Tuesday, January 21, 2020

Tuesday Tweets

I know you're all tired of hearing this, but there's something strange about equating electability with white and male in 2020.



Glad it's not just me.


Remember when we would have assumed this was a joke?


I have a feeling that this is one of those issues we should be talking more about.


This one goes in the Wages of Strauss file.
 
 We have to check in with the unicorns.



You just knew there was a McKinsey connection.



Finally.

Posted by at 2 comments:

Monday, January 20, 2020

Back to the 80K question – part 1: why it matters

A while back, we had a discussion on the closeness of the 2016 electoral college vote and how best to quantify it. I argued that the best approach was to look at the minimum number of votes that would have to change in order to get a different outcome. For the last election, that was around 80,000 votes which, I argued, constituted a close race. I got some pushback so I thought I’d fill oout some of the details, starting with why we should bother with the question at all.

Different kinds of wins have different implications, both strategic and psychological. If there’s a disconnect between the way people are reacting and the actual events, that is in and of itself a phenomena worth noting and studying, but if they are making decisions based on those misinterpretations and false narratives, the problem becomes both important and urgent.

The disconnect around the closeness of Donald Trump‘s victory has manifested itself in at least a couple of ways. First is the theory that Hillary Clinton didn’t just lose, but that it wasn’t’ even close. The argument goes on to say that, in order to win, the Democrats would need to add millions to their popular vote advantage to overcome the biases of the electoral college.

Second is the belief that the margin of defeat was so massive that the Democrats need to do everything differently this time. This has particularly centered on going back to what is traditionally considered a more electable candidate. Specifically a white male from north of the Mason-Dixon Line.



Since every woman who has been successful enough to compete at that level will have to deal with charges of unlikeability, this is basically saying the Democrats need to nominate a man, 

Obviously, any conversations about the closeness of an electoral college win are going to be problematic. Every statement will rest on an edifice of challengeable assumptions and definitions, none of which can ever rise above the standard of merely reasonable.

Nonetheless, this is an absolutely necessary conversation. It matters whether or not Hillary Clinton lost by 80,000 votes, 2 million votes, or somewhere in between, both in terms of our analyses and our decisions.
Posted by at No comments:

Friday, January 17, 2020

"Now, if we wanted to take the children to see a Komodo Dragon..."

The moderators of this week's Democratic debate have taken some well deserved heat for asking Sanders to respond to alleged comments about the electability of women, then asking Warren a question with a premise that contradicted the first answer.

Putting aside for the moment questions of bias against Sanders, this reminded me of the platonic ideal of inattentive interviewing.





Posted by at No comments:

Thursday, January 16, 2020

Two reposts



One on the electability of women, the other on the lack of electability of a demographic that includes Biden, Sanders, Buttigieg... (apply all appropriate caveats)

Wednesday, July 31, 2019

Any conversation about electability has got to start with the fact that a white male has not won the popular vote since 2004.

The electoral discussion among pundits and data journalists has been taking some especially silly turns of late and before the bullshit accumulates to the same dangerous level it did in 2016, we need to step back and address the bad definitions, absurd assumptions, and muddled thinking before it gets too deep.

We should probably start with the idea electability. While we can argue about the exact definition, it should not mean likely to be elected and it absolutely cannot mean will be elected.

Any productive definition of electability has got to be based on the notion of having reasonable prospect of winning. With this in mind, it is ridiculous to argue that Hillary Clinton was not electable. Lots of things had to break Trump's way for him to win the election and, while we can never say for certain what repeated runs of the simulation would show, there is no way to claim that we would have gotten the same outcome the vast majority of the time.

This leads us to a related dangerous and embarrassing trend, the unmooring of votes and outcomes. This is part of a larger genre of bad data journalism that tries to argue that relationships which are strongly correlated and even causal are unrelated because they are not deterministic and/or linear. In this world, profit or even potential profit is not relevant when discussing a startup's success. Diet and exercise have no effect on weight loss. With a little digging you can undoubtedly come up with numerous other examples.

The person who wins the popular vote may not win the electoral college, but unless you have a remarkably strong argument to the contrary, that is the way that smart money should bet.

Monday, November 18, 2019

Fun with Political Trivia

This picks up on a recent thread (telling which one might be too much of a clue). The ones and zeros represent a trait of Democratic candidates from 1964 to 2004. Take a look and think about it for a moment. Here's a hint, the trait is something associated with each man well before he ran for president.

