MIST Final Exam
Characteristics of self-organizing structures
People decide for themselves which tasks to work on Contributors can organize themselves and their work (ex: Linux)
Krug's first law of usability
"Don't make me think" When you look at a website, it should be self-sufficient
AIC Triad
"Heart of information security" • Access/Availability: ability for authorized parties to access data and systems when necessary • (Data) Integrity: maintaining and assuring the accuracy and reliability of the information and systems over its lifecycle • Confidentiality: property that information disclosed or otherwise made available to unauthorized individuals, entities, or processes
Triple constraints
"Scope, time, cost"
Rule-based systems vs. pattern recognition systems (development and timing)
Rule-based systems (symbolic approach) refer to how AI learns • Ex: an adult learns a language by memorizing explicit rules • These AI systems generated deeply underwhelming results • There are lots of rules, and there are lots of exceptions Pattern Recognition Systems • Ex: a child learns a language by being immersed in it and receiving feedback and corrections for mistakes • Learns tasks by experience, repetition, recognizing patterns, and receive feedback Currently, neither system has been able to achieve (AGI) or artificial general intelligence which can apply intelligence to a variety of unanticipated types of problems
Entity-Relationship model (entities, relationships, attributes)
*relational database model is the most popular set of standard rules for database organization 1. Data is stored in one or more tables corresponding to entities 2. Entities are connected through relationships 3. Tables consist of records which correspond to rows in the tables 4. Records store data on a single instance of an entity 5. A record consists of one or more fields 6. Fields are often referred to as attributes and correspond to columns of the tables
Red Dress Boutique - shark tank pitch
-$600,000 for a 5% equity state -Wanted a new website and back-end system for data analytics -raise capital by selling equity -price sold per share simple way to determine value -venture capitalists are opportunistic; want to sell later for money
4-step process that makes up "analytics"
1. Asking meaningful questions 2. Acquiring relevant data 3. Analyzing the data quantitatively 4. Presenting the results
3 major categories of skills required for data science
1. Hacking skills 2. Math and statistics knowledge 3. Substantive expertise
Problems arising from the non-hierarchical/messy crowd
1. It can be hard to find what you're looking for in a mess of uncontrolled information 2. Members can behave in hurtful ways (bad actors) 1. Ex: fake news Crowd is more unruly than the core
When and why outsiders are more effective than experts
1. Organizational dysfunction 2. The core is often mismatched for the kinds of challenges it faces, while the crowd, because it's so big, almost never is Example from the book: Researchers wanted to find a faster way to sequence the genomes of large numbers of human white blood cells. Karim Lakhani concluded that we should not look to experts when things get complex. Instead, look to the outsiders. • Tried to find a way to model white blood cells • The best algorithms took almost 5 hours with only 72% accuracy • Turned to the crowd for answers • Submissions from the crowd proved faster and more accurate
Two Phases of 2nd Machine Age
1. Phase 1 (mid-1990s): the takeover of routine tasks, such as processing payroll, welding car body parts together, and sending invoices to customers 2. Phase 2 (2010-present): when "science fiction technologies" began to appear in the world, like autonomous cars
Six-step UX design process
1. Research on users 2. Build user personas 3. Create user stories 4. Start creating wireframes and interaction prototypes 5. UI, Visual Design, and Delivery 6. Metrics Analysis: validate design
Information security best practices
1. Rotation of duties 2. Mandatory vacations 3. Split knowledge 4. Dual control (2+ people perform the same action) 5. Strict procedure for employee termination
The distinction between "bits" and "molecules" (properties of information goods); moving from atoms/molecules to bits
A bit is a digital unit while a molecule is a physical unit Bits are easier to reproduce at a low cost • Nonrival • excludable
Definition of business model
A business model describes the rationale of how an organization creates, delivers, and captures value
Definition of business process
A goal-directed sequence of tasks or activities that take a set of inputs and convert them into desired outputs
General advantages/disadvantages of platform-based business models
Advantages • Reduces barriers that keep people from transacting with each other • Can influence the flow of transactions to absorb more of the benefits of those transactions Disadvantages: chicken or the egg problem
Waterfall vs. agile software development methodologies
Agile methodology: • focuses on the incremental development process and timely delivery of working software • High emphasis on limiting project scope • Sets a minimum number of requirements and turns them into a working product Waterfall methodology: • SDLC tasks occur sequentially, with one activity starting only after the previous one has been completed
DEFINITION OF MIS
An organized integration of hardware and software technologies, data, processes, and human elements to produce timely, integrated, relevant, accurate, and useful information for decision-making purposes
Basic information about the venture capital industry
Basic concept: founders are trying to raise capital for their operations by selling equity in their business Venture capitalists are opportunists who want to sell their share of the business at a later date
Cryptocurrencies and "free, perfect, instant" properties of information goods
Bitcoin • Used digital signatures to verify when one person transacted bitcoins with another • Transactions were recorded in a ledger to avoid the "double-spending" problem • Not free, perfect, or instant because this would allow forgery to occur (bad actors could spend the same bitcoins over and over again **bitcoin needed to operate in a completely decentralized way, not under any control of the government or other institutions • Blockchain solved this problem as a way to moderate the ledger • Blockchain kept a historical record of all transactions
