Last Saturday night, after I’d already finished that week’s blog post, a bunch of us gathered at Number Two Son’s house for his birthday celebration. The younger girls were there, and at one point, Middle Daughter announced a disturbing climax to some sorority relationship drama. It affected her personally, and she started crying. I went to the bathroom for tissues, and when I came back to the living room, Number Two Son was kneeling beside her, doing his best to hug the tears away.
My heart melted.
My brother and I were never close. We had a couple of knock-down fights in our younger years, and these days I haven’t spoken to him in at least ten years.
I haven’t seen him in person since 2010.
I wish I knew why.
I’ve said before I’d give my eyeteeth to have the kind of relationship my dad had with his brother. I know it’ll never happen. But I wish it could.
My kids know how much that’s hurt me over the years. I’ve begged them to make sure they communicate with each other, even when they’re pissed at each other. I don’t think I could handle it if they ended up the way my brother and I have.
I don’t know that I can put into words how much it meant to me that he comforted his sister the way he did. More importantly, Middle Son and Youngest Son did their part, too.
I know it was Number Two Son’s birthday, but I don’t think I could have asked for a better gift. Yeah, it absolutely sucks that Middle Daughter was hurt the way she was by someone she should have been able to trust, but I was thrilled with the way family stepped in to take care of their own.
Folding
One of the first things I did when I set up HemisphereDance was install Folding@Home.
I’ve written a couple of times about Folding. It’s a tangible way to contribute to various research causes without having to worry about where your money is going.
The new machine has a little more horsepower than the laptop did and it’s not ideal to run F@H on a laptop anyway. They generally don’t have the power to crunch all that effectively. Then again, the old MSI was designed as a gaming rig, so it was a little beefier than your average laptop.
After running cancer research for a while, I moved over to diabetes, which is obviously a big deal for me. But this week I switched my primary project back to Alzheimer’s, which has been my main cause for a long while. It’s one of those things that scares the daylights out of me, even more than cancer does. I think it’s because there’s basically nothing we can do to prevent it and damned little we can do to treat it. The idea of being lost in your mind just terrifies me.
Learn more about F@H here, and if you’re inclined to sign up and join a team, mine is the 234447, the Quiet Professionals.
Writing
I started using generative AI a while back to help me create headlines and social media posts. I knew I’d never use it beyond that because I don’t believe in using it beyond that. People who use it to create stories aren’t writing, IMO. Part of writing is sharing your experiences from life, and generative AI can’t have life experiences. It’s never seen a sunset or watched a lover say goodbye or held someone’s hand as they died.
It’s not alive.
Writing headlines though is something I don’t necessarily do well. These days it’s less about making something funny as it’s about trying to satisfy an algorithm so your post gets shown to more people. That makes it somewhat mechanical. Formulaic, even. That’s almost a perfect reason to use generative AI for a headline.
I’ve justified my use by saying that I’m willingly allowing my post to be used by whatever Large Language Model I happen to be using – mostly OpenAI’s ChatGPT. I’m aware that the post could be used to train the LLM. Indeed, much of my blog already has been. I can’t recall where I found the search tool, but I remember in the fair use/pirating uproar shortly after ChatGPT came out finding a place that cataloged what sites had been used to train OpenAI’s bot, and this site was listed.
Now there’s a new series of complaints, this about the LibGen dataset that’s been used to train various LLMs including Meta’s Llama. The Author’s Guild is running a class-action suit against a number of companies who have used the Library Genesis dataset, claiming copyright infringement. A reporter at The Atlantic created a tool to search book titles, and I found The Sad Girl listed there.
I’m not surprised one of my books was used.
Can I be slightly miffed that only one of my books was used? Is that a legitimate complaint? I’m curious too as to which edition was used. I self-published it, then Booktrope released it, then they closed and I re-released it. So I kinda want to know what version got pirated.
And it was pirated. I never gave any entity permission to use any of my books to train LLMs. The companies involved are claiming that using books this way is a transformative use, thus exempt from copyright law.
You can read The Atlantic’s coverage here.
I’m not buying the transformative argument, personally. I don’t see how whatever these companies did with The Sad Girl transformed it.
And the more I learn about how AI models are created, the less inclined I am to use them, even in the limited fashion I have been. We’ll see.
Thanks for reading! Feel free to share a thought in the comments. Sign up for my infrequent newsletter here. Find some of my other writing at The Good Men Project, too. Subscribe to the blog via the link in the right sidebar or follow it on Mastodon. You can also add my RSS feed to your favorite reader.
Share your thoughts!