Johnson           1
Humphrey       0
McGovern       0      
Carter              1           
Mondale          0           
Dukakis           0           
Clinton            1                 
Gore                1                      
Kerry               0


As you might have guessed, the relationship between this trait and the popular vote didn't hold in the previous or following elections. The trait is not at all obscure. It was well known at the time and figured prominently into their political personas, This is not a trick question.




Read more »
Posted by at No comments:

Wednesday, January 15, 2020

The Golden Globes are a perfect example of journalism's non-news addiction

[Sorry about sending out the draft form earlier]

Just to get this out of the way, all major entertainment awards are largely worthless. Poor as measures of talent, even worse as indicators of artistic quality, but even by this debased standard, the Golden Globes stand alone in their lack of merit. It has always been and remains a joke.

Winning something like an Oscar or an Emmy probably has less to do with the work and more to do with lobbying and popularity, but it does, at least, represent the approval of the industry. A Critics Choice Award is given by people who watch and write about films for a living. Even the People’s Choice  award indicates a level of popular support. The Golden Globes is given out by a small group you've never heard of. >
The award started out as an excuse to party with movie stars but it quickly evolved into one of Hollywood’s most notorious scams. Here are a few highlights from Wikipedia:

1968–1974 NBC broadcast ban

The HFPA has had a lucrative contract with NBC for decades, which began broadcasting the award ceremony locally in Los Angeles in 1958, then nationally in 1964. However, in 1968, the Federal Communications Commission claimed the show "misled the public as to how the winners were determined" (allegations included that winners were determined by lobby; to motivate winners to show up to the awards ceremony winners were informed if they did not attend another winner would be named). The FCC admonished NBC for participating in the scandal. Subsequently, NBC refused to broadcast the ceremony from 1968 until after 1974.

Pia Zadora awarded "New Star of the Year in a Motion Picture" in 1982

In 1982, Pia Zadora won a Golden Globe in the category "New Star of the Year in a Motion Picture – Female" for her performance in Butterfly, over such competition as Elizabeth McGovern (Ragtime) and Kathleen Turner (Body Heat). Accusations were made that the Foreign Press Association members had been bought off. Zadora's husband, multimillionaire Meshulam Riklis, flew voting members to his casino, the Riviera Hotel in Las Vegas, which gave the appearance that they voted for Zadora to repay this. Riklis also invited voting members to his house for a lavish lunch and a showing of the film. He also spent a great deal on advertising. Furthermore, Zadora had made her film debut some 17 years earlier as a child performer in Santa Claus Conquers the Martians.

The Tourist for Best Musical/Comedy nominations in 2011

The nominations for the 2011 Golden Globes drew initial skepticism as the Hollywood Foreign Press Association nominated The Tourist in its Best Musical/Comedy category, although it was originally advertised as a spy thriller, and also as one of the most panned films of the season with host Ricky Gervais even jokingly asking the main star of the film, Johnny Depp, if he had seen it. Rumors then surfaced that Sony, the distributor of The Tourist, had influenced Globes voters with an all-expenses-paid trip to Las Vegas, culminating in a concert by Cher.




Keep in mind that Zadora and The Tourist only got attention because they were already punchlines before the Globes. There’s no reason to assume the opaque process was any less corrupt the rest of the time.

Everyone who follows the industry knows the Globes are of no news value to the Golden Globes, but every year we get actual news organizations with actual reputations runs something like this:

ARI SHAPIRO: Was there anything among the winners last night that really raised your eyebrows and you think might hint at the future of TV?
The Golden Globes are a perfect example of journalism's non-news addiction. Everyone knows they are absolutely meaningless, at best telling the viewers nothing more than who wrote the biggest check, but no one can seem to resist them.







Posted by at No comments:

Tuesday, January 14, 2020

Tuesday Tweets











 











Posted by at No comments:

Monday, January 13, 2020

Efficient infringement -- when it's your fiduciary responsibility to steal stuff

This is a decidedly unhappy sequel to our previous stories on patent trolls, a grim little good news bad news joke. The good news is that we have found a way to prevent trolls from holding companies hostage with illegitimate claims of intellectual property; the bad news is that we did it by taking away the rights of people with legitimate claims.

