The Spreadsheet and Cellular Autonomy
“Finite players play within boundaries; infinite players play with boundaries.” — James Carse
Cellular autonomy, as it relates to spreadsheets, is the notion that the cell is the most fundamental unit of a spreadsheet rather than the sheet. The idea is that each individual cell should be able to exist as an independent entity, with or without the construct of the X-Y. Thus, the concept of “spreadcells” emerges.
The Spreadsheet
The all-mighty spreadsheet. While electronic spreadsheet technology is mainly applied in business computing, Microsoft Excel has become as common a household name as iPhone, Keurig machines, and Netflix. Beloved to some and vilified by others, regardless of affinity, the spreadsheet is applied daily in practical applications worldwide by billions of people, helping to power businesses, drive innovation, and track trillions of dollars globally. The electronic spreadsheet is one of the most under-celebrated achievements of the modern information era, with numerous use cases. Where would we be without this contribution to civilization?
The Power of Spreadsheets and Creative Genesis
The fundamental centrifuge of the spreadsheet’s heroic power is the sheet itself, cast as both victim and villain. Wielding the power of the Cartesian X-Y coordinate system as its sword, it provides the electronic relationship between geometry and algebra. Voila, the table. Rows, columns, on an X-Y plane. One of the most influential and transformative structures the planet has ever witnessed.
We can thank the French Enlightenment thinker Renee Descartes for the X-Y coordinate system principle, developed in 1637. Viva la France. However, it was the French Canadians who were the first documented creators of electronic spreadsheet technology, most notably Rene K Pardo and Remy Landau. The French lineage, and their affinity for geometric and algebraic unification, developed LANPAR (US patent 4398249, Filed Aug 12, 1970.) LANPAR was an application used for budgets and forecasts on the FORTRAN mainframe at Bell Labs in Canada. [1] A Mari Usque Ad Mare!
Furthermore, more famously, Dan Bricklin, independent of Bell Labs, came up with the idea of an electronic spreadsheet while at Harvard Business School in 1979. With this notion, he developed the VisiCalc application, which became the template for what we use today, Microsoft Excel and Google Sheets. His electronic spreadsheet version was a significant improvement over LANPAR, as it was not limited to a mainframe. It was developed on the personal computer platform, making it affordable to deploy on a mass scale. [2] By all accounts, it seems as if Dan Bricklin developed VisiCalc independent from knowledge of LANPAR, and was unaware of the work done at Bell Labs, and thus, he also gets credit as a first creator.
The Future of the Spreadsheet?
As we sit here today celebrating the past accomplishments of extraordinary individuals, what about the future? In my 30-year career as a technologist, when I gaze upon the volume of books on my bookshelf, full of mostly old, antiquated technologies that are now redundant, I reflect on the notion that the spreadsheet is still here, vibrant, and well. While the aesthetics and brands have changed, the essential technology is still the same. A technology that is taken for granted, no different than a sidewalk in New York City.
Today is the year 2023. The current spreadsheet technology we all know in love is over 50 years old. Sure, the surrounding features and interfaces have improved dramatically, such as navigation, scalability, performance, cloud enablement, collaboration, security, and usability. However, at its core, we have a hybrid fusion of 1600 and 1970 technology. While the IT developer herds are chasing the cloud, blockchain, no-SQL storage, AI, Web 3, 4, and 5 (all viable), what about some introspection? What about looking inside one of the fundamental use cases of computing and asking: how can we improve it?
The X-Y Weakness
I propose that the current electronic spreadsheet’s fundamental shortcoming is the reliance on the 1600s Cartesian coordinate system. Why are we always limited to an X-Y constraint? While X-Y is an everyday use case, in the advanced communication of ideas, it often does not reflect the true nature of expressions. Patterns of data can be tabular, such as the X-Y, but they can also be hierarchical. Furthermore, they can be sequential, linked, embedded, or even free-form.
