Experimenting with AI

Posted on June 11, 2023 by Sakke Myllymaki

I have wanted to experiment with artificial intelligence for some time. This is more because of my professional (non-writing) interest than any real desire to profit from it in my writing. My most recent experiments have been with Chat GPT, a language model that is designed to understand and generate human-like text. As the chat part implies, it uses a conversational interface, i.e. it generates replies to questions and prompts and remembers the context.

Chat GPT is supposed to help generate ideas, drafts, and brainstorming, among other uses. Therefore I tried it to generate ideas for characters. The results have been mixed. My attempt to create a minor character is a case in point. Giving ChatGPT prompts detailing what kind of character I wanted, did provide something close to what I was looking for.

My first impression is the output it generates is extremely verbose. This is, in fact, one of ChatGPT’s known weaknesses, though not one that is often mentioned. For my experiment, finding the usable bits among the verbiage is a challenge. Another limitation is, that ChatGPT is known to write plausible-sounding but incorrect or nonsensical answers. This is of course a problem for fiction writers, too.

Of course, my main objection to using AI to write is that it is not me. I don’t know where all the stuff comes, and it doesn’t sound like me. And I shouldn’t forget AI generated text is not copyrightable.

But there is a different type of AI-based writing too I have been using for some time: Autocrit. It is more directly useful, as it is designed for novelists. I have used it to edit my stories and found it useful for identifying repetition, redundancies, or bland writing. I feel it has improved my writing both by pointing out some of my bad habits, but also by enabling me to focus on the bits in need of improvement.

I have learned though that I must scrutinize most recommendations from Autocrit. I need to use my judgment to decide if I can find a better way of saying what I intend. Sometimes what I wrote is for a reason, for example in dialogue as part of the character’s voice.

Posted in On Writing | Tagged , , | Comments Off on Experimenting with AI

Prices and Historical Fiction

Posted on June 8, 2020 by Sakke Myllymaki

In researching for my stories set in 1920’s London, I have become conscious of the importance of knowing the prices and wages. Not much of this will ever be mentioned, I expect. But I feel it necessary to get it right.

I can, of course, avoid mentioning any exact prices, but it goes beyond that: To create the authentic feeling, I need, for example, to know if my characters would be able to afford something.

I recently read a story set in my period. When the heroine, who was supposed to be more or less penniless, treats fifty pounds as if it wasn’t half years salary for a shop assistant, I felt immediately out of the period. You can’t use an inflation calculator to ‘convert’ prices. The relative prices have changed, as have wages.

How do I find my information on prices? Mainly from three sources: Period advertisements, period books that mention prices, and on-line resources. Of these, the period novel seldom mention prices, but when they do, like Orwell’s Keep the Aspidistra Flying, I make note of it.

Of the others, UK advertisements don’t always quote prices, but when they do, it has given me another insight: many prices are quoted as shillings and pence rather than pounds, shillings, and pence: 36/6 instead of £1 16s 6d. Unless they gave it as guineas as when charging for professional services. That’s why I have Jack Hart quote his fee in guineas.

Posted in On Writing, Rants | Tagged , , | Comments Off on Prices and Historical Fiction

On a Commuter Train

Posted on February 3, 2019 by Sakke Myllymaki 
Years ago, back in my student days, 
The hour was late, I should explain,
When at last I was homeward bound, 
And boarded a commuter train.

I had with me a paperback:
Marlowe’s play ‘bout King Edward’s Reign
(It had been assigned for my class)
To read on the commuter train.

I saw a man across the aisle 
His age uncertain, his clothes plain,
His staring eye fixed on me,
Sitting on the commuter train.

I paid him no heed, held aloof
And hid behind my book, urbane.
Minding my own, that is the thing
To do on a commuter train.

As he passed me by on his way,
“ I think,” the man said with disdain, 
“Earrings do not become a man.”
And left the damn commuter train.

Even though this was the eighties, 
And blokes with earrings, I maintain,
Were only seen on TV screen,
Not riding on a commuter train,

Why he said so, I’ll never know.
He must have earrings on the brain;
I sported no ring in my ear
That day on the commuter train.
Posted in Verse | 1 Comment

What Makes a Good Story?

Posted on November 3, 2017 by Sakke Myllymaki

—that my endeavours should be directed to persons and characters supernatural, or at least romantic, yet so as to transfer from our inward nature a human interest and a semblance of truth sufficient to procure for these shadows of imagination that willing suspension of disbelief for the moment, which constitutes poetic faith.