From Joe Nocera:
If you’re not a patent aficionado, you have probably never heard the phrase “efficient infringement.” Not to blow the punch line, but it’s yet another example of how big tech companies use dubious means to squeeze their smaller rivals. When critics complain that Facebook, Google, Apple and Amazon are squashing innovation, this is the sort of tactic they’re talking about.

I first heard the phrase some years ago when I looked into a patent dispute between Apple Inc. and the University of Wisconsin. The Wisconsin Alumni Research Foundation, which owns the university’s patents, had sued Apple over its use of an innovation that university scientists had dreamed up and patented in the mid-2000s. Apple had installed it in iPhones and iPads without bothering to negotiate a license and had been using it with impunity for years before the case finally went to trial. A jury found for the foundation in 2015 and ordered Apple to pay $234 million.

That was actually a victory for Apple. Consider: Apple got free use of valuable technology that it took from a smaller, less powerful entity. Losing at trial was unfortunate, perhaps, but the $234 million was just another business expense. Pocket change, really, considering Apple’s size.

How common is this kind of move? Boris Teksler, Apple’s former patent chief, told the Economist recently that “efficient infringement, where the benefits outweigh the legal costs of defending against a suit, could almost be viewed as a ‘fiduciary responsibility,’ at least for cash-rich firms that can afford to litigate without end.” In other words, stealing patented technology is practically required if you’re focused on shareholder value. And who isn’t these days?

...

“What is the definition of a patent?” asked Brian Pomper, the executive director of the Innovation Alliance, which represents small and medium-size patent holders. “It means you have the right to exclude others from your invention.” He added, “How can a patent be meaningful if you don’t have the right to exclude others?” But that’s where we are. Companies like Sonos have virtually no leverage to stop bigger companies with deep pockets from infringing their patents.











Posted by at No comments:

Friday, January 10, 2020

A media post on Star Wars

This is Joseph.

So a few thoughts on the newest Star Wars movie.

Needless to say: SPOILERS ahoy!

One, the movie was beautiful. The special effects were stunning and the visuals were memorable.

Two, are there no proof-readers in Hollywood? Can we not decide what the antagonist's plan is?

But the part that was the most challenging to me as a viewer was that there was simply no continuity with previous plot points in the other movies. Some of this is forgivable, when you decide on an awesome twist it can be okay to have to patch over a rough spot. Think of Obi-Wan's retroactive deception about Darth Vader, once the writer puzzled out the family tree (or the even more awkward Leia is your sister moment).

But Kylo Ren is just confusing. The second movie seems to give him every chance to decide on his allegiance, even insofar as there was much doubt after the first movie. His history of misdeeds is . . . extreme. He was involved in a mass murder terror attack (Starkiller base). He murdered his fellow students at a school because he disliked the teacher (and focused so much on the other children that the teacher survived). He massacred a village in a terror attack. He murdered a family member as a part of a gang initiation. So his final arc is redemption?

It isn't that it couldn't have been plotted well.  It is just that writers seemed to keep changing their minds, and that is only worth it if the overall outcome is exceptional. The movie was fine, but the plot had a lot of obvious holes (why where the ships buried and were the crews on them when they were)?

Finally, it detracts from the previous movies. What is the point of the middle part of the trilogy now? The heroes fail at pretty much everything and not in a way that teaches lessons or because of a character flaw. No, they fail because the Emperor can survive being blown up and the protagonist doesn't sense the deception.

So it was nice to watch and I am glad I saw it. But it will generate a month of people wondering exactly what happened.
Posted by at No comments:

Two more tech spots from CollegeHumor

This one remains sharp and on-target.




This parody of start-ups, on the other hand, can't hope to compete with the real thing in the age of unicorns.













Posted by at No comments:

College Humor -- "If Internet Ads Were Salesmen " -- repost

I keep meaning to do a post about the terrible state of targeted marketing. When I get around to it, remind me to embed this. At least half of the points I want to hit are illustrated here.





Pre-update:

After I scheduled this in the form above, Josh Marshall posted a piece on internet advertising and the death of Gawker. It contains an informative primer on how this stuff works.

Many people think that the more popular a publication gets the more ads it will sell. The bigger the audience, the more eyeballs, the more ads wanting contact with those eyeballs. That's not how it works.

There are a million dimensions to the advertising economy, just as many ways of describing it. But you can understand a whole lot about how the whole thing works by thinking in terms of three factors: 1) endemic sales proposition, 2) controversy and 3) influence.

