Note: Oneslate is experimental software. Implementation of the self-service cloud or other services are recommended on a trial basis to determine what the Oneslate application can do for projects, individuals, teams, departments, or organizations.
The Oneslate application is designed to generally enable improved goal achievement through error reduction in decision making. Its focus is on decisions which cannot be automated using traditional business intelligence tools.
Oneslate's development was provoked by the open opportunity to address a number of common error sources which had been identified as existing, but for which no realistic solutions had been made available.
A number of related problems whose individual solutions seem elusive may be partially addressable together in a software application to be used in suitable situations. Oneslate may be used to attempt the following:
Help information overload to be overcome.
Help anthropic bias (observation selection effects) to be overcome.
Help common cognitive biases to be identified and mitigated.
Help common sense to be found in tricky scenarios.
Help knowledge leakage to be reduced.
Help the tapping of cognitive surpluses.
Help establish stronger corporate sustainability.
What capabilities does Oneslate incorporate? The core functionality is as follows:
Add and search ideas.
Encode and navigate the structure of ideas.
Relate ideas to many supports and many conclusions.
Include the same exact idea in many contexts.
Associate any ideas with any other ones in the same knowledge base.
Discover all contexts in which an idea occurs.
Heatmap and update idea ratings on a 5-bin scale.
I know I know this.
I think I know this.
I do not know this.
I think I know this is not.
I know this is not.
Unobtrusively check for and log cognitive biases contextually.
"Personal knowledge base" (Davies, 2011) and "knowledge workshop" (Engelbart, 1972) are perhaps the most fitting terms from the literature to describe a system with the above functionality. While a small number of other available systems come close to accommodating the required functionality, they are too complicated for general use and require either facilitation or an uncommon inclination to learn to use and then apply them.
Some additional capabilties of the Oneslate system include the following functionality:
Work with knowledge graphs in the comfort of mainly dealing with tree sections.
Add sources to ideas; autosuggest sources to reuse them in multiple places.
Maintain private compartmentalized knowledge bases with guest access provisioning.
Embed linked media and document links.
Use any modern web browser to interface with the database driven application hosted in the cloud or locally in a virtual machine.
There are some notable comments from the literature upon which the Oneslate system can be reflected, as follows:
Aumann, R. (1976) suggests that rational Bayesians with common knowledge should not disagree. Oneslates provides a framework for establishing common knowledge and for enabling an approach of updating evaluations based on the best available information.
Aurisicchio, M. and Bracewell, R. (2013) describe the opportunity and challenges for design rationale capture systems. The capture of design rationale is an area of potential applicability for the Oneslate system.
Bostrom, N. (2002) identifies that people intrinsically experience the effects of observation selection. Oneslate allows the melding of observations.
Davies, M. (2012) finds that Computer Aided Argument Mapping (CAAM) may be more effective than traditional techniques at instilling critical thinking skills. Oneslate supports CAAM, and may be used as a tool for applied critical thinking.
Davies, S. (2011) identifies characteristics needed by a system to achieve the intended effects of a memex (Bush, 1945) type system. As described in the blog post Oneslate as a personal knowledge base, Oneslate meets a number of these criteria, and it is built in such a manner so as to be expandable to meet more of them.
Engelbart, D. (1962) describes a conceptual framework for augmenting human intellect. Oneslate shares a similar approach to a similar task.
Grudin, J. (1985) identifies reasons that Computer Supported Collaborative Work applications fail. Oneslate is at a stage where many of the pitfalls may be avoidable.
Grudin, J. (2008) shares lack of emphasis on ease of use as the primary reason that the focus of Engelbart's work was defunded, resulting in the lack of realization of much of software's capabilities given digital computing. Oneslate's design has been kept minimalistic in an attempt to facilitate complex tasks without being overbearing.
Grudin, J. (2012) provides a good background on the evolution of human-computer interaction. Oneslate's vision is along the lines of those had by Otlet, Bush, Engelbart, and Nelson, though its approach has been to start with a minimal implementation deployable in the enterprise.
Ioannidis, J. (2005) writes on the majority of research findings' being false. Collaboration in research is an area where Oneslate may be highly applicable.
