattractive to the masses. These collaboration environments are often semi-structured in nature. Portals Portals recently exploded in popularity due to such attributes as their ease of information access and rich WYSIWYG capabilities. As with e-mail and messaging technologies, this was a major breakthrough. Portals: However: So portals do not solve the crisis that has plagued the enterprise since the dawn of the information age: multiple data sources with multiple data models (information silos). For instance, a “person” object in the HR reporting system (available through the portal) may not have the same attributes as a “person” object in a contracts database (perhaps also browsable via the portal). Browsing and reporting information across these separate data sources may be impossible without adding a set of complex application integration interfaces to let cross reporting bubble up to the portal layer. Modeling Tools Many approaches have been developed and commercialized attempting to help people and organizations understand, analyze or modify how their processes operate. This continues to be a very fluid area. In fact, support and user communities wanting to understand their workflow and work processes are often hard-pressed to find a “comfortable” solution as all approaches force the user community to adopt “unnatural” paradigms. Workflow Management This category of business information is “non-transactional” in nature and essentially describes the business model and how it is to be executed. Workflow engines have provided the means to control human / system interaction in series of discrete work steps. These mechanisms enforce repeatability within the enterprise’s implemented business model. Document Management In order to comply with various regulatory mandates (ISO 9001, Sarbanes-Oxley, et.al.) many companies have implemented document management systems. These systems typically hold “official” copies of documentation in their repositories. While robust, these systems are often highly proprietary, difficult to use or systemically integrate. Knowledgeware We find “Knowledgeware” offerings to be similar to those of “Modeling Tools” in that they purport to retain corporate knowledge in a central repository that can be easily accessed. Unfortunately, the people who need this information often do not have access to the appropriate tools. Further, knowledgeware products are specialized for specific purposes and vendors often require multiple modules and “expansion packs” to use their tools in different areas of the enterprise. Artificial intelligence Artificial intelligence vendors have touted AI-based knowledge systems for 20 years. These systems claim that rule-based inference engines help churn through their knowledge bases and answer queries. Unfortunately, the problem with these systems is associated with the “fleshing-out” or populating of the knowledgebase. This can require special skills and be a difficult and tedious chore. Blogs and Wikis Recent history has seen the explosion of the “Blog” (short for “WebLog”) as a medium for sharing information. This utility allows any number of people to edit online “diaries” or documents on the fly using an HTML interface. This model combines the ease of access of the portal with the ability to instantly publish content to the online community. However, a problem with blogs in the corporate enterprise is that they are too unstructured, uncontrolled, and often degenerate into traditional “newsgroup” style stream-of-consciousness” utterances. A “Wiki” is an online web-based encyclopedia that is “open to the public” for instant updating. This forum allows the quick publishing of content without the need for fancy authoring tools or complex approval workflows. A “knowledgebase” can be quickly assembled using this paradigm. The Wiki model allows for very efficient collaboration across any conforming information domain. Blogs and Wikis support the “natural human tendency” to ramble and wander through information. This is the way that humans get many things done. Instead of presenting information in a way that is convenient for a machine (e.g. relational tables/SQL/reporting), this exposes information in ways better suited to the often spontaneous demands of the human mind. RSS Real Simple Syndication (RSS) is a recently developed protocol for culling information from disjoint news feeds. Although its original intent was to make it easy to publish links to content that could be later downloaded and read offline, its popularity has soared as a general purpose utility. So these represent some of the most common - yet sophisticated, elegant and capable information infrastructure components helping us. Unfortunately, they often can only be used independently of one another. So people are required to switch between several different access methods, interaction modes and skill sets to interact with the various pieces of information each system is designed to manage.
