Wednesday assorted links


4a. I come from a fairly quantifiable field, software development. First and foremost, our programs crash or not. But beyond that it is pretty easy to generate functional requirements, testing regimes, error rates, and test coverage.

"Entire genres of prejudice-reduction interventions, including diversity training, educational programs, and sensitivity training in health and law enforcement professions, have never been evaluated with experimental methods."

I think I've come to a surprise that many fields that I expect to use the scientific method in testing for validity do not. More should, and not limited to this.

Well, that depends on the software, and the level of available resources. I know of large (multi-million line), complex technical software products that shipped with many hundreds of known defects, and as demonstrated by subsequent use, many more hundreds of unknown defects, albeit many were minor or associated with uncommon use cases. And with gaping holes in meeting business needs.

Even Level 1 (chaos) software organizations usually make some attempt to verify (meets stated specs) the product, although validation (meets the business need) is often very informal and implicitly delegated to customers. More mature organizations devote a lot more attention to validation.

None the less, I share your surprise. In that area, one suspects lack of resources, lack of incentives (a "reasonable" claim is just as career enhancing as a validated one), delays from input to output longer than the stable lifetime of the involved organizations, and general inability to really prove effect even if they tried, all contribute. And many of the practitioners are probably in a Level 1 world.

We are agreeing though, because when products ship with "hundreds of known defects," testing and quantification are clearly in play. In modern environments those hundreds have been reviewed, assigned a severity, and "allowed" on that basis.

What would be the contrary example, something from (early) Taleb? One of his examples was an oil organization that put out predictions for future prices, simply adjusting them each month, never looking back, and never commenting on errors. He called it "do overs."

Or in the original quote, someone doing "training" with no idea of value added.

Oh oh.

"I share your surprise..."

You only ask questions when you want to know the answer.

Yes, it could be quantified, but apart from a brief period during the 1970s it has been all ego and bluster:

That is a cute enough essay, but it does suffer from a very strange idea that software engineering is one thing, practiced by all organizations the same way in the same decade.

Methods abound.

The big hurdle is for companies to recognize that their process is something to be designed and iterated, just as seriously as the product or service itself.

“Entire genres of prejudice-reduction interventions, including diversity training, educational programs, and sensitivity training in health and law enforcement professions, have never been evaluated with experimental methods.”

Having worked in my share of medium-sized and large organizations, this really isn't surprising to me. It's always appeared to me that organizations engage in these programs as a form of prophylactic signalling, to their employees (especially those who may be harassed or experience prejudice), to the government, and to the court system (should any future cases reach the court). The HR department is busy checking off boxes and performing corporate CYA. Those engaged in the training are generally both sincere and incredibly naive, and have no interest in determining if what they're doing as a career is actually useful. Employees reluctantly sit though these programs, and if anyone actually raises real and actual workplace issues in these diversity/prejudice/harassment/sensitivity meetings and seminars, alarm bells go off and that discussion is immediately shut down. It's theater, designed to appear to address problems that aren't usually even defined clearly, with the real goal of keeping business as usual operating.

#6 The Catalonian people, dispite massive policial terror enforced by Madrids diktat, has risen against the Castillian fascist occupier, who will be driven out no matter what the costs may be.

I was for Catalonian independence until I read, from what TC wrote, that it's not 90% of people for it, but 50%. It's more like Quebec. However, I am for Kurdish independence because there it really is 90%. A land-locked country like a Kurdistan would be doomed to mediocrity but let them have it I say.

The Catalonian people has had its say in this matter. It is time to throw away the Castilian Yoke!

"A land-locked country like a Kurdistan would be doomed to mediocrity but let them have it I say."
It is still better than being America-controlled Iraq!

Have you read Orwell's "Homage to Catalonia?"

The question for Catalans is "What are you prepared to do?"

It seems (History) that 50% active support for an independence or revolutionary movement would result in that movement's victory.

In 1176, American colonists were evenly divided between independence, loyalists, and neutrals. With much blood and treasure and God's help, the American Revolution prevailed. I think 1917 Russia was similar.

