Cathy O’Neil, a data scientist and activist, has written a new book,
“Weapons of Math Destruction.” She is concerned about the proliferation
of certain kinds of algorithms – that help make important decisions, but
that could be based on unfair statistics with hidden biases. She
explains how to look out for them, and what we can do to protect
ourselves.
Cathy O'Neil's website
https://mathbabe.org/
Executive Edge Video
http://video.cnbc.com/gallery/?video=3000549011
YouTube Video
https://www.youtube.com/watch?v=gdCJYsKlX_Y
More or Less Podcast (2 September 2016)
http://www.bbc.co.uk/programmes/p046x243
http://open.live.bbc.co.uk/mediaselector/5/redir/version/2.0/mediaset/audio-nondrm-download/proto/http/vpid/p046n56s.mp3
Extract
https://www.theguardian.com/science/2016/sep/01/how-algorithms-rule-our-working-lives
Interviews with the Author
http://www.salon.com/2016/09/08/the-case-against-big-data-it-is-like-youre-being-put-into-a-cult-but-you-dont-actually-believe-in-it/
https://www.theguardian.com/books/2016/oct/27/cathy-oneil-weapons-of-math-destruction-algorithms-big-data
Reviews
https://www.math.columbia.edu/~woit/wordpress/?p=8732
http://blogs.scientificamerican.com/roots-of-unity/review-weapons-of-math-destruction/
https://www.infoq.com/articles/weapons-of-math-destruction-review
http://www.nybooks.com/articles/2016/12/22/they-have-right-now-another-you/
Related Posts
The Shelf-Life of Algorithms (October 2016)
The Transparency of Algorithms (October 2016)
Pay as you Share (November 2016)
Foundations of Business Organizations and Management, viewed from economic, ethical, social, and systems-thinking perspectives.
Showing posts with label decision-making. Show all posts
Showing posts with label decision-making. Show all posts
Monday, October 31, 2016
Monday, February 18, 2008
A Time To Act 2
Delaying a decision may simply involve holding out for a better offer. For example, when the directors of a company reject a take-over bid, claiming (as they always do) that it significantly undervalues the company.
It would be wrong to take the first offer that is made, if there is a reasonable chance of holding out for more. But on the other hand, if you wait too long for a higher offer, you may lose out.
In a post entitled Logic Need Not Apply, Microsoft employee John Evdemon comments on the refusal of Yahoo directors to accept a take-over bid from Microsoft. Is this simple greed, he asks, or are the Yahoo directors allowing their emotions to rule their heads? In asking this question, John is not just influenced by his current affiliation with Microsoft, but also with a startup company he was involved with previously, whose directors turned down a high offer and were subsequently forced to accept a much lower offer.
According to the New York Post (via Computerworld, Feb 15th 2008), there is a split in the Yahoo board, with some directors advocating acceptance. There is a threat of shareholder lawsuits if the directors could be proved to have acted emotionally, rather than in the best interests of shareholders. But how could this ever be proved?
It would be wrong to take the first offer that is made, if there is a reasonable chance of holding out for more. But on the other hand, if you wait too long for a higher offer, you may lose out.
In a post entitled Logic Need Not Apply, Microsoft employee John Evdemon comments on the refusal of Yahoo directors to accept a take-over bid from Microsoft. Is this simple greed, he asks, or are the Yahoo directors allowing their emotions to rule their heads? In asking this question, John is not just influenced by his current affiliation with Microsoft, but also with a startup company he was involved with previously, whose directors turned down a high offer and were subsequently forced to accept a much lower offer.
According to the New York Post (via Computerworld, Feb 15th 2008), there is a split in the Yahoo board, with some directors advocating acceptance. There is a threat of shareholder lawsuits if the directors could be proved to have acted emotionally, rather than in the best interests of shareholders. But how could this ever be proved?
A Time To Act
After several months in which alternative solutions to the Northern Rock crisis were explored, the UK government has finally decided to take the Northern Rock "bank" into temporary public ownership. [BBC News, February 17th, 2008]
The delay provides easy ammunition for political opponents.
As I have previously pointed out on this blog (decision-making, time), timing is an important aspect of decision-making. In this case, it is possible to debate not only whether the decision was the correct one, but also whether the decision could or should have been made earlier - or even later.
According to our model of decision-making (based loosely on Lacan), there are three phases in a decision.
It is easy to rival politicians to claim that the Government (or any other decision-maker) got the timing wrong, and to make wild and imprecise claims about their own super-human decision-making prowess: "Oh, we'd have acted straightaway" (whatever that means). We live in a culture that values speed and instant response, and is impatient of slow deliberation.