Let's talk first about endemic sales proposition. Because I think it may have played some role in Gawker's demise (on-going legal liability may have played more of a factor or have been the entirety of the issue). A site about clothes has an endemic sales proposition: selling clothes. A site about books: books. You may say well, I only read sites about news and sports but I still buy a lot of clothes so ... Not how it works.

For a variety of reasons, some good and evidence based, others silly, advertisers want to sell you their product when you are thinking about it and in the mindset to buy. This doesn't just mean impulse purchases, but buying in general. In many cases that makes a lot of sense.

For instance, aside from people being really into tech, why do you think there are so many tech sites? Right, because there's a ton of money in video games, devices, computers, everything under the sun. People also tend to buy those things online. Again, we're not just talking about impulse buying. It can be more nuanced and less direct. But if you stand up a site about tech, gaming, computers, etc. and it does well, you have a ready made channel for ad sales. And in the case of tech an extremely lucrative one.

Sometimes it's a little more amorphous but no less ad driven. Why so many 'lifestyle' publications? Well, we all need a lifestyle, of course. And general interest magazines cover many interesting topics. But by and large that's because you're aiming for an audience of people who are affluent and want to read about cool things affluent people do: travel, toys, aspirational personal development. Not that there's anything wrong with that, as they used to say on Seinfeld. But that's what it's about.

Next, controversy. This largely speaks for itself. Advertisers don't want to be around things that upset people or divide people. They want to be everyone's friend. They don't want negative ideas or stories to rub off on them. This isn't an absolute of course. Plenty of sites which court controversy sell tons of ads. Gawker's a prime example. But controversy is always a constraint on ad sales. You just may have other factors that overcome it.

Next, influence. This is an inherently small and nebulous part of the equation. But it's key for many publications. Many ads aren't trying to sell you anything directly. They're trying to tell you stories, shape your thinking, advocate positions. Political ads are like this. But they're mass market since obviously everyone can vote - at least in states without Republican governors and Secretaries of State. But where the money is is with people who are considered influential in various communities, so-called "opinion-leaders".

Here's an example. Go to the subways in New York you'll see ads for storage rentals, lawyers, grocery deliveries, breast augmentations, ESL courses. Go to Washington DC and you'll see ads for ... Kazakhstan or Northrup Grumman or PhRMA or well ... you get the idea. There are lots of people who care a lot about what people in the nation's capital think. And yes, TPM very much plays in that ad space. TPM and similar sites lose big on #1 and #2. But #3 is where there's a business that can drive ad sales.
As a marketing statistician, I'd like to emphasize the point about "reasons, some good and evidence based, others silly." Most of the people buying these ads, including high-level executives at Fortune 500 companies, have a very weak grasp of how targeted advertising works.

Posted by at No comments:

Thursday, January 9, 2020

The people who brought us the greatest TED Talk ever


We'll have more on this later.
As of today, the venerable comedy concern CollegeHumor Media is mostly dead. In the words of Sam Reich, its chief commercial officer, IAC “has made the difficult decision to no longer finance us,” he wrote in a Twitter thread this afternoon. “Today, 100+ brilliant people lost their jobs, some of whom are my dear, dear friends.” That means the cuts hit every vertical: CollegeHumor, Dorkly, Drawfee, and Dropout — only five to 10 people are left, according to Bloomberg.

The news, however, wasn’t all bad: IAC agreed to let Reich become the humor company’s majority owner. “Of course, I can’t keep it going like you’re used to,” Reich wrote further down in the thread. “While we were on the way to becoming profitable, we were nonetheless losing money — and I myself have no money to be able to lose.”
Until then, here's CH taking on the sacred cows of technology.





















And, as promised, the greatest TED Talk ever.


Posted by at No comments:

Wednesday, January 8, 2020

"Cosmic Crisp" is a stupid for a pretty tasty apple

Just had one of these the other day. It was good (though I'll probably stick with the Envy) and the story of the science and economics behind its development is really interestion.

What a 500 million dollar apple tastes like by Keith A. Spencer

Upon first glance, the Cosmic Crisp apple doesn't appear particularly unique nor inspiring. It is larger than the average apple, certainly, but its mottled exterior could be confused for many other reddish varietals. In other words, one would not know immediately that this humble fruit is the pinnacle of industrial agriculture — encompassing hundreds of millions of dollars of investment, two decades of planning, and hundreds of trees bred, tasted, and culled.