How can we break free from the rigid construct of the Cartesian coordinate system while still maintaining the capabilities we all know, love, and expect from spreadsheet technology? Capabilities such as point and click to change a value, instant automatic recalculation based on formulas, numeric and text formatting, status and formula bars, absolute and relative references, labels and formulas, etc.,? All the qualities that make a spreadsheet great.
A spreadsheet is just an arrangement of cells — a mosaic. The X-Y construct is simply a method of cell organization. Like a modern society where individuals are at the lowest level of intersectionality, is the cell also not the lowest level? The sovereignty of the individual reigns supreme in Western morality. What about the sovereignty of the cell? Should we not allow cells to be arranged from the bottoms-up organically rather than a top-down forced iron construct? This line of reasoning springs forth the notion of “cellular autonomy.” With the cell as the nucleus rather than the sheet, other data arrangements can spring forth naturally while maintaining calculation capabilities.
Spreadsheet Cellular Autonomy
Spreadsheet Cellular Autonomy suggests that the cell is the most fundamental unit of a spreadsheet rather than the sheet. The concept is that the cell should be able to exist as an independent entity, with or without the construct of the X-Y.
One of the definitions of “cellular” from the Google Oxford Languages dictionary is:
“consisting of small compartments or rooms.”
The word “autonomy” from the same said dictionary is:
“the right or condition of self-government.”
“freedom from external control or influence; independence.”
Thus, using the standard definition of the terms, the idea superimposes into the domain of spreadsheets as such: each cell as a small independent self-governing compartment. The operative words here are “independent” and “self-governing.”
Organic Systems and Spreadsheet Design
The concept of cells as autonomous is standard in biological systems. All things alive are made of cells. Cells are known as “the building blocks of life”[3] and are the most fundamental unit in living organisms. The sophistication of living organisms can vary. In the simplest of forms, they can consist of a single cell. Take, for example, yeast. More advanced forms are multicellular, such as dogs, cats, lamas, sharks, and people. Biology is just an advanced system.
Is an electronic spreadsheet not an advanced system? Since nature has mastered this process, why not take a page from the established playbook — one that has been pressure tested for ages of time across millions of species and amongst billions of people?
Biological systems present the following hierarchy: cells, tissues, organs, organ systems, organisms, populations, communities, ecosystems, and biosphere.[4] How does this hierarchy of living things map to the paradigm of cellular spreadsheet autonomy?
Here I present a map of the bottom four nodes as such:
- Cells = Cells
- Tissues = Sheets
- Organs = Workbooks
- Organ systems = Files Systems
In the current spreadsheet paradigm, the base of existence is the sheet. Workbooks are a collection of sheets. Sheets are composed of unbendable X-Y arrangements. Thus, in this classic mold, the cell is not autonomous. Imagine a biological system with the tissue as the base level while ignoring the cellular level. What kind of life form would materialize? This state is the condition of modern electronic spreadsheets — tissues without cells. Paper without pulp.
To move from the abacus to the calculator, from the calculator to the modeler, and from the modeler to the next stage of expression of ideas, the construct of the sheet needs to be adjusted. One that more closely mirrors organic systems. The cell needs dominion — it needs to be treated as an independent entity. From there, the different cellular arrangements can manifest themselves naturally with more sophistication and meaning.
Practical Implementation of a Cellular-Based Spreadsheet
Within a cellular spreadsheet, the cell is a base class that is instantiated with its own unique properties, methods, events, and interfaces.
Properties of a single cell would include considerations such as shape (circle, rectangle, triangle, etc.), fill (color, pattern, transparent, etc.), border (style, visible, color, etc.), and content (number, text, etc.). The interfaces would include the input and output of text and numbers, control menus, status bars, drag-and-drop management, resizing, and point-and-click value changing. Methods would encapsulate the data storage mechanisms, capabilities such as math and word processing, updates-adds-deletes, copy-paste apparatuses, push notifications, and integration to external data stores, such as blockchain and RDMS.