Thing is the those examples likely are impossible in these degenerate times.

The Catalans got religion right quick when the EU said "Actually no, you will never be in the EU, ever, as an independent country". Hence last nights "just kidding" back-peddle.

Plus, the snow flakes and degenerates would have surrendered to an invasion by a troop of weaponized, Spanish cub scouts.

I rehearse the dbltap at the range once a week.

> when the EU said “Actually no, you will never be in the EU, ever, as an independent country”

Why is the EU trying to sweeten independence?

@ #1 - this paper is good, you should read it. Basically all of economics is normative, an an extension of political science, and should be (ought to be) read as such. They don't call the dismal science "political economy" for nothing.

No comment on the Kevin Bryan post? It's all about patents, I'm honestly curious to read your take

@msgkings - you are referencing #3. I just read it and found nothing really newsworthy in it. I do agree we need a better incentive system, for example, laid-open patent applications (no examination, like today's US copyright applications, and as done for patents in Japan and Germany for certain types of minor inventions), and, as in Japan and partly in West Germany, a potential government granted prize for breakthrough patents, going directly to the inventor, regardless of what the inventor signed to their employer. That's a start, along with 'crowd sourced' patent examination, and a British-type "losing plaintiff pays the defendants legal bill" (aka "English Rule") to cut down on patent trolls. But the article never even mentioned this at all, instead it was very squishy. Thanks for asking. BTW nothing I say above is novel to me, it's been said by many in the patent field. The fact it never sees the light of day among the general public is only because most people have no interest about inventions, aside from an occasional fantastic story about Nicola Tesla or some movie like "Tucker" and so on. Utter garbage. But GIGO, and that's our innovation system, basically depending on altruistic Good Samaritan nerds for the most part for breakthrough inventions (market failure to a degree). As TC has said, public praise of inventors goes a long way and should be encouraged, and that's the traditional (and IMO flawed) model for engineering innovation (aka Nobel Prize, aka Field Medal, etc etc).

No, I was referencing this:

Like politics, all economics are local/micro- and post-modern economics are politicized, dismal-pseudo-science. For you victims of public schools, that means I have no respect for the field.

3. Carroll and Frakt make the observation that innovation in health care tends to be at the margin, an example being a (very expensive) drug that might add a year or less to a terminally ill patient (my example not theirs) - the macabre television commercial for a new cancer drug with the jingle "tomorrow, tomorrow, tomorrow", the promise being a few more tomorrows for patients who take the drug. Of course, such expensive drugs aren't equally available to all. In an earlier thread today I made the point that, while everyone is waiting for the next big innovation (flying cars and spaceships to Mars), in the real world it's the sum of the marginal (mini) innovations that add up to the big innovation from which economic growth flows. Whether I'm right is debatable (I'm merely repeating what economists have concluded), but in health care it's much the same: everyone is waiting for the next big innovation, such as a (non-existent) "cure" for cancer, while in the real world it's marginal (mini) innovations that broaden the delivery of life-saving and lie-prolonging drugs and treatment that have the greatest impact.

#6: What's interesting is that people's responses are influenced by the position of the People's Party. This rather affluent region of Spain rejects the starboard party by 10-1 margins. In Scotland, a mildly affluent region, the starboard party is rejected by more than 3/4 of the electorate. In Greater London, the most affluent part of the UK, 70% of the population lives in Labour boroughs. In the United States, the most affluent commuter belts - the Bay Area, Washington, New York, Boston, give a huge advantage to the Democratic Party.

As famous Chinese Communist leader Chairman Mao Zedong pointed out, the cities are surrounded from the countryside.

The Ghost of Manuel Fraga is hovering over Catalonia.

What is more interesting is the rest of spain -- if elections were called we could expect a PP majority, but I suspect that PSOE would be forced into third position.

The Ghost of Manuel Fraga is hovering over Catalonia.

He was a cabinet minister in the latter third of the Franco regime; his portfolio included public relations and tourism. That would be of special interest in Catalonia just why?

Fragra was the founder of PP.