Role of humans in algorithmic development (proactive vs. reactive)
Computers have great difficulty gathering more or different data than what their builders and programmers allowed Reactive: using humans as monitors that watch over the decisions and actions of AI algorithms Proactive: using humans as a source to train supervised AI algorithms by identifying exceptions to rules
Definitions of the core vs the crowd
Core: dominant organizations, institutions, groups, and processes of the pre-internet era Crowd: new participants and practices enabled by the net and its attendant technologies
Machines and creative production
Creativity is no longer uniquely human; **machines have creative potential as well Computers can create art and generate music, but humans describe AI-generated art as "haunting"
Parts of BMC
Customer segments, value proposition, channels, customer relationships, cost structure, revenue streams, key partners, key resources, key activities
Examples of tensions in self-organizing structures (DAO, consolidation of power in bitcoin)
DAO: recent efforts to substitute a crowd for a company, decentralized autonomous organization First DAO • Open-source, distributed software for executing smart contract • No human or institutional layer, no CEO, board of directors, or employees • Software and software alone • Hacker turned the DAO into an ATM Bitcoin • Became controlled by a handful of people over a period of time, going against its original purpose • "Large concentration [of power] was undesirable since it could lead to disproportionate influence over the system's evolution, and the whole point was to avoid such influence by maintaining decentralization"
DANCE of the robots
Data: explosion in the volume generated Algorithms: deep learning Networks: essential to collect data for centralized processing Cloud: Environmental improvements in digital hardware (Moore's Law)
What is a database? DBMS?
Database: an organized collection of data Database management system: a software application that lets you create and work with a database Big data: 1. Volume: large amounts of data 2. Variety: data is messy and doesn't fit neatly into a database 3. Velocity: data arrives rapidly and continuously
Consequences of poor organizational coordination
Delays, excess inventory, lack of visibility
BPMN and basic notation
Emerging standard for specifying business process model The goal is to provide standard notation readily understandable by all business stakeholders Event: something that happens instantly in time (circle) task/activity:Represents a unit of work that has a duration (rectangle) Flow: Connects objects with each other and shows the order of elements in the process (arrowhead) Parallel gateways: Object that represents things that separate and/or recombine flows → diamond Exclusive Decision / Merge: Indicates locations within a business process where the sequence flow can take one or more alternative paths Only one of the paths can be taken Depicted by a diamond shape that may contain a marker that is shaped like an "X" Lanes: resource classes in the same organizational space and sharing common systems Pools: independent organizational entities
Ways to organize the crowd
Essential things to bring together the crowd 1. Openness: do not limit based off of experience 2. Noncredentialism: anybody can contribute, regardless of job titles, diplomas, credentials, experience 3. Verifiable and reversible contributions: easy to identify when software doesn't work, and easy to reject it and revert it when it doesn't 4. Clear outcomes 5. Self-organization: people and organizations decide for themselves which aspects to work on 6. Geeky Leadership: technically proficient leadership Linux was the prime example of this mentioned in the book. Other examples included Nupedia, which initially failed because it required creators to have credentials. Wikipedia was created after Nupedia failed, and allowed any users to contribute.
Re-engineering movement and business processes
First example: when businesses began to "electrify" (transition from the steam engine and power to electricity), many successful incumbent companies failed The re-engineering movement refers to how technology is changing modern businesses. In the context of business processes, the re-engineering movements have changed the way that businesses execute their most important processes. • The Enterprise System allowed businesses to conduct cross-functional processes with one large piece of software • The World Wide Web allowed these enterprise systems to extend reach to customers via their computers (and E-commerce allowed the transfer of money) "Reengineering the Corporation" is a book that was written that explained how businesses should think of themselves not as conducting tasks within departments, but instead as executing business processes
"Free, Perfect, Instant" and "Network Effects" as two kinds of disruptive economies
Free: free to make an additional copy of digitized goods Perfect: copies are every bit as good as their digital originals Instant: networks allow distribution of a free, perfect copy of information goods from one place to another virtually immediately Network effect: refers to a phenomenon by which a product or service gains additional value as more people use it • WhatsApp
Functions vs. processes
Functions: different functions responsible for different things Processes: cross-functional, relies on each individual group for execution
Relationship types
How entities are related 1. One-to-one: one entity has a relationship with only one instance of the other entity 2. One-to-many: when an instance of the first entity can have a relationship with one or more instances of the second entity, but instances of the second entity can be related to only one instance of the first 3. Many-to-many: when instances of each entity can be related to one or more instances of the other entity
Machine Learning?