I don't actually know whether this particular decision was taken at the best possible time, and I guess you probably don't know that either. Just look out for statements about how quickly or slowly a given decision ought to be made, and see how these statements are underpinned not by reasoning but by instinct. Which means that if other people act faster or slower than you think they should, this doesn't necessarily mean they are stupid, it may simply mean they have different instincts.
The delay provides easy ammunition for political opponents.
- "After months of dither and delay we have ended up with this catastrophic decision," said George Osborne ( Conservative shadow chancellor).
- Liberal Democrat treasury spokesman Vince Cable said that the right decision had been taken, though "belatedly", and that the government should have walked away from the prospect of a private takeover some time ago.
As I have previously pointed out on this blog (decision-making, time), timing is an important aspect of decision-making. In this case, it is possible to debate not only whether the decision was the correct one, but also whether the decision could or should have been made earlier - or even later.
According to our model of decision-making (based loosely on Lacan), there are three phases in a decision.
- The instant of seeing - the moment it was recognized that Northern Rock was in trouble
- The time for understanding - a period of reflection, negotiation, allowing time for collaborative solutions to emerge
- The moment of concluding - the point at which all the available options are understood, and a choice must be made.
It is easy to rival politicians to claim that the Government (or any other decision-maker) got the timing wrong, and to make wild and imprecise claims about their own super-human decision-making prowess: "Oh, we'd have acted straightaway" (whatever that means). We live in a culture that values speed and instant response, and is impatient of slow deliberation.
I don't actually know whether this particular decision was taken at the best possible time, and I guess you probably don't know that either. Just look out for statements about how quickly or slowly a given decision ought to be made, and see how these statements are underpinned not by reasoning but by instinct. Which means that if other people act faster or slower than you think they should, this doesn't necessarily mean they are stupid, it may simply mean they have different instincts.
Monday, November 26, 2007
Decisions and Timing
One of the key aspects of decision-making is timing - judging when to commit yourself to something. Two contrasting styles/cultures can be identified
Some people / organizations are impatient with anything that inhibits action.
Lacan offers an interesting analysis of hesitation. Logical time is divided into three "moments".
The correct deduction appears to depend on the hesitation of the group. "If I had a red disk, then each of the other prisoners would not hesitate to deduce immediately that he was green. Since neither has done so, I must also have a green disk."
Delay, doubt, hesitation, procrastination, the ability to make nothing happen (ungeschehenmachen) - these characteristic features of decision-making are grounded by Lacan in the phenomenology of obsessional neurosis.
For a detailed discussion, see John Forrester, The Seductions of Psychoanalysis: Freud, Lacan and Derrida (Cambridge 1990), Chapter 8. The prisoner story can be found on p178 ff.
Related Post: Time to Act (Feb 2008)
- Delay / Hesitation / Procrastination
- Haste / Impatience
Delay / Hesitation / Procrastination
It is always possible to find a reason (excuse) for deferring a decision.- More information needed
- More options could be developed
- More stakeholders consulted
- Committees and subcommittees
- Referrals and due diligence
Haste / Impatience
Because it is always possible to find a reason for delay, some people / organizations are inclined to dismiss such reasons, or are reluctant to take them seriously.Some people / organizations are impatient with anything that inhibits action.
- JFDI - "just xxx do it!"
Lacan's Theory of Logical Time
Lacan offers an interesting analysis of hesitation. Logical time is divided into three "moments".
- the instant of seeing
- the time for understanding
- the moment of concluding
The correct deduction appears to depend on the hesitation of the group. "If I had a red disk, then each of the other prisoners would not hesitate to deduce immediately that he was green. Since neither has done so, I must also have a green disk."
Delay, doubt, hesitation, procrastination, the ability to make nothing happen (ungeschehenmachen) - these characteristic features of decision-making are grounded by Lacan in the phenomenology of obsessional neurosis.
For a detailed discussion, see John Forrester, The Seductions of Psychoanalysis: Freud, Lacan and Derrida (Cambridge 1990), Chapter 8. The prisoner story can be found on p178 ff.
Related Post: Time to Act (Feb 2008)
Wednesday, November 08, 2006
Market Timing 2
I have referred on this blog before (Market Timing 1) to the importance of time/timing in making decisions. In Lacan's theory of Logical Time, there is "a moment to conclude". Lacan is one of the few thinkers who appreciates the need to explain why decisions and actions take place at a particular point in time - helping us understand, among other things, how decision-makers are affected by haste and hesitation.
Now marketing guru Seth Godin has a post on his blog on the importance of timing in marketing. He points out that the effect of a message depends on the state of the recipient when the message arrives.