Thus, a workbook could contain a single-cell structure with all of these capabilities.
In the cellular autonomous world, the sheet is a collection of independent, self-governing cellular objects arranged in patterns. A sheet can possess at least six distinct arrangements. These are:
- Single Cell
- Tabular
- Hierarchical
- Embedded
- Flow
- Loose
The following diagram illustrates the six cellular arrangement patterns and some of the associated subcategories.
In the remaining sections, I will briefly overview each sheet type. Note: each of these following segments is just a basic summarization. More will be written expounding upon these patterns.
Sheet Arrangement #1 — Single Cell
The term spreadsheet should be reconsidered to “spreadcell.” Cells are the building blocks of sheets. This arrangement is what it professes to be, a single block. Imagine if I opened a spreadsheet, and a solitary cell was in place rather than a seemingly endless X-Y coordinate plane. The use case would be analogous to a calculator display. The contemporary basic calculator has a single cell panel. A single cell function within a calculator would be a+b=c to get a quick answer.
When opening the modern spreadsheet, users are presented with a grid of cells. While not endless (row and cell limitations are tool-dependent), most cells presented go unused. This seemingly limitless array comes at a cost: memory, CPU, screen real estate, and delayed opening, the Achilles heel of quick, just-in-time, ad-hoc calculations. Have you ever needed to do a rapid single-cell math calc in a hurry and tried to open Excel to accomplish it? The splash screen seemingly takes forever. Arg! How does one gain single-cell focus instantly? On the other hand, an autonomous single-cell structure, as a default, requires less memory and CPU and can fire up like lightning.
The single-cell mantra is taking root in other domains of modern computing. Take, for instance, blockchain. Each transaction is represented as a single cell block.
Sheet Arrangement #2 — Tabular
Tabular is the classic spreadsheet state. The X-Y coordinate plane in various degrees of sophistication. Tabular could be as simple as a list (row-based stacking), an array(column-based stacking), or a full-scale table. The table is both durable and detrimental, a strength and sickness, an enhancer and inhibitor.
The limitations tables present are not novel. They have been recognized as a barrier in other realms of computing — for example, RDMS systems. The requirement to store nonstructured data and the manifestation of Hadoop and NoSQL database technologies indicate such limitations.
One interesting characteristic that should be emphasized here is once the rigid X-Y plane philosophy is abandoned, a cell is not limited to rectangles, as in the current spreadsheet paradigm. With an autonomous cellular spreadsheet, if we so desired, we could have a tabular composition comprised of circular cells.
This example is just one of an infinite amount of visual expressions that can emerge.
Sheet Arrangement #3 — Hierarchical
A hierarchy is a tree structure that includes a trunk, branches, subbranches, and leaves. Trees can represent an organizational chart, flow diagram, or even advanced workflow that sub-branch. The nature of the hierarchical formation is fundamentally different from the tabular one. They are both structured data but have different behavioral patterns that rule their existence.
An excellent practical business use case of a hierarchical structure is a financial chart of accounts. There is a base level of accounts, a trial balance: the leaves. The leaves are grouped and summed based on branch nodes, which ultimately roll up to net income. A structure like this is expandable and collapsable based on the perspective the viewer is seeking and the desired level of granularity.
In the current spreadsheet paradigm, how is the hierarchy represented? In Excel, for instance, a group/ungroup feature exists that expands and collapses the grid, but it is an artificial means of forcing the grid to conform. What if cells could be arranged in true hierarchical nature? Not a grid, but a tree. One with expandable and collapsable nodes, automatic aggregation and disaggregation capabilities, and even data-spreading? These mechanisms exist today but only within the confines of expensive and advanced performance management systems. What if there was a desktop spreadsheet that could do this? These ideas and more can all be achieved within an autonomous cellular structure.