The phrase “apple farmer” inspires images of pastoral orchards, straw hats and wooden buckets full of fruit; yet modern agriculture practices fall far from that imagined tree. A Darwinian tasting process brought the Cosmic Crisp into being: in the late 1990s, Bruce Barrett, an apple researcher at Washington State University, picked the Cosmic Crisp out of 10,000 crossbreeds. The Cosmic Crisp, then known as WA 38, ticked all the boxes: its ability to survive in different Washington state climates, its taste and texture, and how long it lasted without decaying. (Supposedly the Cosmic Crisp can last around a year in cold storage.) A consulting firm tested the apple with focus groups, where one participant commented on the apple’s appearance as star-like, which led to the name. As with many modern seeds, Washington State owns the rights, and thus growers must pay royalties in exchange for growing the apple. The apple was so widely believed to be "the apple of the future" (as the New York Times put it) that 13 million trees were grown at a cost of $500 million. To extend the selling season, producers store the apple in a refrigerated, controlled atmosphere of 1% carbon dioxide and 2% oxygen after treating it with fungicides. Instagram influencers were hired to help hype its release, including a retired astronaut.

Posted by at No comments:

Tuesday, January 7, 2020

Tuesday Tweets




This is basically what I had in mind with the Magic of will/belief/doubt section of the magical heuristics thread, specifically doubt.











 
 






 





Posted by at No comments:

Monday, January 6, 2020

Part of my framework for (what we used to call) the nightly news

On the reactions of various GOP officials to and increasingly erratic Trump administration, here is a quick outline of the assumptions I’ve been using. So far they’ve done a pretty good job explaining the situation and they are reasonably consistent with the takes of people like Josh Marshall (which always makes me feel better).

Much of the GOP and most of its base must now be treated as a cult of personality.

As of 2019, GOP elected officials can be broken down almost entirely into two groups: believers and nonbelievers trying to pass themselves off as believers.

Nonbelievers make constant displays of loyalty to trump out of both personal interest and concern for a party.

Trump demands constant praise and lashes out at even mild criticism. Given his control over primary voters, he is in a position to destroy the political careers of any party member coming up for election in the next 2 to 4 years, possibly even longer.

In addition to fear of political suicide, Nonbelievers also have to contend with the two existential threats that Trump represents to the Republican party.

In the short term, Trump could have a massive public breakdown, or act out in such an extreme way that a solid majority of the country (more than 60%) insists on his removal. As previously discussed, that makes it almost impossible for a party to be nationally competitive.

The long-term threat is that the party continues to double down on policies that cost them support from every major demographic group except for rural white men born before 1960.

The first order of business for the nonbelievers is to keep Trump calm at all cost. This is why so many senators and congressmen who had seemed reasonably sane in the past now talk like characters from that Billy Mumy Twilight Zone episode.

The long-term, on the other hand, explains why those who aren’t singing praises are so reluctant to say anything at all, why the desire to spend time with one’s family has reached such unprecedented levels, and why we are starting to see surprisingly regular trial balloons about an anonymous Senate vote.
Posted by at 1 comment:

Friday, January 3, 2020

The greatest film comedy ever made just entered the public domain




And quite a bit more (1924 was a very good year)::

On January 1, 2020, works from 1924 will enter the US public domain,1 where they will be free for all to use and build upon, without permission or fee. These works include George Gershwin’s Rhapsody in Blue, silent films by Buster Keaton and Harold Lloyd, and books such as Thomas Mann’s The Magic Mountain, E. M. Forster’s A Passage to India, and A. A. Milne’s When We Were Very Young. These works were supposed to go into the public domain in 2000, after being copyrighted for 75 years. But before this could happen, Congress hit a 20-year pause button and extended their copyright term to 95 years.2

Now the wait is over. How will people celebrate this trove of cultural material? The Internet Archive will add books, movies, music, and more to its online library. HathiTrust will make tens of thousands of titles from 1924 available in its digital library. Google Books will offer the full text of books from that year, instead of showing only snippet views or authorized previews. Community theaters can screen the films. Youth orchestras can afford to publicly perform the music. Educators and historians can share the full cultural record. Creators can legally build on the past—reimagining the books, making them into films, adapting the songs.







Posted by at No comments:
Subscribe to: Posts (Atom)