- |
Page Links |
|
||||
The "physics" of information.
Both formally in systems and intuitively in our daily lives we struggle with information workings of which we aren't fully aware.
Information Origins and Precedence
How we become informed.
Information Dimensions
The four dimensions of information .
Information Vectors
Paths followed within information dimensions to derive new information from existing information.
Supercession
Just because information is no longer current does not mean it is no longer information.
Delta Vectoring
What did we know, when did we know it, and what did we do about it.
Correlating Information
Beyond hierarchies and relationships. How information correlates.
Information Origins and Precedence |
   |
How we obtain information (origination) reflects the order in which we obtain it (precedence).
1 |
1a |
2 |
|
ORIGINATION |
Determination |
I-Chronology |
Derivation |
PRECEDENCE |
Primary |
Collateral |
Secondary |
Determinant-Primary Information is obtained through direct experience (observation, interrogation,or measurement) or it is defined by appropriate authority (a child's name given by their parents). It is first received information. In conventional terms it is "input" (but input to us, not to a computer).
I-Chronology-Collateral Information is chronological information about primary information . It is information about information. We deal with collateral information intuitively in our daily lives but it is mostly ignored in our formal efforts at managing information. The two collateral facts about primary facts are:
• Effectivity-FT - When something did/will happen (and thus it and primary information about it became/becomes effective). Effectivity has past, present, and future possibilities. We can know what will be (postage or cost of living increases, for example) before it is effective. • Applicability-FT - When we obtain primary information (at which point we can apply it). Applicability only has past and present possibilities. We can't know what we know before we know it. Applicability is always simultaneous with obtaining primary information. It is "now." Effectivity is simultaneous with or subsequent to obtaining primary information. Effectivity sometimes reflects the same point in time as applicability, but often does not. It could be earlier, now, or later. Whether they reflect the same or different points in time, effectivity and applicability are always separate and distinct facts reflecting separate and distinct meanings.
We don't worry about not acknowledging a new friend's January birthday if we didn't meet them until February. Nor are we concerned if, having met them in November, we don't learn their birthday is in January until April. What was true (what and when something was/will be effective) and what we knew (when we found out about it) is critical to properly managing information.
Because we don't consciously understand it, conventional systems pay almost no attention to collateral information. When they do, we choose one over the other and often create considerable confusion over which we are paying attention to. The difference between Accrual and Cash accounting, for example, is essentially the difference between Effectivity and Applicability ... two separate perspectives on the same information. Choosing one and then disallowing the other is a decision to NOT manage information to its full potential. While either perspective may be sufficient to satisfy accounting rules, choosing only one sacrifices informational realities. Whether ignoring collateral information, choosing one but not the other, or confusing the two, the result is poor information management.
Derived-Secondary Information is derived from Primary and Collateral information. It is always time (effectivity, applicability) sensitive. The rules of derivation are intutivtely understood but poorly organized. In conventional terms secondary information is "output" (what we can know if we pay proper attention to what we do know).
Secondary information is consequences, not causes; shadow, not substance. It is often the focus of conventional systems leading to poor development and organization of primary and collateral information. The penalty is systems that manage results instead of providing the capability to get results. We build the car for only one trip. The only reason to focus on secondary information is to assure that the development of primary information is sufficient for ... but not limited to ... expected results.
If we get primary and collateral information right, all secondary information naturally follows. If we can't get desired secondary information from a system it is because we didn't get primary and collateral information right.
Information Dimensions |
 |
Understanding collateral information reveals that information functions in four dimensions; two definitional and two chronological.
Definitional Dimensions(Information at a point in time) |
Chronological Dimensions(Information over time) |
||
Occurrences |
Facts |
Effectivity |
Applicability |
Occurrences and facts roughly correspond to concepts of rows and columns. Occurrences and facts, however, are about how information-meaning applies in the real world while rows and columns are about organizing data-symbols with tools and media. Information topography identifes and organizes these two defintional dimensions by paying attention to the real world instead of to tools and media.
Effectivity and applicability facts establish chronological dimensions which in turn establish effectivity and applicability vectors through which we can derive considerable volumes of secondary information. Making effectivity and applicability as intrinsic to information bases as row and column are to databases, allows for secondary possibilities without having to plan them out for every circumstance in every would-be system. We build the car prior to and without having to know the trips for which it may be used.
The difference between conventional data base systems and fully dimensioned information bases is the difference between a snapshot and a movie. With data bases, considerable overhead effort is required to keep a system current ... to constantly modify the sanpshot to reflect what's happening "now." What the snapshot looked like last week or the ability to anticipate what it might look like next week, legitmate usages of well managed information, are impossible. These systems don't work with time or in (the flow of) time but against time. Business and users must continually react to a system's timing instead of the system reacting to their timing.
We use two dimensional databases to prepackage answers.
Four dimensional information bases allow us to
spontaneously respond to questions.
Full dimensionality eliminates the software and operations overhead needed to modify the snapshot. Yet, with it we can flash back (and sometimes flash forward) just like a movie that shows a different persepective often revealing what we know now about then that we didn't know then about then.