For the cell to achieve the capacity to meet these criteria, the notion of connectors necessitates mention. Connectors can be direct visual constructs or indirectly inferred. The connectors configuration is imperative to bring together diverse cellular arrangements. In the case of a hierarchy, nodes are presented as a visual tree. The connector is used as a method to display the relationship, as depicted in the following image:
Sheet Arrangement #4 — Embedded
An embedded arrangement is a cell within a cell. What is an example of this arrangement? What about a page in a word-processing document? When you look at a page, does this not fit the definition of a cell? Is a page not a large self-governing compartment? The goal of the word processor is to fill the box with text, but what if, within the writing, we could formulate and calculate numbers? If I change one number in the calculation, the formula totals automatically update within the sentence, just as a traditional spreadsheet would.
Take, for example, the following sentence:
“Expenses for Q1 are projected to be 4.3 million. Revenue is currently projected at 5.5. This forecast has our net income targeted at 1.2 million.”
This sentence expresses information. It is a blend of words and numbers. Embedded within the sentence is a calculation. 5.5–4.3 = 1.2. After this sentence is written, the expense forecast comes in slightly lower than expected, 4.1 million, and may need revision. By changing the 4.3 in the sentence to 4.1, in an embedded cellular model, the projected total of 1.2 would automatically update to the correct total of 1.4, the same way a spreadsheet would. In this paradigm, the author can focus on the quality of the writing without worrying about rerunning math calculations. The integrity of the numbers would be higher and more consistent.
It would require embedded cells within a central cell to make this scenario possible. Put differently, a sub-cellular structure. The primary cell is the word processing sheet, and the embedded cell is the calculation.
Sheet Arrangement #5 — Flow
Any cell bound by a linear, chronological sequence is a flow arrangement.
Using blockchain as a use case, is a blockchain not a sequential linear flow? An assembly of transactions strung together as a chain? Cannot spreadsheets be a chain of blocks also?
Using another example, in a 10-page PDF document, are these ten pages not just a chronological flow of rectangles? Are these rectangles not just large cells? They are sequentially bound in a linear workflow. A PDF is, in essence, a large flow arrangement of cells. Thereby, word-processing documents are a flow sheet.
What about successive business process maps? Again, flow sheets.
Sheet Arrangement #6 — Loose
A loose cellular structure is a complete freestyle assembly of cells. The loose arrangement could still have calculable references between and among its members, but the designer has adhoc control of the cell organization. Due to its utterly unrestricted nature, this arrangement may be the most potent and boundless aspect of autonomous cellular sheets. The end-user’s imagination is the only limit.
In Closing
While the development of such a spreadsheet has not been fully realized, I am optimistic it is just a matter of time. Prototypes are underway to illustrate the power of these ideas. In the meantime, I will continue to use this forum to frame the core philosophies and thought processes that will help drive the future. If it is not digitalized, it does not exist. This blog is the womb.
As Alan Cooper, the “Father of Visual Basic,” once said:
“How do I design if not with prototyping? An excellent question. The short answer is ‘on paper.’”
References
[1] Rene Pardo & Remy Landau Introduce LANPAR, the First Commercial Electronic Spreadsheet : History of Information
[2] Was VisiCalc the “first” spreadsheet? (bricklin.com)
[3] https://byjus.com/biology/cells/
[4] 1.8: Themes and Concepts of Biology — Levels of Organization of Living Things — Biology LibreTexts
About the Author
Scott Gehring has over 30 years of experience in global enterprise information systems and holds several patents for his work in varying industries. As a pioneer in the field of analytics, he has been an influential industry leader in defining best practices around system design, implementation, integration, and operations. Scott has built hundreds of solutions for companies ranging from small-mid-business to large-scale enterprise organizations, helping to drive process improvement, tighten the link between business and IT, and provide the latest innovations in information technology.
www.scott-gehring.com
www.linkedin.com/in/scott-gehring/
Scott Gehring — Medium
Technology Whiteboard