Wrestling with the Blended Legacy Information Infrastructure With so many independent and isolated information infrastructure components, it is no wonder that so many of today’s organizations are still thinking in terms of legacy information systems. Under these traditional paradigms they can’t effectively break out of the information silos that have both served and plagued them for years. This outcome is reasonable and expected since it is in our nature and capacity to implement, and then incrementally evolve our systems - individually. Therefore, while they independently provide a benefit, they usually do not integrate well with other systems (which are following their own equally unique missions and evolutionary paths). So enterprises are left with a “blended” legacy information infrastructure. A complex construct where people attempt to creatively minimize existing multi-system impedance mismatches such that the interests of the enterprise are protected and the user community is served. However, due to inherent issues flowing from such complexity, value returned is not reliably measured or typically responsive to manipulation. Still many organizations have had reasonable success improving their particular infrastructures, aggregating information through such accommodations as web portals, single sign on technology and untold application integration programs. However, even when these improvements are well deployed:
While organizations continue to devote significant effort in attempts to improve it, the current information infrastructure remains a complex mix of loosely associated tools that are not collectively integrated to best serve mankind’s fundamental requirements. This is an expected outcome since these tools were created out of necessity and without deference to longer term integration and impedance considerations. Simply stated, they are all niche functional elements applied to specific and limited areas of human pursuits. However, knowing what we know now and having the technical capabilities we have now – would we have built our information infrastructure this way? Would we have allowed this outcome to materialize? No. We would not have. In a fantasy “do-over”, we would have first stepped back and pondered the value equation as impacted by such things as impedance matching. This would lead us to investigate how human’s best interact with themselves and their support infrastructures. We would then have seen the long term value of designing matched and well integrated functional elements – all leveraging a universal information model – a model that was closely associated with how we humans interact and organize ourselves to accomplish things. With that foundation in place, we would begin to build. Well it is too late for that now. But, it was never a real possibility anyway. It just isn’t in our nature to sit and wait around until every t is crossed. However, perhaps a review along these lines would help to shape our thinking and benefit any future efforts.
So How do Humans Organize Themselves to Get Things Done? Whether we are speaking about governments, business or social organizations, when we step back and devote effort toward understanding how our species organizes itself to accomplish things, we find that throughout our history there have been only two fundamental approaches. While facets of these have been mixed in all manner of combinations and applied to manifold circumstances, they are indeed the structure mankind has imposed on itself to produce value through organized efforts. Specifically, since our beginning, people with power (or those sanctioned to wield it on their behalf) committed to some goal they wished to achieve. They then either:
Of these two approaches, the first has a long history of being practiced within authoritarian and often oppressive organizations. Here the potential contribution of the individual is always diluted by the overhead of any imposed servitude. Further, individual initiative is actively suppressed, since it is often correctly interpreted as a threat to the established chain of command or power structure. The later approach is most widely applied by free and enlightened organizations. Individual initiative is encouraged and promoted as a way to reduce costs and improve results. In fact, this approach has proven itself to be generally more effective and certainly more sustainable. Notice that it is also in harmony with mankind’s inherent desire to improve their individual circumstance according to their individual needs and sensibilities. Simply because we are human, suppression is always expensive and unstable, but promotion is seductive, profitable and well tolerated. In other words, the second approach is the only one that is aligned with how humans are wired. It should not surprise us then that this particular fit of man and structure facilitates a more natural and productive cooperation.
The Universally Preferred Human Process Mankind’s long running efforts to cope with our complex existence, has forced us to come to terms with the fact that we are imperfect beings. In fact, during our uniquely human adventure, we have discovered that we are often in control and are able to deftly steer the course to arrive at our desired goals. Unfortunately, other times have demonstrated to us that forces beyond our original perception, understanding or influence somehow conspire to throw us off course, forcing some adjustments to be made or persuading us to abandon our objectives outright. Accepting this reality of our common existence, untold individuals and organized groups steadily trudged along. In spite of the intrinsic risks, pain or uncertainties they faced, they accomplished great things. To our great collective benefit they also preserved and shared their information across the generations. Today, win, lose or draw, we too always muddle along – even grudgingly bearing any additional costs, resulting delays or associated unintended consequences. This is because we know that if we apply ourselves we can expect to eventually make real progress. We can and do accomplish things that are important to us and to our children. These experiences have taught us something else as well. Specifically, we see that our most effective structures are the ones that are in harmony with wired-in human attributes. Constructs that compliment our wiring’s perceptions, methods, expectations and quirks are often the most easily accepted and productively utilized by us. It is just the way it is. So humanity prefers to cooperatively accomplish things in the following way.
This then is the essence of organized human behavior. It is stunning how such simple elements have supported the development of such a complex and rich outcome.