?Samuel Taylor Coleridge

What makes a story gripping, so that it is almost impossible to put it down? It is not merely sufficient to have a series of interesting or entertaining events, or an important and engaging theme.  So what is it that not only hooks the reader and makes them want to learn more?

When Aristotle said that a story that is whole has a beginning, middle, and an end, he didn’t mean to state the obvious; what he meant is, that within the story everything must be a necessary consequence of some action of the story, contained between the beginning, which starts the chain, and the end, which resolves it.

Another insight of Aristotle’s is, that “the plot, being an imitation of an action, must imitate one action and that a whole, the structural union of the parts being such that, if any one of them is displaced or removed, the whole will be disjointed and disturbed.” Or, to put it somewhat less elegantly, a story should hang together and contain nothing that is not part of the story.

But this is only the structure of the story, it is not alone sufficient to make a story good. The main ingredients in any story are the characters. We the readers are human; it is people we are interested in: their actions, feelings, and ideas. Even if the story is about something other than a human being, it still reflects humanity.

Your character doesn’t need to likeable; Tom Ripley is a conniving murderer without a conscience, but an engaging character nevertheless. We want to follow his progress and see how far he will go to get what he wants, and will he get away with it. Most importantly, he is a convincing character, with complex motives and fears. If the character’s traits seem contrived, the story fails to grip us.

The characters are central to the unfolding of your story; everything should follow from the actions and choices of your characters. This is harder to do than it sounds. Those actions must be consistent with the character. If the character does something without an apparent reason, the reader is at best puzzled, at worst throws your story away in disgust. But as long as the character’s motivation has been established, it doesn’t matter whether he is trying to find out who stole the Brasher Doubloon, or abduct Lord Emsworth’s prize pig, the reader is willing to suspend disbelief.

A story is a work of fiction, an imitation of an action, a semblance of truth. For a story to be engaging, it must ring true, when it isn’t.  It is not enough for the reader to believe in the world of the story: You, the writer, must believe in it as well.

Posted in On Writing | Comments Off on What Makes a Good Story?

My Ten Not Quite Rules for Writing

Posted on June 4, 2017 by Sakke Myllymaki
  1. Start with the idea that a blank screen is your friend; it is there waiting to be filled with words. Don’t let it intimidate you, even if it can have a menacing appearance.
  2. Everybody has some kind of plan before they start to write. Do you make a detailed outline or have just a vague idea of what is going to happen? It doesn’t matter as long as you know what you are doing.
  3. Keep notes of everything. Did you get a great idea at work? Make a quick note before you forget it. Are you sure that you mentioned, say, the colour of your character’s toothbrush, somewhere. I hope you made a note of it (I didn’t – it didn’t seem so important at the time) in the appropriate character sheet, because it is much easier to check that way when you need that information. I use OneNote for this; I can access my notes from all my devices.
  4. Descriptions are hard to do. It’s not just describing something that exists only in your mind’s eye, but you must allow for just enough detail to allow your reader fill in the rest.
  5. Names are very important. I can use months to decide what to call my main characters to bring out something of their qualities. Conversely, when I name an incidental character quickly, they can suddenly gain more importance in the story just because of their name. Here name really is an omen of what the character is like.
  6. Metaphors are important. They are the stuff creativity is made of. You might forget that your whole story is a metaphor.
  7. Google is nice if you need to check something quickly, but don’t use it for procrastination. It’s all too easy to spend the whole evening ‘doing research’.
  8. If you get stuck, try writing something else. This could be another scene or chapter in the same story that you find more inspiring at the moment, or it could be another story.
  9. Raymond Chandler once said: “When in doubt have a man come through a door with a gun in his hand”. It doesn’t have to be a man, and they don’t need to have a gun, or even come through a door; just make something happen. You can always delete it, if it doesn’t fit, but keep on writing.
  10. Sometimes your characters don’t want to do what you have in mind. This is because your plot wants them to behave in ways that are inconsistent with their way of thinking and doing. If it is crucial for your plot, think what might motivate them to behave in the way you want?
  11. Rules are to be broken. Actually, these are not even rules. They are just my ideas, good or bad. Mainly they work for me, at least part of the time.

Posted in On Writing | Comments Off on My Ten Not Quite Rules for Writing