On Being a Cruciverbalist

According to Merriam Webster, a cruciverbalist is “a person skillful in creating or solving crossword puzzles.” I started doing the New York Times crossword occasionally several years ago — typically just once or twice a week. Since the pandemic started, I have usually done each day’s puzzle from Monday to Friday.

The easiest puzzles are published on Monday, and they get harder as the week goes along. Originally, I found Friday puzzles so hard that I wouldn’t even attempt them. But “practice makes perfect,” as they say, and eventually I didn’t find them to be so scary. You just have to find those one or two clues that you can answer to get started, and then keep working at it. On the harder puzzles, I will sometimes set them aside if I can’t finish; an answer will often pop into my mind the next day when I go back to the puzzle.

I occasionally use the online Times Crossword app, but I mostly like to solve the puzzle with pen and paper. (Like many people, I think I probably spend too much time staring at screens.) It seems that others are sometimes impressed that I’m doing the puzzle in ink. But of course, sometimes I make mistakes — I just have to write a different letter in the square. Alas, make more than one mistake in the same square, and the mess probably becomes unreadable. For harder puzzles, I often write answers in faintly, until I’m certain that they’re correct.

Last year, I decided to try creating a crossword puzzle, which I submitted to the Times. Now a year and a half later, my puzzle has not been accepted. I figured that my chances of publication were pretty slim; I understand that they get about 200 submissions per week. And of course, they only publish seven each week.

It turns out that creating a crossword puzzle is hard! I didn’t keep track of the time spent, but I’d guess I spent 20 to 40 hours on it. If I needed to change one letter to make an Across word, it also affected a Down word (and vice versa). Sometimes I had to undo several words and start over on a particular section of the grid. After getting the entire grid of letters done, the writing of clues was not nearly so time consuming.

A friend of mine tried the puzzle and thought it was too easy. I doubt that is the case, as there is a very wide range of difficulty during the course of the week. My concern is more regarding consistency: is the difficulty level about the same for all the clues? However, I understand that the editors can adjust clues as necessary to make it consistent.

Assuming that the puzzle is not going to be published, I have posted it on my website, along with the answer key. You can find it on my puzzle page at July4.net/puzzles if you want to try it.

I have also enjoyed doing the Wordle puzzle nearly every day for over a year now. I have spent probably too much time analyzing the puzzle! I have looked at letter frequencies and usage of letters in various positions, in order to pick the best starting word, etc.

I also created a tool on my website where you can analyze a Wordle play; you can find it on my puzzle page. Type in the list of guesses, and it shows you what words are still possible after each guess. I check every day after I have solved the puzzle, and often find that there are more valid words at a particular point than I had realized.

April 11, 2023

“What Do You Do?”

“What Do You Do?” That’s a common question when you meet someone new at a social function. When I worked full-time in New Jersey, I would say that I was a software engineer, and of course, provide more elaboration depending on the circumstances. When I lived in Champaign from 2006 to 2017, I was generally “semi-retired” — sometimes working part-time. I often supplemented my answer with stories from my various volunteer activities that were mostly tied to my alma mater, the University of Illinois.

(By the way, in 2005, after I had left my career job at Telcordia Technologies, and was making plans to move from New Jersey to Illinois, I met an older gentleman on a vacation trip. Rather than the usual question, he asked me “Do you work for a living?” That’s the only time I recall it being asked that way. It happened to be quite appropriate for me, as I was in the midst of planning a transition, and not working at that time.)

I’m now fully retired, in the sense that I plan to never again work for a paycheck. Since moving to Chicago in March 2017, I’ve been enjoying the new experience (new to me!) of living in the big city, but I’m sometimes unsure how to answer the question. What do I do? (I have a friend who doesn’t like the word “retire” because it sounds like you’re just sitting in a rocking chair waiting to die. I’m not doing that!)

Because I don’t go to a job every day, I should have 40 hours a week of more free time (not to mention commuting time, and those occasions when the job took more than 40 hours). Sometimes I wonder — where does the time go! So, on a chilly rainy afternoon in April, I decided to figure out what I do, by looking at the data. I went through my calendar for the twelve months ending on March 31, 2018, and here’s what I found.