Elsewhere in Spain, I don’t see the ghosts of Franco. Zapo tried a few years back, the blowback (and the nationality issue) was enough to give Rajoy the office.

But in Catalonia, PP is still considered a fascist party.

PP has many issue; corruption the biggest. Rajoy has many flaws. But only in Catalonia is PP considered fascist.

Is the PP popular at all in Pais Vasco?

If Rajoy was a true leader, he'd try to meet with the Catalans and figure out a compromise- without having to give into a referendum. But he isn't and it appears that the PP is just a Spanish nationalist party with little regard for the outer autonomous regions of Spain. Whether he'll be able to crush them or not remains to be seen.

But in Catalonia, PP is still considered a fascist party.

OK, Catalonia has a lot of meatheads in its electorate.

If Rajoy was a true leader, he’d try to meet with the Catalans and figure out a compromise- w

That's silly. A compromise is what his interlocutors do not want. They might accept it after having their insteps smashed enough times. You have to run over their feet first.

The basque nationalist parties are backing his government; they take away support and it goes.

So yes, EAJ doens’t seeem to be hurting by that partnership.

#6 A useful reminder of the Washington Post's truly rotten quality. The piece is based upon one series of polls by one organization that ends back in July. Thank god for wikipedia where we can easily see that the data point the piece hangs on is unrepresentative and that there other multiple polls that are more in agreement on higher levels of support for independence:,_2017#Opinion_polls

File alongside "Brexit will obviously be rejected" and "Trump will obviously not get even 200 electoral votes."

Let's hear it for WaPo polling, ladies and gentlemen!

The Washington Post is a lapdog of the CIA and the Department of State.

Yes, it is a lapdog of the CIA. Bezos' CIA contract is worth much more to him than the newspaper. But don't bypass the CIA. The elder Sulzberger was essentially a contract employee for the CIA - he signed a confidentiality agreement and posted CIA people to newspaper posts. I don't know what the son is doing, but the history is there. (Just Google Carl Bernstein and CIA and you'll get his article explaining all this in detail.)

I don't know who the writer of the Post article is, but he's from U of MD and that's a university heavily tied into the DOD.

On a less conspiratorial note, the comments to the Post article have it right. Madrid tried to suppress turnout and then argued it wasn't a representative referendum. The writer may be correct that the Madrid opposition has polarized the situation. But he leaves out a huge fact - the Spanish police brutalized the Catalan population. Of course that's going to push people into an opposition camp. That's similar to US foreign polity blindness -- "they're bombing us -- why are they so angry at us?"

The CIA has a long history of supporting fascist governments. I would not be surprised if this article was proposed by them. Read "The Devil's Chessboard" about Allen Dulles, and how so many people within the OSS, the later CIA and the State Department were supporters of fascism. And, specifically, Spanish fascists.

Next the Washington Post will explain how a small minority of wackos in Massachusetts maneuvered the British governor into sparking a Revolution in 1775.

That was very clever. Good on you.


"In this paper, we investigate whether a statement like “the reform ought to be favored” is practically the same as “the reform would yield positive net consequences.” The first statement, featuring “ought,” is one that many economists would deem “normative,” the second, “positive.”"

The second statement implicitly assume you agree with the objective function being maximized, hence it is still normative.

That is spot on. They set up a straw man. Positive economics tells you 'if you do A, then B is the likely outcome'. It doesn't tell you whether B is more valuable than C.

Economists are terrible philosophers.

Agreed. In my field (accounting), I think a survey would show that most of us would see all of these claims as normative. Positive would be a claim like "we find that Financial Accounting Standard 394 caused stock prices to be more correlated with future cash flows." A claim like "This proposed standard would increase the correlation between stock prices and future cash flows" would be viewed as 'predictive', but would almost never pass peer review because that is so hard to establish. Usually the only evidence brought to bear is what happened in the past, so we ask that authors focus on that.

No reviewer or editor would ever accept a paper whose major claim focused on the future net benefits of the standard, whether or not they used the word "should", because some people will benefit and some people will suffer from just about any change in accounting standards, and balancing their preferences is not our expertise. We would leave that for lobbyists, though occasionally we'll publish that kind of thing as a peer-reviewed thought piece.

