Discuss: Give Us Your Data To Save the World

The motion for discussion is...
"Give Us Your Data To Save the World! discuss"
We need a process to make both public and private data transparent and available for researchers and firms to use.
The motions are to take positions for or against legislating for organisations to have access to public and private data to "save the world."

Goal
Data mining is possible, who benefits, who loses? What are the benefits versus the risks of bringing different data sets together. To assess the value versus the cost of making public and private data available to all.

Instructions
Two groups (red-against, blue-for) will present an analysis and then debate the merits of their different positions, involving the whole class in the closing stages.
The readings given for initial investigation are listed below. We strongly encourage the use of additional secondary sources and providing examples as evidence supporting arguments.

Readings

• Hey, T. "The Next Scientific Revolution: How data mashups can help save the world," Harvard Business Review), November 2010 2010, pp 56-63.
• Give us back our crown jewels (www.guardian.co.uk) Thursday 9 March 2006

Exercise: Using spreadsheet 'Goal Seek' for break-even analysis

Goal seek is a spreadsheet function that attempts to match the function value of a 'goal' cell to a 'target' value by changing the value of an 'input' cell. Simply put, goal seek tries to solve and identify a function 'output' in terms of a single input variable.

Break-even Anaysis
Break-even analysis attempts to solve or identify the point at which a total cost (TC) function equals (i.e. cancels out) a total revenue (TR) function. The usual demonstration is to model TC and TR as simple line equations (functions of the form y = m*x + c). By plotting values (cost and revenue) against quantity, where quantity is the number of units produced or service delivered. Naturally, at some level of production (quantity) TR must exceed TC for the business to be financially viable.

The following examples illustrate the use of goal-seek for break-even analysis.

1. In the first case identify the production level (quantity) that needs to be sold for the business to achieve break-even in year 1.

2. The business model in this case is loss making. What unit price should be charged over the next period to break-even?

3. Goal seek can be applied to seek a solution output for any single input variable. In this case identify a price per unit in a scenario where the break-even point is 200.

4. Finally, using the model from question 1 identify the break-even point graphically by plotting a range of values for TC and TR against quantity, i.e. graph the range of net profit/loss outcomes for a range of production levels (quantities). This formulation does not use 'goal seek'.

Playing with DSS, Modelling and AI

The cool stuff: 
Havok for AI, destruction, cloth, behaviour, physics and more.

Possible Medical examples:
WebMD (link)
YourDiagnosis (link)
Differential Diagnosis (link)

Identify the Decision Support Systems

Gerd Ludwig's Long Shadow Project

Preparing for Flight (Oct 29, 2010)

Artificial Intelligence around the corner?

The media has celebrated IBM's 'Watson' program after its recent win in popular US TV game show Jeopardy. But should we temper this success with a more pragmatic view of the so called 'intelligence' of what is after all, simply a computer program?

...and why isn't AI or artificial intelligence ready for the real world, if indeed it is possible in the first place? Watson doesn't know what the questions mean. Watson doesn't know if it is right or wrong. Watson can't explain why its answers might be right or wrong. Watson doesn't care about winning or losing, or feel responsible for its success or failure. Watson doesn't think or come up with a new idea. Watson can't ask original meaningful questions that matter to it. Watson doesn't care.

Articles

• IBM team that developed Watson (online at reddit.com)
• IBM Press kit "Watson to challenge humans at Jeopardy" (ibm.com)
• Irish Times (online) 19th Feb
• Watson thick as two planks: critique/comments on CW (www.computerworld.com)

Which is the longest London Tube Line?

Classic London Tube Map

vs a geographic depiction

A Geographic Map from Maptube.org

And for fun have a look at animals on the underground (link)

The future of Artificial Intelligence!

AI vs AI. Unintentionally funny perhaps. But more seriously, is this the dawning of a new age of intelligent computing?

Mobile data: where for, what for?

Matt Biddulph from Nokia offers a thoughtful overview of the potential for aggregate data. Mostly it confirms what we already know, the data is dead unless we ask interesting questions. What he shows us however is our growing ability to present data in interesting and engaging ways. What is less clear is how these things add value; are these representations merely pretty pictures or do they say something deep about our cities and behaviour?

'Cognitive Cities: City analytics' on Slideshare homepage. http://slidesha.re/gUIAFX

Read the questions at the end... 'privacy', 'who benefits', etc.

Also see Oliver O'Brien's Suprageography blog and pages.


Image credit: screen shot of Oliver O'Brien's London Bike Share Map (2/3/2011)

Global Pulse, who benefits?

Who benefits from data Philanthropy? Individuals, organisations and corporations have long dreamed of the free availability of social data. We now imagine that Orwellian vision to be so much closer by accessing messages in social networks, text messages, blogs, micro-blogs and other emerging social exchanges. Worthy causes break down our reluctance to give up or give over something we may have never appreciated fully. But who benefits?

“something is happening that is evident in the digital smoke signals... This is computational behavioral economics,” he said. “We’re part of a whole new science here.”