Kunz, W. and Rittel, H. (1970) developed an Issue Based Information System (IBIS) for tackling so-called wicked problems. Oneslate supports the IBIS notation and may be used to address this class of problems.
Minto, B. (1987) described a straightforward system for making thought and writing clear, logical, and natural, for maximum effectiveness in business communication scenarios. Oneslate facilitates the construction, updating, sharing, interconnecting, evaluation, and referencing of infinite logic trees.
Munger, C. (1995) spoke on the psychology of human misjudgement, calling out specific mental shortcuts as irrational thinking errors and describing a number of antidotes. Since the errors are often subconscious, Oneslate has a built-in mechanism for raising a list of many biases including those mentioned by Munger for direct reflection and contextual notation.
Simon, H. (1960) describes a four-phase management decision process including intelligence, design, choice, and review, noting that automated decision making is not suitable for "hopelessly qualitative" problem areas. Oneslate is designed to organically enhance each of these steps, and it can be considered to compliment traditional business intelligence tools since it is particularly suitable for the qualitative problem domain.
Tversky, A. and Kahneman, D. (1974) discuss a class of cognitive biases resulting in judgement errors due to frequently but not universally useful shortcuts, as well as the reason why such biases usually go undetected. Oneslate provides a mechanism for reviewing and contextually noting whether a number of such heuristic-based, as well as other types of, biases are acting, so that their action may be mitigated where appropriate.
Yule, D. and Blustein, J. (2013) note that web browsers have only recently become capable enough for usable implementations of a memex or NLS/Augment type system. Oneslate's development was in part instigated by the ability of modern browsers to afford these types of tasks.
Decisions and goals
How is Oneslate designed to provide for improved goal achievement? It assumes that firms have goals, which may be achieveable through decisions necessarily made by executives, managers, or professionals. Where tools can be applied help to reduce error in decisions, goal achievement should be improvable.
According to Simon (1960), the decision process includes the following phases:
Intelligence (searching the environment for decision making conditions),
Design (inventing, developing, and analyzing possible actions),
Choice (selecting from the available actions), and
Review (assessing prior choices).
Oneslate may be able to enhance each of these decision phases at the appropriate moments.
Errors and bias mitigation
Humans make errors as the result of a number of conditions (resource limitations, tool limitations, observation selection effects, sensory limitations, and other sources of cognitive bias). Much suboptimal decision-making comes from inadequate alternative consideration. A major cause of inadequate alternative consideration discussed by Bostrom (2002) can be "observation selection effects" or "anthropic bias", which might be called simply insufficient sampling or impossible sampling requirements. The intermediary results of suboptimal decision processes can be, compared to an ideal case, a lack of knowledge access or lack of learning from experience, repetitive suboptimal decision-making, or the metaphoric reinventing of the wheel. Hard to avoid intermediate errors can result in inefficiencies in the achievement of higher level goals. Some specific problem sources and how they are addressed in the Oneslate application are outlined below:
Error sources and mitigation methods implemented in Oneslate
Error mitigation method
Limited alternative consideration (observation selection; imperfect sampling; "anthropic bias") is a major cause of suboptimal decision making.
Have a knowledge repository where alternatives can be efficiently built over time as deemed important and discovered in all relevant contexts.
People have limited working memories, generally able to follow five to nine things at once (Miller, 1956).
Enable creation of visible, non-linear mental model representations of important information on an ongoing basis as needed that can be accessed in a highly focused manner to present only and all available related information in relevant contexts.
Cognitive biases are rampant in people.
Permit the unobtrusive identification and contextual tracking of identified cognitive biases (e.g., in situations of changing validity assessments).
Knowledge leakage is a real problem for organizations.
Enable efficient capture of structured knowledge.
critical thinking skills
College graduates are under-equipped in critical thinking skills.
Deploy tools supportive of a process (e.g., Computer Aided Argument Mapping) shown to potentially improve critical thinking skills better than traditional methods.
Office workers exhibit workplace dissatisfaction at alarmingly high rates in the US (Gallup, 2013).
Provide a suitable outlet for otherwise trapped insights and expression.
There is a significant cognitive surplus in people.