• I was away from home, on seven trips, for a total of 34 nights (two trips of more than a week each, and five short trips). Although many retirees talk about doing a lot of travel to “bucket list” places, my travel consisted of visits to out-of-state friends and family, and some shorter trips back to Champaign. The one trip that might be considered “exotic” was a two-day trip to southern Illinois to see the total solar eclipse.
• I attended a total of 45 performances during the year (including concerts, plays, standup comedy, and the occasional lecture). About a third of these performances were at the little jazz club in my neighborhood.
• I attended 25 events associated with the University of Illinois — some were on my visits to Champaign, and some in the Chicago area, including athletic events and other alumni activities. (In order to avoid double counting, this does not include performances by UI groups.)
• There were a total of 22 visits to museums and similar places, about a third of which were to the nearby Art Institute. And I enjoyed a total of 17 Scrabble afternoons, typically playing three or four games each time.

This list doesn’t include a number of other social occasions — lunches, dinners, and so on. I also was riding my bike on the Lakefront Trail (during warmer weather), as well as doing a lot of walking around the city. And of course, there are still mundane activities like grocery shopping and doing the laundry. Retirement isn’t all fun and games!

Like many people, I “waste” some time every day looking at Facebook and other online stuff. I also maintain an overly elaborate spreadsheet for estimating my taxes and optimizing my participation in the Obamacare health insurance market. (One friend told me that’s where all my time goes!) Oh, I almost forgot — I write an article for this blog, once in a blue moon. Although considering how (in)frequently the articles appear, I don’t think I can answer the question by saying “I’m a blogger.”

April 25, 2018

A Mathematical Flaw in the ACA

The Affordable Care Act (ACA, often referred to as “Obamacare”) established a government-run marketplace for health insurance, to be used by those Americans who do not receive health insurance from an employer. The Act also provides for a Premium Tax Credit for lower and middle income taxpayers to help pay for health insurance. (Taxpayers near or below the Federal poverty line are offered Medicaid and cannot use the tax credit.)

The tax credit is designed so that taxpayers pay an increasing share of their health insurance premium as their income rises. When income exceeds four times the poverty line, the taxpayer is no longer eligible for the credit. Unfortunately, there is a serious mathematical flaw in the design of the credit for some taxpayers. (It should be noted that the tax credit computation is based on the cost of the “Second Lowest Cost Silver Plan” in each geographical market. Insurance premiums are also based on age, so everyone’s situation is different. The flaw described in this article may not apply to every taxpayer.)

The following chart shows the net insurance premium (after applying the tax credit) that would be paid by an example taxpayer, for the calendar year 2017, at various income levels. The red line in the chart, and the green entries in the table, show the annual insurance premium. The data was obtained from the government’s HealthCare.gov website, for a 58 year old resident of Champaign County, Illinois, buying the least expensive policy offered in that market. (For comparison, the blue line shows the amount of Federal income tax, assuming that the taxpayer takes the standard deduction; more about that later.)

With an annual income of $20,000, the subscriber would pay almost nothing, as the tax credit would cover nearly the entire insurance premium. At an income of $40,000, the annual premium is $1855. And at an income of $50,000, the annual premium payment is $7466 because the taxpayer is no longer eligible for the tax credit.

That “cliff” in the chart indicates the serious flaw in Obamacare. With an income of $47,519, the premium is $2584. Add one more dollar of income, and the premium jumps to $7466 — an increase of over $4800!

As everyone knows, the income tax increases with income. But there is usually no scenario where a $100 raise would result in the tax bill increasing by more than $100. Unfortunately, if this taxpayer’s salary increases from $47,500 to $47,600 (a $100 raise), he will see an increase of over $4800 in the insurance bill. It would actually make sense for this worker to refuse the raise, or not take on extra hours in December, or not make those last couple of sales.

Because the prices are different from place to place, and also vary by age, I don’t know how many people could be affected by this problem. Still, I’m surprised I haven’t seen it mentioned in news reports more often. There are some avoidance strategies for people who could hit this cliff. For example, eligible contributions to an IRA account or a Health Savings Account can reduce a taxpayer’s Adjusted Gross Income. (This article discusses the cliff and some strategies for dealing with it.)

I can only assume that the designers of the Affordable Care Act didn’t intend to impose this cliff on us. However, it stands as another example of the unintended consequences often created by the political process.

September 21, 2017

What Year Is It?

Happy New Year!  Maybe you’ve accidentally written 2016 on a check or other document recently, but my headline question isn’t really a stumper — it’s 2017, of course.  (Although, they say that after you put something out on the internet, it stays there “forever.”  Maybe it’s a different year by the time you’re reading this.  But I digress.)