Robert Kirkpatrick, director of Global Pulse (quoted in Lohr, 2011)

References and further reading
Global Pulse: Harnessing innovation to protect the vulnerable. Website (link)
Steve Lohr, pSR1 The New York Times. December 18, 2011. (link, may require registration and/or fee)

Discuss: Customer Data; Use It or Lose It

The motion for discussion is...
"Customer data; use it or lose it! discuss"
The motions are essentially to take positions for or against using customer data to the maximum possible, at the edge of the legal limits.

Goal
To review the status of data protection in Ireland, the rights of individuals and obligations of organisations. What are the implications for using and acquiring customer data, for direct marketing, data accuracy, faults, selling on, etc. The Office of the Data Protection Commissioner provides the following characterisation; customer rights, organisation's obligations. Is it possible to also characterise data management in terms of an organisation's rights and a customer's obligations?

Instructions
Two groups (red-against, blue-for) will present an analysis and then debate the merits of their different positions, involving the whole class in the closing stages.
The readings given for initial investigation are listed below. We strongly encourage the use of additional secondary sources and providing examples as evidence supporting arguments.

Readings

• Ireland's Office of the Data Protection Commissioner (http://www.dataprotection.ie/)
• Davenport, T. H., Harris, J. G., Jones, G. L., Lemon, K. N., Norton, D. & Mccallister, M. B. (2007) The Dark Side of Customer Analytics. Harvard Business Review, 1-9.
• Clarke, R. A. (1999) Internet privacy concerns confirm the case for intervention. Communications of the ACM, 42, 60-67.

Relational DB (mockup)

(link to Google Spreadsheet)

From Scrapbook Photos

Exercise: VLOOKUP - what it is, what it's good for, how to use it

What is VLOOKUP? A function that performs 'vertical lookups' in a spreadsheet.
VLOOKUP and HLOOKUP allow spreadsheets simulate the functionality of relational databases. LOOKUP functions employ database-like functionality and concepts like array indexing (column or row) over a number of tables.

The syntax is generally as follows:
lookup_value (the value you want to find in the table or array)
array or table_array (the range of cells comprising the table you are using e.g. $A$1:$C$9)
row_index_num or col_index_num (the content of the corresponding array location as indexed from the array you defined earlier, e.g. col_index_num value of '2' will return the value from the 2nd column)

The following illustrates - an example in Excel for Mac.


Goal
The objective of this exercise is:
to introduce the vlookup function;
Secondary goal is to use basic spreadsheet calculations: sum, addition, multiplication, bracket rules, cell ranges, relative cell references, absolute cell references.

Materials
At least one laptop computer running Excel or another spreadsheet application in each group.

Instructions
Allocate 20" to complete exercise.
Groups of between 2 to 3 students working on a single laptop.
Open the exercise spreadsheet and enter formulae, values and calculations to answer the set questions.

Comments
What to do if the student is running a Mac without MS Excel installed? Use Google docs spreadsheet in that case.
1. login to Google, select Docs and click the 'create new' button (LHS).
You can use the following example on googledocs.

Exercise: Pivot Tables

The following figures illustrate answers for assignment 5.
Note: Present each report in a separate, new worksheet.

Starting the pivot with the example spreadsheet:

a). Create a pivot table displaying a breakdown of the sales for each month in each state.
(i). In an attempt to identify trends, group the Month field by Quarters only.
(ii). Now adjust the Pivot Table so that it shows the sales in each quarter as a percentage of the full year.

b). Create a new pivot table displaying the type of product that was sold in each state for each month. Arrange the fields in a manner that most easily displays the information.

c). Create a PivotTable Report showing breakdown of the sales for each month in each state with pages formatted by Type.
(i). Filter the Type to show just White products.
(ii). Identify the state that does not have a good sales record. Filter this data to show what total sales would look like without this State.

d). Create a Pivot Chart displaying the sales per state for each month. Allow this chart to be filtered by product group.

Discuss: Senior Management Knows It All

The motion for discussion is...
"Senior Management knows it all! discuss"

Goal
To critically assess Nonaka's claim that innovative knowledge-creating corporations can be modelled as 'organisms,' with Senior Management envisioning, Managers interpreting and Employees experiencing.

Instructions
Two groups (red-against, blue-for) will present an analysis and then debate the merits of their different positions, involving the whole class in the closing stages.
The readings given for initial investigation are listed below. We strongly encourage the use of additional secondary sources and providing examples as evidence supporting arguments.

Readings
NONAKA, I. & TAKEUCHI, H. (1991) The knowledge-creating company : how Japanese companies create the dynamics of innovation, New York ; Oxford, Oxford University Press.

Open source programs open to scrutiny

It gives me a sense of comfort to know that open source programs are open to scrutiny by people who just want to know stuff and to people who need to know and be able to critique it. In this example Tali Garsiel, a developer in Israel who likes to figure out how things work, began exploring the internals of Firefox, Chrome and Safari (which is partly open source). She documented what she found... (link)

Dutch aim to explicitly protect "Fair Use"

What does the Dutch stance on relaxed copyright protections mean for Europe as a whole? A relaxed position on copyright or a lazy one? (link)

Ireland’s clouds a computing advantage?