But this is not a good example of Lacan's logical time, because an election is held according to an external schedule. What is more interesting in elections is the timing of a campaign. Seth is absolutely right to point to the importance of groundwork, but no amount of groundwork can prevent your candidate from peaking too soon.
Technorati Tags: Lacan marketing systems thinking timing
Now marketing guru Seth Godin has a post on his blog on the importance of timing in marketing. He points out that the effect of a message depends on the state of the recipient when the message arrives.
"All marketing analyses that ignore time are wrong. There's a big difference between a message that arrives when I'm full and when I'm unfull. And there is a big difference between a first impression and tenth one. Even if I can't remember the first nine."What happens when people have to make a decision on a particular day - for example an election? Some people make up their mind at the last minute. Perhaps it is not rational to commit to a candidate too early, since any candidate may say something unwise, and damaging revelations may emerge at any time before polling day. (A Republican candidate dropped out five weeks ago when inappropriate emails came to light [source: BBC News]. And after yesterday's mid-term election in the USA, the balance of power in the Senate depends on the outcome of the vote in Virginia, where the Republican incumbent was well ahead until he was caught on video uttering a racial slur [source: BBC News].)
But this is not a good example of Lacan's logical time, because an election is held according to an external schedule. What is more interesting in elections is the timing of a campaign. Seth is absolutely right to point to the importance of groundwork, but no amount of groundwork can prevent your candidate from peaking too soon.
Technorati Tags: Lacan marketing systems thinking timing
Thursday, February 24, 2005
ChoicePoint
ChoicePoint is in the news this week, because it "mistakenly" sold personal credit reports for about 145,000 Americans to criminals. (I like the word mistakenly - it leaves us wondering whether this was an error of intention or of execution. After all, if you get found out, it's always a mistake.)
Bruce Schneier (Feb 23) writes: "ChoicePoint's behavior is a textbook example of how to be a bad corporate citizen. The information leakage occurred in October, and it didn't tell any victims until February. First, ChoicePoint notified 30,000 Californians and said that it would not notify anyone who lived outside California (since the law didn't require it). Finally, after public outcry, it announced that it would notify everyone affected."
Interesting decision-making process here. Public trust in a company is critically affected by the way it deals or dithers with a crisis such as a product recall or website error. See my notes on Kodak, Sudan 1.
Adam Shostack (Feb 23) suggests we might expect shady behaviour from firms that don't expect to be around for very long. There is therefore a possible link between ethics and viability - poor viability leads to poor ethics. But what about the causal link the other way - what effect does bad corporate behaviour have on corporate (economic) viability?
Adam Shostack (Feb 24) also raises the question, "who notified whom? Reuters claims that the authorities notified Choicepoint, while Choicepoint claims they notified the authorities. Let's see...who has motive to lie?" Gary North posts one theory (Feb 19) "According to ChoicePoint, there was no announcement because law authorities prohibited it." to which IanG (Feb 19) replies "OK, so maybe they wanted to set up a sting. They're the good guys, and they're in control, right?"
Lots of information from EPIC and SourceWatch, and further links from Bruce and Adam.
Bruce Schneier (Feb 23) writes: "ChoicePoint's behavior is a textbook example of how to be a bad corporate citizen. The information leakage occurred in October, and it didn't tell any victims until February. First, ChoicePoint notified 30,000 Californians and said that it would not notify anyone who lived outside California (since the law didn't require it). Finally, after public outcry, it announced that it would notify everyone affected."
Interesting decision-making process here. Public trust in a company is critically affected by the way it deals or dithers with a crisis such as a product recall or website error. See my notes on Kodak, Sudan 1.
Adam Shostack (Feb 23) suggests we might expect shady behaviour from firms that don't expect to be around for very long. There is therefore a possible link between ethics and viability - poor viability leads to poor ethics. But what about the causal link the other way - what effect does bad corporate behaviour have on corporate (economic) viability?
Adam Shostack (Feb 24) also raises the question, "who notified whom? Reuters claims that the authorities notified Choicepoint, while Choicepoint claims they notified the authorities. Let's see...who has motive to lie?" Gary North posts one theory (Feb 19) "According to ChoicePoint, there was no announcement because law authorities prohibited it." to which IanG (Feb 19) replies "OK, so maybe they wanted to set up a sting. They're the good guys, and they're in control, right?"
Lots of information from EPIC and SourceWatch, and further links from Bruce and Adam.
Labels:
decision-making,
ethics,
risk-trust-security,
trust
Subscribe to:
Posts (Atom)