Information Vectors |
|
"Paths" followed through informational dimensions to derive secondary information from primary and collateral information. Information vectors reflect the conditions and parameter requirements for applying math to the real world.
| 1. | Immediate | Formulas and their (immediate) results derive new facts from existing facts. Formulas apply to individual occurrences (Sales-Price = Unit-Price x Quantity-Ordered) and to summaries (Average Salary = Salary-Total / Employee-Count). Note: Multiplication and division and their various forms (exponentiation) only occur within immediate vectors. |
|||
| 2. | Summary | Counting (how many occurrences) or algebraic summation (how much of something for many occurrences) at a point in time for a period of time. | |||
| 3. | Balance | Counting (how many occurrences) or algebraic summation (how much of something for many occurrences) at a point in time for all time. | |||
| 4. | Factor Distribution | Calculating how much is to be distributed to an/each occurrence (salary, payroll deduction). Factor distributions are often derived as a factor-share of a total (stock dividends) | |||
| 5. | Comparison | Deriving the difference between one occurence and another (taller, heavier, older) | |||
| 6. | Effectivity |
|
|||
| 7. | Applicability | ||||
| 8. | Time Distribution | Calculating how much is to be distributed to an/each occurrence over time (Teachers earn salary across nine months but often receive distribution across twelve months) | |||
| 9. | Comparison Summary | Calculating summarized comparisons over time (How many more widgets were ordered in March than in February). | |||
Difference vectors (5, 6, 7 & 9) compare two different things at a point in time or the same thing to itself over time.
Distribution Vectors (4 & 8) are the result of Events - A company or organization (ownself) doing something that derives and establishes additional information from existing information (makes the information explicit) . For all other vectors the derived information is not established (it is implicit, thus always derivable). |
|||||
Supercession |
|
Just because information is no longer current
does not mean it is no longer information.
In the real world information is superceded, not forgotten. In computer systems, however, data is routinely changed, ("forgetting" the old by replacing it with new), or deleted ("forgetting" the old and replacing it with nothing). Replacing instead of superceding data-symbols results in the wholesale destruction of information-meaning. Systems lose touch with the real world.
We replace or delete data because it is no longer current which is to say, it is not currently effective.
Women don't forget their maiden names just because they get married. When contacting the high school reunion committee they know which of their names was effective when they graduated. We would think it ridiculous if a woman didn't know or understand this information about herself, yet accept planned amnesia as standard practice within conventional systems.
Respecting the full dimensionality of information puts information management in touch with realities conventional systems deny. When new data is introduced into systems the informational reality is simply that the effectivity of newer information ends (supercedes) the effectivity of older information. As of last week the older information is still effective ... it is still information. When deleting data we end the effectivity of prior information but don't replace it. The essential information is that there is no superceding (effective) information, not that the prior information never existed.
Changing or deleting data is anti-information management.
(This discussion of data deletion is not about archiving information which is a technical issue of storage capacity, not a "meaningful" consideration. It is important when archiving data, however, not only to preserve meaning but to preserve information about the archived information.)
Delta Vectoring |
|
What did we know, when did we know it,
and what did we do about it?
Chronology dimensions allow us to manage information in systems the same way we intutitvely manage it without the interference of technology.
When we are told something:
• We know if we have been told before • If what we are now told is different from what we were previously told • If we acted on previous information that is now different, whether we now need to do something else
John's "time card" (activity context type) for the third biweekly pay period of 2008 (effectivity vector 2008-03) is for 80 hours. The time card is entered into the system twice. Even so the system properly pays John for only 80 hours. A week later, after John questions his paycheck, it is confirmed that John actually worked 92 hours. A new time card is submitted for the third pay-period (effectivity 2008-03) for 92 hours along with a time card for the fourth pay-period (effectivity 2008-04) for 80 hours. John's next paycheck correctly pays him for the 80 hours of the fourth pay-period and for the 12 hours of overtime for the third pay-period.
The important factors about this example
• Documenting John's activity was accomplished the exact same way everytime whether it was the first or last time or whether it was before or after he was paid. • John's pay came out right regardless of the order in which information was received or acted on.
Because information is properly delta vectored there is no need for operational (user) or programmatic (system) adjustments to what and how John is paid. This saves time using systems and costs in developing systems. Traditional exception adjustments are a form of delta vecorting based on how we use (and automate) pencil and paper instead of by what we accomplish with information.
Information Correlations |
|
Identified occurrences associate as part of a group (A man and a woman, John and Mary, are correlated as the group married couple.) John could be, say, a customer. Mary could be a customer. Or John and Mary Smith together can be a customer. John and Mary as a couple can also be members of a group identified as a church which can also be a customer. The hierarchy builds because correlations within the same context establish groups which can also be grouped and grouped again.
The difference between hierarchies and relationships is whether
the things related are in the same or different contexts.
Strategically, relationships and hierarchies are exactly the same thing, but we have allowed ourselves to be tactically confused by how we draw them on paper instead of by looking at what they really are. The long and loud debate between hierarchical and relational techniques is based on incomplete understanding of both of them.
Information requirements for an individual who is a customer and a group of individuals, a church, that is a customer, are different but the meaning, customer, is the same. This requires that the same sub-context ("row") can have different forms, a violation of conventional IT concepts. Not, however, a violation of reality. It is conventional wisdom that violates reality.
This is not to imply that a system with a church customer has to keep a full congregational roster. The requirement is to know who can commit the church as customer. If Mary is treasurer and Joan is president, they are the individuals within the church group customer separate from John and Mary as customer. Mary just happens to be a member of more than one group.
-