Friday, July 26, 2019

Why a Lottery Scam Should Inform Our Use of Electronic Voting Machines

Eddie Tipton, the former information technology manager for the Multi-State Lottery Association, was convicted in 2017 of rigging lotteries. The organization he worked for provides number-picking computers for lotteries in 33 states, the District of Columbia, Puerto Rico, and the U.S. Virgin Islands.

He concocted a simple scheme: he added code to the computer programs that run the lotteries so that on certain specific days instead of producing a random draw of winning numbers, the program picked numbers known to him ahead of time. He had his brother and friends buy winning tickets and shared in the “winnings.”

The first known instance was the Colorado Lotto in 2005. The last payout was 2011. The one that triggered his arrest was a $14 million payout from the Iowa Hot Lotto in 2010. And that happened more by luck than anything else.

What does this crime mean for electronic voting machines?

We must realize that what Eddie Tipton did while working for the Multi State Lottery Association someone else could do working for a voting machine manufacturer

The lottery scam presents a parallel argument that all electronic voting machines should have a paper backup so voters can confirm the machine has correctly tabulated their vote and election officials can perform an accurate recount or audit.

Here is one example in which someone who wanted to affect a presidential election could make a difference:

Several states, including Pennsylvania, allow straight ticket voting in which by making a single choice, the computer records you as voting for the candidate of your chosen party. Although Pennsylvania law requires new machines to have paper backup, most old machines do not. Imagine if some programmer had the foresight a decade ago to stick in a little bit of extra code in the programming that changes one of every 100 straight party votes from Party A to Party B for the 2020 election only.

Pennsylvania is a swing state . . . that small bit of code might be enough to change who is elected president.

Don’t want to change votes, how about not counting all the votes? Not counting one of every fifty straight-line Party A votes accomplishes the same result.

This fantastic scenario requires a single programmer in the right spot at the right time, just as Eddie Tipton was in the right spot at the right time.

Hackers with access to voting machines can accomplish the same thing on a more targeted basis. They can commit after-the-fact voter suppression, say by rigging machines to ignore one of every 100 votes in certain districts that vote primarily for Party A and maybe flip an election to Party B. Oh sure, comparing the number of voters to the number of votes cast could uncover the irregularity, but then what? Would the courts order an entire new election? Would voters lose confidence that their votes mattered?

The opportunities are myriad. We must have secure elections. We should require electronic voting to have paper backup. If Congress were serious about preventing election fraud, it could purchase electronic voting machines with paper backup for every voting precinct in the nation. Would it cost money? Yes. But Congress has no problem increasing our budget deficit, and at least we’d get something for the money spent.

* * * * *

James M. Jackson authors the Seamus McCree series. Full of mystery and suspense, these thrillers explore financial crimes, family relationships, and what happens when they mix. False Bottom, the sixth novel in the series—this one set in the Boston area—is now available. You can sign up for his newsletter and find more information about Jim and his books athttps://jamesmjackson.com.


Monday, July 1, 2019

Amazon, Authors, and Asymmetric Information


By James M. Jackson

Asymmetric information occurs any time two parties to a transaction have different levels of knowledge about a situation.

Sometimes that’s a very good thing: When I go to the doctor with bothersome symptoms, I expect her to know a lot more than I do about diagnosing the problem and how to treat the underlying issue.

Sometimes it is not a good thing: for example, when you buy a used car from a stranger, the seller knows much more about the car than the buyer does. The seller knows whether the car had all its routine maintenance, whether the son red-lined the engine drag racing with friends, whether it sat for a week in a flooded garage. Unless you are a mechanic, or hire one to inspect the car, you will suffer from asymmetric information.

Let’s Focus on a self-published author’s relationship with Amazon when using KDP (Kindle Direct Publishing).

What does Amazon provide to self-published authors who use KDP (Amazon’s publishing platform)?

You know your royalty rate depending on choices you make about pricing your book and what markets you want Amazon to sell into. There are no hidden costs.

They provide information about your sales—it’s not always 100% accurate, but it’s reliable, and it’s current (at least reasonably so) to within a day.

If you enroll your books in Kindle Unlimited, you can determine the number of pages members read, again daily, with good accuracy.

Amazon provides these data sooner than what an author receives from traditional publishers.

Where does Amazon use its Asymmetric Information to its advantage?

Book Rankings: No one knows exactly how Amazon determines book rank. Sales have something to do with it: we’re sure of that. But sales over what period? What effect does sales momentum have? Sales memory (past sales)? Pages read by members enrolled in Kindle Unlimited also count toward a book’s ranking, but we don’t know how. Does your book share a level playing field with big publishers? How about with Amazon’s own imprints? (i.e. How level is the playing field level?)

One example of a non-level playing field is this: If you give your book away, Amazon assigns a separate ranking for free books. However, free downloads of Amazon Imprint books included in their “First Reads” promotion count the same as sales and are not segregated into a separate class, meaning these books can (and do) obtain the coveted #1 status in their categories.

Search Results: When you search for a book on Amazon, how does Amazon determine which books to show? Do they favor Amazon imprints? To what extent do they favor those who advertise? Do they look to maximize possible Amazon profit on a book sale?

For example, does Amazon use an algorithm that calculates their revenue for a book purchase, multiplied times the probability someone will purchase the book after seeing an ad + the profit earned from the ad itself?

Amazon does “manipulate” results to reflect its interests: recently some authors have entered their book’s name in Amazon’s search box and discovered the first page of results did not include any of their books. Most people don’t click past the first page, which means when that happens the affected author’s books become nearly invisible, even to people who specifically searched for the author.

Amazon tells authors to choose seven useful keywords to categorize their books to help readers find them when they search. Yet only Amazon knows how they use this information and what they combine it with when they deliver search results.

Amazon usually shows “Also Boughts” relative to each book. How are these determined? Is it a level playing field (i.e. do Amazon imprints get included more often as also boughts that other books?) Do books with ads automatically get different treatment?

When authors advertise on Amazon, the asymmetry becomes worse.

Only Amazon knows how it determines how much to charge for a click.

Only Amazon knows how it determines which ads it presents and where.

Amazon knows exactly who clicked on your book’s link—you don’t.

Similarly, Amazon knows which book everyone bought—you have no idea who buys your books.

What’s an author to do?

Knowledge and boycott are the primary tactics to counter asymmetric information. If you choose to self-publish, Amazon is too big a marketplace to boycott, leaving knowledge as your only choice. Knowledge takes two forms: First one should understand where asymmetric information exits (and hopefully this blog helped). Then, to the extent possible, learn strategies to counterbalance Amazon’s advantages.

It’s not just authors . . . readers, too!

Amazon’s asymmetric information advantage also affects us as readers. It knows what we read (if we use a Kindle or Kindle App, it even knows when we read). It knows what books you’ve searched for, what ads you click on, and which ad placements attract your attention.

Not that I want you to become paranoid about asymmetric information, but based on your purchases, Amazon might even know if you are naughty or nice–oh wait, that’s Santa Claus—and besides, you already know that about yourself.

* * * * *


James M. Jackson authors the Seamus McCree series. Full of mystery and suspense, these thrillers explore financial crimes, family relationships, and what happens when they mix. False Bottom, the sixth novel in the series—this one set in the Boston area—is now available. You can sign up for his newsletter and find more information about Jim and his books at https://jamesmjackson.com.


This blog was first published on Writers Who Kill.