#2 "As with his breakout indie hit Tangerine, Baker’s approach is to immerse you so deep in a place, and the patter of the people who live there"

I tried to watch Tangerine. I really did, but those were the longest 20 sunk cost minutes of my life.

I haven't seen that, but I did greatly enjoy The Florida Project. I wonder what Cowen perceived to be the flaws.

The first paper is very interesting, and it sort of leads me to wonder, if economists beliefs on these issues are really just an extension of their values, then what is the value-added from asking economists about their expertise on those issues? For example, it is a common refrain to note that economists tends to favor more liberal immigration policies, because on average, these policies make people better off. But there are obvious distributional consequences. I am not implying that they ignore these distributional consequences, but the weight you assign to them is going to in some sense depend more on your own values than any kind of economic calculation. Not to mention, immigration is not just purely about economics either.

Economists have a tendency toward efficiency and growth as ultimate objectives. While understandable for a science, it's a rather narrow view of things.

Clearly distribution matters. Someone born in the US has a greater claim to the use of local public goods than someone born elsewhere. You can never discount violent revolution, however irrational, as a limiting factor in even the best intentioned and well supported political machinations. Low skilled people don't like having their wages bid down to nothing. At some point, crime or warfare becomes the path of least resistance.

Why don't any of these articles on Catalonia analyze why the PP is so hated there?

Why not analyze why Catalonia's voters are such maroons?

#1. I've never been clear on the normative/positive distinction. I feel less stupid today.

The is-ought distinction goes back to Hume. There are empirical observations, but they can never (by themselves) lead to a conclusion involving valuation. That is inherently subjective. The problem with the linked paper is that the "positive" statement is actually normative (as pointed out by Jim & Danny Kay above).

Talking of "positive" net consequences is still normative. Positive economics would just tell you what the net consequences are and let you decide how to feel about them. Shitty, shitty study.

#7 How embarrassing to have Obamacare blamed on your “work” if you can call saying “let’s just pass a boatload of laws to make people do what I think they should do” work. On the other hand, perhaps success at achieving the difficult task of making nudgers seem less toxic.

none of this even remotely describes MVP or his work.

naturally, klein's newest crappy survey will be published in the journal he himself edits; naturally without any real peer review. long live the GMU spirit!

#6: an independent Catalonia is possible.

However, how will it look? Present incertitude and lack of leadership in Catalonia opens the door to populist leadership. Will Catalonia go full Cuba, a slow decay of living standards as Argentina or circus style Venezuela?

the Catalonia independence "vote" was a joke...
1/ there were no lists
2/ people could vote as many times as they wanted, in as many polling stations as they wanted
3/ no audit
4/ a final turnout figure under 40% despite all of the above

even the Crimea "referendum" was carried out in a more proper way.

1. This is the Economics 101 distinction between normative and positive.

In any decision making method, there will be an objective function that calculates a preference ordering from input parameters. Assuming well behaved preferences, one can justify selection of almost any feasible choice by weighting the parameters appropriately, except those that are strictly dominated by others. And even those that are strictly dominated can be selected by adding criteria that are unique to that choice and weighting it heavily.

Someone who maximizes expected utility will likely make a different choice than someone who minimizes maximum regret, given identical choices and conditions.

While positive economics concerns the mathematical formulas that govern choice for a particular objective and conditions, normative economics deals with the selection of objective functions, weights, time preference, etc. These are value laden and cannot be reduced to an incontrovertible answer unless one views these choices as endogenous to the overall environment.

Perhaps this distinction vanishes in the meta-analysis, taking all personal values as inputs to a cosmic black box. But this paper doesn't go that far.

#6. Regardless of the exact percent of Catalans who support liberation, it appears that a significant portion of citizens of EU member nations would like to be liberated and to have greater local autonomy. The author at the link below raises the interesting question of whether the EU should use its power to regulate secession within member nations. It would be somewhat ironic if regional autonomy got a boost from a supranational entity.

Comments for this post are closed