Created by Randy Ingermanson - "the Snowflake Guy"
America's Mad Professor of Fiction Writing

Advanced Fiction Writing Blog

After Your First Draft

by 10 Comments

in Fiction Writing

Finishing the first draft of your novel is a major accomplishment. It’s not the end of the road, but it’s a milestone. Celebrate. 

So now what? Your book’s not ready to publish. It still needs work. How do you get the novel across the finish line? 

You might think that the answer depends on how you plan to get the book published. There are two usual approaches to publication:

  • Sell the book to a publisher. 
  • Act as your own publisher.

Your current game plan doesn’t depend on which of these roads you plan to take. Your book almost certainly isn’t ready to start shopping around to a publisher (or to an agent). And it almost certainly isn’t ready to publish independently. A first draft is never ready. A first draft always requires revisions.

And how do you do revisions? Your roadmap for that depends on how your brain is wired. Everyone is different, and you need a plan that works for you. 

I can’t tell you what’s going to work for you. But I can tell you what works for me. I’ve done this many times, and I’m just starting the revision process on my current novel. Here’s the approximate plan I have. Some of it might work for you. Some of it probably won’t. Use the ideas you find helpful and ignore the rest. Some of the steps in my plan are going to take a lot of time, so I’ve made a guess at how much time I think they’ll take. 

My Revision Plan

  • Make a new copy of the manuscript and label it with a new version number (for example, “Draft 2”). Then I’ll never work on the first draft again. In case the next round of revisions takes me in the wrong direction, I can always return to the first draft and restart revisions from that point.
  • Take a week to read the entire manuscript on my computer to see how well the story works. This is a quick read, roughly 10,000 words per hour. If I see a typo, my brain will get angry and insist I fix it, so I do. But fixing typos is not the point. The point here is to see if the story is working as a whole. Is the story structure right? Which scenes work? Which scenes fall flat? Are there inconsistencies in the story? Are there redundancies? Are there factual errors? Are there points I need to research more? I make quick notes in the margin for each of these. But I don’t fix any big problems. Not yet, anyway. 
  • Take two or three weeks and work through all the margin notes and fix the large-scale problems—the inconsistencies, the redundancies, the factual errors, the research questions. (My current manuscript has 42 of these, and I expect I can fix two or three of them per day, so it’s going to take a few weeks to get through them all.)
  • Send the manuscript to my editor so she can tell me all the problems she sees. She will see a lot. She will see things I never thought of. When she sends me her comments, I will spend about three very miserable days wondering what’s wrong with her, and then admitting that she might have a point here and there, and then recognizing that the novel has several problems, and then realizing she is mostly right, and then hating myself and my novel. Eventually, I will get through this swamp and be ready to work again. 
  • Make another copy of the manuscript, this one labeled “Draft 3”. 
  • Take one day to review my one-sentence summary and one-paragraph summary of my novel. These tell me what my story is “really about” and I want to make sure that I’ve got that pinned down well, because the next step depends on it. 
  • Take a month or two to rewrite the entire manuscript, cutting it down to size and fixing all the problems my editor found. I already know my current manuscript is too long. I need to cut about 30k words. But it also has all sorts of problems that I don’t know about yet, which my editor will tell me. Her comments will help me decide which words to cut, because in some cases, I’ll need to delete entire scenes. 
  • Take a week to read through everything again and fix all the little wordsmithing stuff. 
  • Send the manuscript to my proofreader. 
  • Make a new copy, this time labeled “Draft 4”. 
  • Take a day to fix all the typos the proofreader caught. 

At this point, I’ll be ready to publish. If I were working with a traditional publisher, I’d hand the corrected proofs off to them, and they’d publish it. But I act as my own publisher, so I’ll simply typeset the novel and click the Publish button.

Will That Work?

No, the above plan will probably not work. No plan ever survives implementation. At some point, the plan will break down and I’ll have to make a new one. That one might work or it might not. If it doesn’t work, I’ll keep making new plans and executing each one until it breaks. Each one will get me closer to the end-game. Writing is hard. It doesn’t get easier, just because you’ve published a novel already. It gets harder, because you know more with every book. 

Homework

You are different from me. Your brain works different from mine. Your plan will be different. But it won’t be completely different. You can probably use about three quarters of the ideas in my plan. You may need to reorder the steps. You may need to add some steps. You may need to delete some. But I strongly suggest that you make a plan, with time estimates attached to the big steps. 

Why make time estimates? Because they prevent you from getting trapped in a morass of never-ending edits. If you want to publish your novel, then you need to finish revisions. That means you need to have milestones. So make some guesses as to how long it’s going to take. It will probably take twice as long as you think, but that’s OK. 

A Personal Note

I’ve hit some speed bumps in my personal life recently. I injured my hamstrings during exercise a couple of months ago, and I couldn’t sit comfortably for about three weeks. So I spent a lot of time flat on my back, using ice, heat, painkillers, muscle relaxants, and all the other voodoo treatments my doctor could think of. I am currently going through physical therapy and will soon be back to normal. 