Machine learning is the art and science of building software that can detect patterns and formulate winning strategies after being shown many examples • AlphaGO
Definition of a platform
Online environments that take advantage of the economics of free, perfect, and instant
The relationship between transaction cost economics and self-organizing -- why might companies be passe?
Overall, markets have higher costs than firms do • Higher costs of searching and discovering relevant prices, costs of negotiating and making decisions, contracts Companies are not shrinking, they're growing larger TCE • Markets usually have lower production costs (cost of making goods and services) • Hierarchies (firms) have lower coordination costs (costs associated with setting up the production and keeping it running smoothly) • As coordination costs go down, markets become more attractive (comparative disadvantage shrinks), implying that we should be seeing more use with markets and less with hierarchies However, firms are growing stronger and larger, going against the TCE
Difference between the two machine ages
Phase 2 technologies can perform nonroutine tasks and learn things on their own → People have also begun to carry these powerful, flexible computers with them at all times → Information is encoded as bits instead of molecules • Analog vs. digital • Print vs. E-Book
Primary and secondary keys
Primary key: an attribute that can have a unique value for every instance that you store in a table (social security #, student id #, employee id #) Secondary key: other defining characteristics of the attribute Foreign key: always has a primary key from another table
Relationship between revenue and financing
Revenue might not be enough at first to help business operations so a business might turn to a type of financing to get capital.
Organizational impacts of muting HiPPOs
Shifting to data-driven decision making and pulling power away from HiPPOs
social network theory
Social network theory is a way to collect and analyze data that represents a social network. The primary "unit" in this type of analysis is the individual node, and you can collect data that represents the various relationships between nodes ("ties"). In class, we saw an example of how data scientists collected social network data about passing networks on soccer teams and used a statistical tool to transform that data into an easily interpretable graph. For those of you that follow European soccer leagues, you noticed that Barcelona is an outlier on the graph--that represents their unique passing game, which is known as "tiki-taka." What's important here isn't the math or the statistical principles that are used to actually apply social network analysis. What's important is that it represents a different way to think about collecting and using data that opens up new types of analyses that are based on the links between nodes. Think: transform data to a plot to easily find outliers (which "nodes" have the most connections and are the most central)
Strengths and weaknesses of human intelligence
Strengths: Humans have more common sense than computers and can see distinct and unanticipated factors • Broken leg role • Look out the window: a quick "double-check" on the computer's decisions by humans Weaknesses: Humans tend to think that they are smarter than they are, and often have biases that affect system 1 and 2 thinking "Data-driven, system 2 decisions (such as those made by a computer) are better than a blend of system 1 and system 2 (the human brain)"
System 1 vs. System 2 thinking
System 1: "intuition", fast, automatic, evolutionarily ancient System 2: Slow, conscious, evolutionarily recent **System 1 is prone to biases Standard Partnership: refers to the division of labor between man and machines • Humans are responsible for exercising judgment and intuition in order to make smart decisions • Computers are responsible for the man and record-keeping **The Standard Partnership places too many emphases on human judgment • Relying solely on algorithms and formulas often yields better results • computers are the better decision-makers, based on the fact that there are many biases and malfunctions of system 1 and 2
Characteristics of jobs least likely to be affected by progress
Technological progress will not touch jobs/tasks that tap into social drives: compassion, pride, embarrassment, envy, justice, solidarity • Ex: soccer coach • Patients do not want to receive a healthcare diagnosis from a machine
SDLC (Systems Development Life Cycle)
The SDLC is inappropriate whenever specifications are hard to come by. This can happen if the problem is an ad-hoc problem. In this case, I am using the term "ad hoc" (which is a latin term and the connotation can be confusing if you're not familiar with it) to mean that there's a impromptu sort of problem that springs up without much planning or forethought that demands a solution immediately SDLC can be used for a new information system that needs to be built from scratch or one that needs to be redesigned/updated. Requirements specification is the key step that translates the business needs (from the functional units, e.g., marketing) to the developers themselves, who are building the software but may not have domain knowledge in the functional area.