What I’m really wondering is how you read the number.  Was it “two thousand seventeen” or “twenty seventeen”?  I hear the names of years spoken in each of these ways.  Does it matter which way we say it?  Is one way better than the other?  Or is it just a case of “you say puh-TAY-toe, I say puh-TAH-toe… tuh-MAY-toe… tuh-MAH-toe,” and so on (as suggested in the lyrics of that old jazz standard “Let’s call the whole thing off!”).

I prefer the latter pronunciation.  If I said I was born in 1968, I believe nobody would read that as “one thousand nine hundred sixty-eight.”  Of course, it’s “nineteen sixty-eight.”  We could have simply transitioned from “nineteen” to “twenty” as we moved into the new century.  To my ear, this second pronunciation sounds better, and just flows more easily off the tongue.  (By the way, if I said I was born in 1968, that would be a lie — er, I mean, a typo.  But I digress.)

Also consider this.  The second form of pronunciation contains only five syllables, whereas the first version has six.  So if you follow my suggestion, you save a syllable every time you name a year.  It’s more efficient.

(If you have thought about this as much as I apparently have, you have noticed that the efficiency argument didn’t help in the first decade.  “Two thousand seven” and “twenty oh seven” both have the same number of syllables.  And worse, “two thousand” is shorter than “twenty hundred” — but that case only arises once per millennium.  Besides, we were concerned with the Y2K bug back then; nobody was worrying about how to say the year.)

Finally, and I suppose I’m getting a little ahead of myself here, but let’s think about 2117 — it’s only 100 years away.  Will we say “two thousand one hundred seventeen” or “twenty-one seventeen”?  In the 22nd century, we could be saving three syllables each time!  So we might as well start using the best pronunciation now.

January 18, 2017

America’s Unity Candidate

I really don’t like politics, but I feel the need to comment on this election cycle, which many agree has been crazier than most elections.  The Democratic Party has nominated Hillary Clinton, and the Republican Party has nominated Donald Trump.  It has been widely reported that both candidates have unfavorable ratings in the range of 50% to 60%, or even higher, making them the most unpopular major party candidates in a very long time.

Meanwhile, the Libertarian Party has nominated Gary Johnson.  Despite the fact that he has received considerably less media attention than the major party candidates, there is still about 10% or more of the electorate saying they would vote for him.  Probably as a result of wanting more information, polls have indicated that a majority of voters wanted to see him included in the first presidential debate on September 26.  Unfortunately, the debate commission decided not to include him.

It seems that a lot of voters feel the need to vote for Trump, just to make sure that Clinton is not elected.  And similarly, a lot of voters will pick Clinton, just to avoid a Trump presidency.  Of course, there are some voters who truly believe their favorite is a good choice.  So as a thought experiment, let me make up some numbers here.  Let’s suppose 20% of the voters really want a Clinton presidency, and 20% really want to elect Trump.  And as I’ve noted, let’s say 10% are supporting Johnson.  That leaves 50% of the voters in the “reluctant voter” pool, who are mostly voting against a “bad candidate,” rather than for someone they’re enthusiastic about.

I suspect it’s almost impossible to find someone who likes both Trump and Clinton.  This means that Johnson is at least the second choice, if not the first choice, for almost everybody.

Here’s an idea.  I think Gary Johnson would make a great “unity candidate.”

What if the 50% (the “reluctant voters”) choose Johnson instead of a major party candidate?  He would be at least their second choice.  Add that to the 10% support he has now, and he wins the election in a landslide.

(Actually, we don’t need a landslide.  Maybe Johnson gets 40% and the others 30% each.  After he completed two terms as governor of New Mexico, Johnson climbed Mount Everest in 2003; surely he can get to 40% by November 8!)

Let’s compare the results.  If Clinton or Trump wins, there will be a large number of Americans who think we have elected one of the worst presidents ever.  And the new president will start off with an unfavorable view by a majority of voters.  In that case, I would expect four (or eight) more years of political divisiveness.  On the other hand, if Johnson wins, most people will say he wasn’t their first choice, but he’s not so bad, and at least we avoid the “calamity” of electing Trump/Clinton.

Also consider this.  One hundred and fifty-six years ago, the nation was even more divided than it is now.  We elected a unity candidate from a recently formed third party.  His name was Abraham Lincoln.  And I think that choice worked out pretty well.