The Current State of the Process Paradigm So since the beginning of time, when we decided to get something done, we somehow organized and engaged in and/or operated through processes. In fact, processes are still the vehicles we use to translate our efforts into desired results. And our processes are still created with the intention that they will help us accomplish something. If properly crafted, supported and executed it is our experience that they can be expected to deliver their intended value - at least in the near term. Since all processes are created in a prior moment of need and subordinate to the vagaries of their specific time, they are unfortunately static instruments with a limited effective life. Yet we need them and we often force them to support us during very long and dynamic associations. Nevertheless, from the time that they are instantiated, their efficiency/impact will change. Left to themselves, they will likely degrade during their long term use. However, if we are skilled, experienced and dedicated to periodic and incremental intervention as we play our role in their development and maintenance, they can continue to have a positive impact. If we neglect our responsibilities, the outcome will be less favorable. To explore this, it is sometimes helpful to consider a process to be a living creature. With that metaphor, we can say that processes are born, they are shaped, nurtured and eventually die. Until then, they produce, they consume. They can be productive, they can multiply. We can shape them. We can ignore them. We can benefit through our interactions with them. As with an animal, a process can serve us, or just as easily - we can end up serving it. Under different conditions they can become either a blessing or a burden. Most importantly, to remain “healthy” and to provide a positive impact, processes need to “evolve” and adapt as internal and external pressures modify their “environment”. As in life, environmental pressure is the key driver of this evolution. We all have learned about the impact of the environment on shaping species through the process of natural selection. Somehow, most creatures find a way to incrementally modify themselves to better meet the challenges imposed by their changing environment. Unfortunately, as sophisticated as today’s technology is, man made processes are still “unaware” of their particular situation. Worse, they are largely unable to modify their response to pressure of any sort. Unlike animals, processes rely solely on us to evolve and adapt them to the current requirements of their environment. Unlike living creatures, they can’t do it on their own - at least not yet. So, due to environmental pressure/change, all processes need some of our attention to “prosper” and help us over the long term. As their “masters” it is our unique responsibility and obligation to periodically:
In general, the frequency and magnitude of our attention varies with the need, impact, environment and pressure surrounding each circumstance. Critical, high impact processes are likely to be more regularly attended to than those of less significance. However, in all cases, it is up to us to determine the levels that once crossed will trigger our intervention. We decide what we are willing to invest to evolve them. There is a real cycle here. We have learned that:
For now, through our direct intervention, many processes will continue their service for years, even decades out. This is possible as on the occasions when conditions increased pressure and called into question the benefit/impact of them, people intervened and the processes were eventually adapted to reclaim the intended benefits. So we continue to periodically step up and do our part of this cooperative. Sometimes we are on the ball and make many small and timely adjustments preempting any significant negative consequences. Other times we choose to hold back and make fewer but larger ones. Sadly, once in a while we are too distracted, ignorant or worse and we only become engaged after being put on notice by some poorly performing process. This notice can be served up in various serious public manifestations – customer service problems, declining sales, increasing turnover and so on. For example, especially when times get lean and profits slump, people with oversight responsibilities notice the negative contributions of dated, inefficient processes. Specifically, organizations continuing to utilize flawed processes bring attention to themselves as they often waste the funds that are really better directed toward meeting other important obligations. This increased visibility quickly translates into pressure bearing on the local management. Soon outside “Help” in the form of analysis and intervention is often forcibly delivered. Later when the newly evolved process demonstrates that it can better cope with current requirements, some measure of trust and confidence in local management will be restored. Only then will the “extra” oversight attention be redirected. While common, this scenario is to be avoided. Therefore it is generally in our collective interest to periodically target less efficient processes for review – and evolve them before they publicly falter, causing us embarrassment or attracting outside intervention. So it goes for all current man-process relationships. It is work, but the return makes our investment worthwhile. We see that all processes are created in a unique moment and under unique circumstances. They are chronically influenced by internal and external pressures. Each one needs to be adapted through our intervention to continue to make a positive impact. Currently, they rely solely on us to find, implement and audit the proper adaptations. It is likely that until our systems become sentient, we will continue to happily devote ourselves to evolving our processes. After all, once they have been created, it just isn’t a heavy burden to occasionally step in and give them a tune-up. Yet it remains another example, albeit more subtle, of effort demanded of us by our own construct’s current limitations. While necessary, our efforts expended in these areas still negatively impacts our costs, diluting the value returned to us. Here too, it is in our collective interest to continue to shift more of this overhead burden to our technology.