The role of good management in the companies of the future and how to lead
The amount of managers is increasing worldwide. Why? • The world is complex and always changing, and managers have the ability to make problems go away before they show up • Social skills are more persuading and appealing to us than algorithms or statistical data • Humans are social beings and want to work together How to lead • Enforcing the egalitarianism of ideas: all ideas from any department are equally important • High level of transparency **Companies are NOT becoming passe
Failures of long-term prediction & how to be good
The world is too chaotic and unpredictable to rely on long-term predictions A shift is taking place away from the long-range forecast, long-term plans, and big bets, and toward constant short-term iteration, experimentation, and testing • Ex: casino Harrah's emphasis on machine-slot users over big time "whales" or betters, changing certain wording on a website to increase sales Foxes vs. Hedgehogs • The metaphor that describes foxes as better forecasters based on the fact that they consider multiple viewpoints • "A fox knows many things, but a hedgehog one important thing"
Definition of user experience (and contrast with user interface design)
User Interface describes graphic design
How humans use apps/sites
We don't make optimal choices because a) we're in a hurry, b) there's only a very low penalty/cost to "guessing" wrong, and c) weighing the options carefully to make an optimal choice may not even improve your chances of successfully navigating a site if it's poorly designed. Create a visual hierarchy on the page! Visual precedence: How our eyes scan content The "Z" pattern in designs without much text The "F" pattern in designs with more text Take advantage of conventions These are often obvious: A phrase in very large type is usually a headline that summarizes the story underneath it Break pages up into clearly defined areas Eye-tracking studies show that users decide quickly what parts are useful and then never look at the rest of it Make it obvious what is clickable Minimize noise
1st vs. 2nd order questions
What do we put on our shelves? Vs. How do we decide what to put on our shelves?
Types of work robots succeed at; where they may overtake human skill
Where work is dull, dirty, dangerous, or dear Robots succeed at • Making predictions based on information
Wireframes (where they fit in the UX design process what should be included, how they are used)
Wireframes: a visual guide that represents the skeletal framework of a website or application • Includes navigation, input components, key interface elements
Debt financing
is when you borrow money from a lender that you have to pay back with interest • Banks • Monthly payments must be made to these debt obligations
Equity Financing
is where you trade partial ownership of some of your business to investors in exchange for their capital • If your business fails, you don't have to pay back investments • Others have ownership of your business, meaning you give away decision-making power
Security functions and frameworks
security controls (with the main types being administrative, technical, and physical) have different functions and that in a company, all the different types of security controls they have make up the company's overall security framework Functions: • Deterrent: intended to discourage a potential attacker • Preventative: intended to avoid an incident from occurring • Corrective: fixes components or systems after an incident has occurred • Recovery: intended to bring the environment back to regular operations • Detective: helps identify and incident's activities and potentially an intruder • Compensating: controls that provide an alternative measure of control Frameworks • Made up of many entities, protection mechanisms, processes, and procedures that all work together and rely on each other to protect the company • Each organization has its own security model
silo effect
workers complete their tasks in their functional silos without regard to the consequences for the other components of the process
Security Controls
• Administrative controls: policies, standards, procedures, guidelines, personnel screening, training • Technical controls: authentication, firewalls, biometrics • Physical controls: locks, monitoring, mantraps, environmental controls
Cost Structure
• All costs incurred to operate a business model • Ex: "no-frills" airlines is an example of a low-cost structure
Characteristics of a project
• Definite beginning and end date • Produces a deliverable or outcome (unique product, service, or result). Creates value. • Often limited by triple constraints
Customer Segments
• How a company groups customers into distinct segments with common needs, behaviors, or other attributes
Channels
• How the company communicates with and reaches the customer to deliver a value proposition • Web sales, owned stores, partner stores
Project planning documents
• Project charter • Work breakdown structure • Project schedule
Scope creep, scope leap, scope grope
• Scope creep: the most common term and problem. Adding new features (not part of the original scope) incrementally over the course of the project • Scope leap: drastic increases in the projects' scope • Scope grope: the project team can't define the scope of the project
Why we use the BMC
• Shows the logic behind how a business makes money • Shared language
Key Partners
• Strategic alliances to optimize the business model, reduce risk, or acquire resources • **it is illogical for a company to own all resources or perform every activity by itself
Revenue Streams
• The cash that a company generates from each customer segment • One time transactions or recurring transactions from ongoing payments • Asset sales, usage fees, subscription fees
Key Activities
• The most important actions a company must take to operate successfully
Key Resources
• The most important assets required to make the model work • Physical, financial, intellectual, human
Value Proposition
• The reason why customers turn to one company over another • How the business solves a customer problem or satisfies a need • The product/service offered • Ex: newness, performance, customization, design, brand/status, price, convenience
Customer Relationships
• Types of relationships that a company establishes with specific segments • Personal to automated • Tailored to each segment