Have a space that physically enables contribution through focusing on new and growing out related concepts.
knowledge dissemination and handling ability
It is not feasible for much knowledge to be very widely disseminated or handled.
Utilize a usable "real hypertext" knowledge base tool that matches knowledge input and display scenarios to only those relevant.
methods for handling wicked problems
Some complex problems are not understood until solution formulation and require unique, one-shot solutions that are neither right or wrong, and have no given alternatives.
Utilized a tool which supports a notation identified as appropriate for wicked problems, such as Issue Based Information Systems, and which is usable in non-facilitated scenarios.
consensus awareness and utility
Consensuses can be highly undervalued, next to impossible to ascertain, and dynamic.
Use a knowledge base system in which everything is a rolling 5-bin poll with instantaneous and trending consensus breakdown displays.
Bayesian updating is not widely habitual.
Provide context for Bayesian updating with an appropriate scaffold.
How it works
The Oneslate application functions by offering a generalized interface for mental model capture, review, and updating. Its idea space visualization capabilities are intended to offer the flexibility and simplicity required for utility. With the right tool, you can efficiently interact directly with the idea structures that matter to the decision-related tasks at hand, in a repository serving as a living knowledge base.
In Oneslate, individual ideas can be pieces of text, or embedded linked media or documents. In Oneslate, ideas related to other ideas can be connected as causal supports or conclusions, and the same idea can occur in many places with all of its connected ideas discoverable. In Oneslate, the updateable assessed validities of ideas can be encoded as a color or intensity on the representations of those ideas. And in Oneslate, rating changes prompt an optional bias survey.
How can you tell if a new idea space visualization tool is worthwhile in a situation? Determine the metric that reflects the goal that your function is helping to achieve. On a trial basis, deploy the new idea space visualization tool, and see what result is reflected in your metric of choice.
A number of people have produced texts providing background information relevant to the Oneslate application and its functionality. The following texts are referenced above, and most of them are available online:
Aumann, Robert J. "Agreeing to disagree." The annals of statistics (1976): 1236-1239. [scholar]
Aurisicchio, Marco, and Robert Bracewell. "Capturing an integrated design information space with a diagram-based approach." Journal of Engineering Design 24.6 (2013): 397-428. [scholar]
Bostrom, Nick. Anthropic bias: Observation selection effects in science and philosophy. Routledge, 2013. [scholar]
Bush, Vannevar. "As we may think." Atlantic Monthly (July 1945). [scholar]
Davies, Martin. "Computer-Aided Mapping and the Teaching of Critical Thinking." Inquiry: Critical Thinking Across the Disciplines 27.2 (2012): 15-30. [academia]
Davies, Stephen. "Still building the memex." Communications of the ACM 54.2 (2011): 80-88. [scholar]
Engelbart, Douglas C. "Augmenting human intellect: A Conceptual Framework." SRI, 1962. [scholar]
Engelbart, Douglas. "The augmented knowledge workshop." A history of personal workstations. ACM, 1988. [scholar]
Gallup. "State of the American Workplace: Employee Engagement Insights for U.S. Business Leaders". Report, (2013) [online]
Grudin, Jonathan. "Why CSCW applications fail: problems in the design and evaluationof organizational interfaces." Proceedings of the 1988 ACM conference on Computer-supported cooperative work. ACM, 1988. [scholar]
Grudin, Jonathan. "TIMELINES Why Engelbart wasn't given the keys to Fort Knox: revisiting three HCI landmarks." interactions 15.5 (2008): 65-67. [authorized link]
Grudin, Jonathan. "A moving target: The evolution of human-computer interaction." The Human-Computer Interaction Handbook--Fundamentals, Evolving Technologies, and Emerging Applications, 2nd ed. Erlbaum (2008): 1-24. [scholar]
Ioannidis, John PA. "Why most published research findings are false." PLoS medicine 2.8 (2005): e124. [scholar]
Kunz, Werner, and Horst WJ Rittel. Issues as elements of information systems. Vol. 131. Berkeley, California: Institute of Urban and Regional Development, University of California, 1970. [scholar]
Miller, George A. "The magical number seven, plus or minus two: some limits on our capacity for processing information." Psychological review 63.2 (1956): 81. [scholar]