Getting pooped out of the pop-ups.

If an app can be a friend, Grammarly was one of mine for years. I can’t recall when I first discovered the program that installed my Oxford commas, kept me from using “their” instead of “they’re,” and reinforced the lesson of my high school English teacher that “it’s” means “it is.”

On the fly, Grammarly redlined my awkward constructions while giving my simple posts a polished appearance. I could not imagine life without her. Yes, she was definitely female.

But as of today, our relationship has ended. I have scrubbed her from my Chrome browser and from everywhere else on my desktop computer. I will see her no more.

Her sin? Greed.

It started with a pop-up ad that couldn’t be shaken or avoided. Everywhere I placed my cursor to type, whether in an email, a Facebook comment, or this blog, an ad pitching Grammarly Pro appeared. The pop-ups were not huge, but they were just large enough to block what I was typing, and relentlessly persistent. I would close one, only to have it reappear with my next keystroke.

I understand that a company cannot survive forever by giving away its product for free. I subscribe to and pay for a number of programs that make my online life more pleasant and productive. I might even have considered paying a modest subscription fee to Grammarly, but the in-your-face advertising, as obnoxious as any street vendor we encountered last week in New York City, was a deal breaker.

So please forgive the occasional lapse in spelling and grammar. I promise to do my best. It won’t be Grammarly perfect, but I will try to make it Raff immaculate, without getting in anybody’s face.

From airlines to AI, does the decision really matter?

I might as well flip a coin.

American or United to New York. Home Depot or Menards for duct tape. Substack or WordPress. Different decisions, same result. It often feels interchangeable.

Now I’m facing another choice, and I’m not at all sure it will turn out any differently.

I’ve been using ChatGPT for over three years. It’s helped with my writing, my consulting work, and even the occasional village issue. It’s fast, agreeable, and, if I’m being honest, rarely tells me I’m wrong.

But after a bike ride the other day, our daughter Laury said I should try something else: Claude.

Laury likes the way Claude helps her pull spreadsheets together and turn them into clean, usable reports. I can see the appeal. When I was running laboratories, that kind of help would have saved hours. Sometimes days. It often felt like it took a week just to assemble a monthly report from all the different sources.

But that’s not my life anymore. Even the consulting I do now is lighter, less data-driven. I’m not merging spreadsheets or tracking trends across thousands of cases. I’m writing. Thinking. Revising.

So the question, for me, is simpler. Which one helps me do that better?

I know what ChatGPT gives me. The tone, the speed, the steady encouragement.

Now I’ll see what Claude does.

Maybe this turns out to be another one of those interchangeable choices. Door #1 or Door #2.

How a quiet Canasta game turned me into a Joker

The rec room is warm and humid, carrying the faint aroma of sugary snacks. At five scattered card tables, automatic shufflers whir, and plastic tiles clack. Players study their hands, waiting. Cries of “Crack!” and “Bam!” and “Splash!” erupt from every corner. There are twenty-five Canasta and Mahjong players. Twenty-four women, and me.

I volunteered to play Canasta with some of Barb’s friends, helping fill out a table at their weekly game. I played a few hands with Barb before she rotated to another table, but now, for the first time in my brief Canasta career, I have a different partner.

I’m holding my own, following the rules and keeping a reasonable pace. I’ve made one mistake, discarding a card that should have gone into my partner’s meld, but she graciously waved it off. I hope it doesn’t cost us the game. Then again, we’re only playing for fun, not money. Not even pennies.

The game continues. I bend forward to pick up my next card, and suddenly something seizes in the back of my right leg. The pain is immediate and intense, and I rise in agony. I steady myself against the table as my partner calls out to Barb, “I think Les needs help!”

Barb sees my face, then my hand reaching for the back of my thigh. We both know what this is: a muscle cramp, severe enough to stop me in my tracks. With her years of experience as an occupational therapist, she also knows exactly what to do.

She steps behind me and places one hand on my shoulder. With the other, she presses firmly into the back of my thigh, working the muscle with steady, practiced movements. We rock slightly as I try to ride out the pain.

I realize the room has gone silent and look up to see twenty-three women staring at us in amazement. I can almost hear what they’re thinking: Are they really doing this here?

Barb senses it too. “It’s ok,” she says. “He’s my husband. And it’s only a cramp.”

The pain eases enough for me to send Barb back to her table. I’m not ready to sit yet, so I play standing. In this awkward position, I knock the refreshments off the table behind me.

As I do, the woman next to me shakes her head and says, “Can’t take him anywhere.”

And I know the others, Barb included, agree.

What the two of us have in common.

In last week’s blog post, I called myself out for becoming too dependent on AI in my writing. While I will still use various “other brains” for proofreading and grammar refinement (skipping the 5th grade leads to various holes in grammarian knowledge), ChatGPT and others will no longer suggest topics, provide outlines, or come up with clever quips to fill my word quota.

It looks like my post has been inspirational to other writers. Today’s New York Times features a Guest Editorial by Colson Whitehead. The title of the Pulitzer Prize-winning fiction writer’s piece: Don’t Use A.I. to Do This. The “this” in question? Creating. As Mr. Whitehead says:

Write the piece, not the prompt.

Of course, the author of The Underground Railroad and The Nickel Boys says it a lot more elegantly and humorously than I ever could. In his defense of creativity, he blends in references to Dostoevsky (yes, I had to look up how to spell the name), Edgar Allan Poe (did not have to look up), and the intro to Law and Order. Even with the help of 6th-generation super-advanced AI, I don’t think I could accomplish that particular weave. I don’t even think that Donald Trump could.

I’m glad to know, or at least hypothesize, that my post inspired an author who has been listed for every existent literary prize since the invention of the printing press to write his essay. Or at least to know that great minds think alike.

Mr. Whitehead will go back to writing all his great novels, and I’ll continue my somewhat shorter blog posts. As for ChatGPT, Claude, Gemini, and all their kin? Go write some code and leave us creators alone!