So What Produces Waste in our Information Infrastructure? Members of our species remain very interested in results and the costs involved in achieving them. It is ingrained into our nature to expect a reasonable return for any effort expended. Since we have finite limits on our time and resources, we often resent being forced to underwrite any waste. In fact, few things frustrate a person more than dedicating themselves to something that proves to have been a waste of their time. Of course it is our time and therefore our prerogative to do with it as we will. So it is our unique privilege and responsibility to make judgments regarding what we are willing to invest toward the support of our current infrastructure. (Remember: value equals benefits/costs.) In fairness, these judgments are influenced by temporal considerations as well. As time passes we become less appreciative of the benefits we enjoy and more annoyed by the limitations that accompany them. For example, we complain about reloading toner in our copying machine. How much sympathy would we receive from even a very recent ancestor? In any event, and in spite of the many blessings our current information infrastructure delivers today, we see that there are certain recurring operating inefficiencies and costs associated with it that we are not happy about and only grudgingly support. For example, these problems manifest themselves when an individual or other entity:
We quickly can arrive at the roots of these costs. They are:
This state is to be expected as our information infrastructure is an amalgam of an incredibly diverse mix of technologies, structures and methodologies. It currently helps us do wonderful things, but due to its shortcomings it demands that we support its activities, address its problems, fill any gaps and plug all leaks. Given that we need information to perform our work and better humanity’s collective circumstances, it is logical that we built and continue to support this enormous information infrastructure. However, it is currently as unwieldy as it is valuable. Therefore, we see great opportunities to continue to evolve it in ways that better fit our requirements and expectations.
The Ideal Universal Information Model
The Need Data is everywhere. Information swirls all about us. Let’s think a bit about how we as individuals and employees swim around in our information infrastructure. Specifically, we often need to “search the enterprise” for all relevant information (policies, procedures, processes, facts, etc.) on how to accomplish something or other. These days we often begin by exploring the corporate portal. While portals can make various, disparate content accessible and hide the complexities of information gathering, they do not “understand” the nature of their individual data sources. Further, they can’t refine or package information at a more abstract and useful level. In fact, portals are typically little more than a seductive facade leaving us thirsty for a more complete and useful solution. However, our current information infrastructure is better off with than without them. So as luck would have it, we find at least some of what we were looking for through it. Then we remember that there are public folders somewhere on the LAN. While they appear to contain a tremendous amount of information, to get anything out of them requires users to have a fair amount of knowledge about their structure. We have floundered around in there before, but still it may be worth some of our time. We probe, but it is hit and miss and we come up a little short. Finally, we employ a local search engine that perhaps has indexed the corporate portal and fileserver on the LAN. To our surprise, our search returns many “hits”. However, surprise leads to disappointment when we realize that what we get back is mostly “noise”- matches that do not contain the information we are looking for - just a meaningless jumble of irrelevant links. This happens because the search engine is “dumb” - it doesn’t understand the entities or the relationships between the various information sources. All it is capable of are word and pattern matching. While the syntax matched, the semantics were all wrong. Frustrated, we soon find ourselves enlisting the help of others. So we make a few directed e-mails or telephone calls - seeing if we can convince them to join with us in our quest. Hopefully, one or two of them will supply the missing information or at least tell us where the silos are. If they do, we should soon have all of the pieces we need to set out and accomplish what we originally wanted to get done. It is important to note that all of these activities have added no value. They are wasted effort. They are added costs. Who wouldn’t rather avoid all this waste? Why can’t we just search and browse through a single location to find all of this information? What currently drives this outcome? Follows some key factors: To solve these problems we need to develop and adopt an “Ideal Universal Information Model” that will address these current information infrastructure limitations, be well received and ultimately return a much higher value. General Attributes The authors recommend the adoption of an Ideal Universal Information Model that addresses current deficiencies while shaping its structures to conform to immutable human characteristics. This particular model or paradigm ideally should: Just how do we design this ideal model? We do so by first paying close attention to the needs and quirks of the intended beneficiaries - us. Specifically, we note that individuals: These attributes appear to be universally human. No matter the vantage point, they seem to ring true. We should not discount their potential impact when we design and adopt support solutions. Then what would the attributes of the ideal universal information model be? They would: Classification Subscribing to the finding that mankind is wired in distinct and immutable ways and that we are most comfortable operating within a limited number of well defined paradigms, we can extrapolate and craft a classification model that is a best fit with us. Consider what one would need to categorize any information related to any man-made organized structures or enterprises. Six general categories are appropriate. These are:
Enterprise Entity Abstractions These abstractions represent a true reflection of how human beings naturally interact in the world. They provide nearly unlimited freedom while promoting just enough structure to protect and leverage information’s value across the population and through time. Remember in all organized human activities we start with an idea, then we communicate policies and directives intended to promote and shape how we wish to reach it. We then build and operate through processes to accomplish it. Whether you are running a household or managing a space program it is the same. This ideal model is impedance matched to how mankind naturally builds, uses and leverages information. We have classified manifestations of this Ideal Universal Information Model to be SKIERs as the acronym of Structured Key Information Enterprise Repository.