(25.09.2009) Microsoft's completion and commissioning of a major European data centre in Dublin West to service the growing use of cloud computing services and applications was announced in 2009. Dublin was well placed to be selected to be one of Microsoft's key hubs for these growing services due to geographical and environmental factors, in addition to Ireland's previously acknowledged advantages for basing high technology industries here (education, language, cultural aspects, access to Europe, clustering of other high tech industries).

It is interesting to note that Ireland's climatic conditions are now seen to offer specific advantages for the operation of computer server farms - large, conditioned, secure facilities housing thousands of rack mounted computer servers - due to Ireland's relatively low ambient temperatures. Annual temperature averages for Ireland hover around 9°C, with coastal areas experiencing a relatively smaller range around the average than inland areas due to stablising influence of the Atlantic Ocean and the Irish Sea. Coupled with Ireland's stable temperature profile, its relatively high average annual rainfall raises the efficiency of geothermal heat pumps, which in combination help to minimise one of the principle cost drivers for data centres, the cost of cooling computers.

For more see SiliconRepublic.com

Discuss: Data Driven Decisions Remove the Human Factor

The motion for discussion is...
"Data driven decisions remove the human factor! discuss"

Goal
To critically assess the claims surrounding rational and intuitive decision making.

Instructions
Two groups (red-against, blue-for) will present an analysis and then debate the merits of their different positions, involving the whole class in the closing stages.
The readings given for initial investigation are listed below. We strongly encourage the use of additional secondary sources and providing examples as evidence supporting arguments.

Readings
LOVEMAN, G. (2003) Diamonds in the Data Mine. Harvard Business Review, 81, 109-113.
BONABEAU, E. (2003) Don't Trust Your Gut. Harvard Business Review, 81, 116-123.

In projects: Fail early, fail often.

The point with fail early and fail often is that it is a total radical upending of the way we generally think about working on projects, particularly projects where the goal we want to achieve has never been done (at least not in our knowledge).

Asa Raskin illustrates this idea with a story about Paul MacCready (designer of the Gossamer Condor) who put it this way “The problem is we don’t understand the problem.” In which case the best strategy is to learn quickly, to discover what the problem is. Often the goal itself seems to suggest solutions, perhaps obvious, but that these may be diversions to a more creative and ultimately successful resolution.

http://www.azarask.in/blog/post/the-wrong-problem/

Trading robots (software)

HFT Algo Machines from January 2007 through January 2012.
The following chart from Nanex displays the date/year and time/value of automated trades (high frequency trading);

http://www.nanex.net/aqck/2804.HTML

Exercise: Analysing data

Review the following data (link to Google docs)

What is the average transaction value based on this data?

This set of transactions is incomplete, estimate how many transactions in total occurred over the period? Assume transactions (see receipt number) are numbered consecutively when produced.

Provide an estimate the total sales value over the period.

Provide an estimate of one week's revenue based on a 5 day week.

Exercise: What is data, what is information?

Goal
Build up an understanding of the difference between data and information.

Instructions
1. Ask members of the class to define information and identify examples. (1 minutes)
2. Write down and display this list of definitions and examples.

Definitions
Something told
Items of knowledge
A thing known
...
...

Examples
News
A point on a scale
...
...
...

Comments:
There are no wrong answers, just build up a list, tabulate and provide back to the class.
What is data?
What is the relationship between data and information?
Consider difference between technical definitions and non-technical.

Date and time values in Excel (and Access etc)

Date/time values are stored internally in Excel as a real number where the apparent integer part and the mantissa part are interpreted as simple counts or fractions respectively.

• The data component is represented internally in Excel as the 'number of days elapsed since 12/30/1899.
• The time component is represented internally in Excel as the fraction of 24 hours.

Consider the following example:
Create your own test by entering January 26, 2011 9:00 PM the date into an Excel cell, e.g.

26/01/2001 21:00

Excel's auto format will usually recognise it as a date and record it (internally) as a date time.
Now change the cell format to General. You should see the following displayed.

36917.875

Date component	separator	Time component
36917	.	875

---

The mantissa part (the 'fraction part', everything after the decimal point or separator, also called a 'significand') is a decimal fraction of 24 hours.

The integer part is (the whole number in front of the decimal point or separator) is simply the number of whole days.

Question: does the actual internal representation in Excel map to a double precision floating point number or two integers?

Discuss: Formal Project Management Contributes to Project Failure

The motion for discussion is...
"Formal project management contributes to project failure! discuss"

Goal
To critically assess the claims surrounding the need, specification, assumptions and efficacy of the 'project management' approach.

Instructions
Two groups (red-against, blue-for) will present an analysis and then debate the merits of their different positions, involving the whole class in the closing stages.
The readings given for initial investigation are listed below. We strongly encourage the use of additional secondary sources and providing examples as evidence supporting arguments.

Readings
THE STANDISH GROUP (1999) CHAOS: A Recipe for Success.
FELD, C. S. & STODDARD, D. B. (2004) Getting IT Right. Harvard Business Review, 82, 72-79.