The Structured Key Information Enterprise Repository™ SKIERs are elegant and direct manifestations of the Ideal Universal Information Model. Built on well established themes and conventions, the SKIER is crafted to fully conform to human impedance requirements. It is structurally in harmony with how our species naturally interacts with each other and with our support infrastructure. Further it imposes just enough structure to guarantee the highest theoretical return on the value equation. As the model dictates, a SKIER is composed of primary entities (classes) that are well-understood within the process modeling community. Notice how these align with human organizational structures described earlier.
Model Type Descriptions The parent type is the Entity. Entities can be one of the following types:
Entity Types It should be noted that “Entity” is equivalent to “Class” in traditional modeling paradigms. All entities have common attributes (name, author, organization, description, etc.) and behaviors but are specialized according to their specific type: Policy A Policy is defined as a fundamental directive (often formally published) which is intended to insure that processes crafted and subordinate actions taken conform with and support the mission through sanctioned means. Policy may imply or suggest necessary processes. Process A Process is defined as a formalized, sanctioned and required flow-of-events detailing some specific value adding pursuit. These instruments typically provide detailed instructions, methods and requirements. Process are implemented by Role(s) Role A Role is defined as a function performed in a particular operation or process. Any individual or entity performing activities as permitted and/or demanded by a process is said to be acting in the context of a defined role. Specifically, to accomplish some type of value added, roles act on or are acted upon by processes. Roles are filled by a Party. Fact A Fact is defined as: Party A Party is a person, group, system, organizational entity, etc. Party represents the “actor” element in the enterprise. Examples would include employees, suppliers, customers, systems, special interest groups, external business partners, etc. Place A Place represents a physical location. Places are defined by their address which uniquely identifies them. This mechanism provides a means to describe and enumerate important places in the enterprise. For example, “3rd floor cafeteria” may be an identifiable place with appropriate information for employees (meals, serving times). Other examples would be “Technical Library”, “Northeast division office”, “customer locations”, “Conference room A”… A Unified Theory of Information Management The SKIER entities alone represent but one dimension of a three-dimensional navigational space for the corporate enterprise. The enterprise organizational structure is considered another dimension and a general set of classifications (taxonomies) the third dimension. All have hierarchical relations amongst their instances.
3-Dimensions of Classification So in total, information is classified into 3 dimensions all of which can be searched or browsed. The x-axis represents typed entity information. The y-axis categorizes the information entity according to the corporate organization/department that is applicable. The z-axis is the functional category (within the corporate taxonomy to which the information entity belongs. For instance, we may have an entity called “Inventory Receive Process” classified as a Process (x-axis). This process may be defined and owned by the finance department; therefore it is defined in the context of the Finance department (y-axis). Furthermore, this entity may be classified into a functional category called “Asset Management” in the corporate taxonomy (z-axis). As depicted below, classification across multiple dimensions is a perfectly natural activity in an enterprise setting. Information is collected and analyzed by people then classified by entity, organization or category (taxonomy):
The Classification Process To summarize then: Relationships Taking our “Receiving Process” example a step further, this process will have associations with other enterprise entities. For instance, within the narrative description of the receive process we can hyperlink to related corporate processes/policies/facts, etc. As an example, Hyperlinks are embedded in the following narrative fragment. Receiving Process: We see that the above process is defined in traditional narrative with references to other entities. These references are in the form of hyperlinks. Note that a relationship exists between the “Receiving” process and other entities (e.g. scanning, barcode, etc.) in the Receiving process definition. Note that the relationship definition is inherent in the narrative text. We could depict our entities and relationships graphically using visual modeling languages like the Unified Modeling Language (UML) or Entity-Relationship-Diagrams (ERDs), but we would loose the power of the narrative approach where the semantics are easily absorbed by fellow human beings. Interaction Metaphors People access information differently depending on context. Most often we choose the fastest method to find the information we are looking for. For instance, at one moment I may be able to quickly find what I am looking for by using a cross-reference. At other times it may be faster to find information by looking for a specific creation date. SKIER allows information to be populated/queried using all important metaphors: What this Means to Business… Knowledge is truly in the mind of the beholder and mankind is “wired” having unique impedance. We rely on a heavily associative thought process, using symbols, alphabets and narrative text. Consider any free-running conversation. During its course, the topics will ebb and flow triggered by such things as specific words and/or phrases. This is because any word or phrase can trigger a new thought, which often has the effect of steering the participants into a different area of discourse. However, when concluded, they may remember “where they left off” and pop back to the original context of their conversation. In this scenario they have likely traveled forward, backwards, sideways and perhaps a bit off the original tangent. The point here is that narrative text, with bi-directional hyperlinking, models the way we as a species think and interact. We think jumbled thoughts, traverse various paths, intuitively drill down and pop back to the “top level” - and this is why we at Zybic believe that a simple information model with bi-directional hyperlinking capability provides the highest value. Once information elements conform to this Ideal Universal Information Model, they are easily navigable and discoverable. “Information at your fingertips” becomes a real possibility. The value equation is fully leveraged throughout the enterprise. Harnessing Collective Employee Knowledge Employees often uncover important business information during the performance of their duties. However, while they may personally expose relevant facts and valuable “institutional knowledge”, they are rarely expected to package and make such discoveries available to anyone else. Certainly people take notes. However, they take them in an amount and level of abstraction according to their own particular needs. They also produce them in a format and media of their choosing. It is no wonder that such manifold information resources are not generally published for use by the general employee population. So any wider benefit of this information is lost. Value is not compounded because this unpublished information is inherently not sharable. Then there are those things that we may have not have ever taken notes about. For example, just prior to an employee leaving the company, a “brain dump” is hastily undertaken to attempt the extraction of the employee’s knowledge and domain expertise. This process is often ad-hoc and incomplete. Many times the employee walks away with valuable information that is never passed along to successors. So effort is chronically expended by countless individuals with an interest or need to recapture the same information - over and over again. Sure there are almost always ways to reconstitute missing information, but there are costs associated with that effort. Ultimately, your employees “own” your business processes. They execute them everyday, supported by rules of thumb and common knowledge gained through their collective experience and training. Unfortunately, the information and knowledge that they possess is often not in a useful form or broadly available. Worse, information is often far more valuable, in different times, in assorted areas and under different circumstances - all of which are not foreseeable by the original “discoverer”. This bliss of ignorance costs lots and lots of money. Why should this be the case? Couldn’t we have each employee “encase” their knowledge in the form of a key information repository instead? Furthermore, what if most employees stored their job functions, procedures, processes, etc, making them: It is clear that this capability would greatly impact the value equation.
Introducing Zybic Enterprise (ZE) You know things. But when you need to learn a bit more, you want to know:
Aren’t these expectations what you really wanted from your information infrastructure in the first place? Haven’t these capabilities remained out of reach for too long? We at Zybic have devoted our efforts toward more effectively aligning the information infrastructure with these fundamental human expectations. Our Structured Key Information Enterprise Repository (SKIER) paradigm points out how human beings should naturally couple to their constructs - thus minimizing infrastructure costs while vastly increasing the value returned on any effort expended. Further we have designed and built an enterprise tool called Zybic Enterprise (ZE). It allows any organization to capture the benefits delivered through adoption of the SKIER paradigm. In other words, it is an information infrastructure application crafted to fit how people are wired into their existence... So, Zybic Enterprise makes organizing, managing, exploring, running or auditing organizational units and understanding their value to the enterprise simple. Specifically, ZE is a unique and flexible on-line application. Like the spreadsheet or word processor, it is a tool that supports and improves how everyone performs their work. It is focused on delivering information that can be easily found, understood and relied upon - with low effort and in the moment that it is needed. While ZE can be used for any Key Information Repository (HR, Operations, Sales, IT, Admin, Engineering, Purchasing, Distribution, Accounting...), it also widely promotes the coordinated management o | |||||