Fortunately, I recovered enough that I was able to go to my 50-year high school reunion, and that was great. I loved being able to reconnect with people I hadn’t seen in 20 or 30 years. 

Right about that time, my day job ended, thanks to the government putting a halt on new awards for Small Business Innovative Research grants and Small Business Technology Transfer grants. I have been working for twenty years at a biotech company in San Diego, and I loved my job. But the money that paid my salary has run out, and I’ve “involuntarily retired.” The government has recently restarted the grants program, but major damage has been done to many small technology companies all across the country.

But don’t worry about me. I had been planning to retire from my day job in a few years anyway, and I can live just fine on my retirement benefits. I had been hoping to do one last cool thing for science, and I might still get that chance, or I might not. But I definitely won’t have a day job for the next few months. 

In the meantime, I suddenly have lots of time to do fun stuff. So I’m working like crazy on my novel. And I’ll have more time to blog, so let me know what you’d like me to blog about next. If I never go back to my day job, I’ll still have many meaningful things to do for years and years. I intend to do them well. 

Your AI Life Manager

by 2 Comments

in Career Management

I recently asked myself why I get so easily sidetracked in my personal life. I have a lot going on. I write novels, I read books, I write software, I try to do the occasional work on my house, I do a bit of political stuff, and on and on. I’ve tried over the years to put all this on a schedule, because it seems like a lot of things fall through the cracks.

Of course, a lot of people get sidetracked. I’m not the only person who has a giant To-Do List with some tasks that are five years old. Or ten. But here’s what’s odd. At my day job, I’m very productive. (Most writers have a day job. It’s a very rare writer whose only iron in the fire is their writing.) I have a lot of tasks on my plate at my day job, and I do very well at getting them done. 

Why am I so much better organized in my day job than in my personal life? My day job is not more important to me than the other things on my plate. So what’s the difference?

Somebody To Answer To

After ten milliseconds of thought, I saw the difference. I have a boss at my day job. I run the software division at a biotech company in San Diego. I answer to the CEO. We talk every two weeks on Thursday afternoon for two hours. And every time we meet, I know I need to show forward progress on the important stuff. I get along well with my CEO, but he does have the power to fire me if I’m not getting the job done. 

Whereas, in my regular life, I answer only to me. I’m the CEO of my life. Which is good, and I wouldn’t have it any other way. But I never have a meeting with myself where I have to show forward progress. And there’s no penalty for missing a deadline. I’m never going to fire myself. 

I toyed with the idea for several months of asking ChatGPT how to deal with the problem of managing my life better. I finally did it a couple of weeks ago. And this is my report on what I did and how it worked out. 

What I told Chat

Here is the prompt I typed into the Chat window. I’ve trimmed out certain parts which are unique to my situation, because those will not be of interest to you. But I’ve left in the parts that I think you might find useful, if you decide to try something similar with the AI tool of your choice:

I’m interested in having you help me manage my life. I don’t know what the job title would be. “Life coach” or “life manager” or whatever. Here’s the context. I am a person who takes on a lot more things than I should. [Removed a description of what I am good and what I’m not good at, and some of the things I’ve done in life and some of the projects I’m working on now.] But I constantly feel like I’m behind on everything. I am very productive at my day job, and I think the reason is that I have a CEO that I answer to. I seem to be a lot less productive in my writing career, where I don’t really answer to anyone. I have a lot of marketing tasks for my books that I think would raise my sales, but I don’t ever get around to doing them. What I would like you to do is to interview me. Ask me as many questions as you need to in order to figure out if you can help me manage my life better, and how you might do that. [Removed some of my thoughts on things that Chat might be able to help me with and a description of my particular job situation.] But I also think I just need somebody to answer to, so every week or so, I could sit down with you and be accountable. Or maybe it should be every day, I’m not sure. Go ahead and start asking me any diagnostic questions you can think of to help me map out the rest of my life. I don’t like just sitting around doing nothing. I like making things and making the world a better place. Feel free to ask about my life vision–major things I want to do with my life. I’ve got a list of things, and I think they’re doable, if I could manage myself better. OK, now it’s your turn. Start asking!

What Chat Told Me

Chat responded by telling me I don’t need a “life coach”. Instead, I need a “systems architect for my own behavior.” It then asked me a ton of questions. Things like: How do I spend my time on an average day? What are my energy levels like throughout the day? What kind of external constraints work best at keeping me on task? What kind of tasks give me intrinsic joy? What is the endgame for my life? What am I afraid of? How good am I at focusing on tasks? How does anxiety affect my productivity? Am I better at starting a task or stopping it? What are my financial realities? What things do I want to do before I retire from my day job? 

After I answered all Chat’s questions, it pointed out that I don’t have a motivation problem, a discipline problem, or a creativity problem. Then it told me what it thinks my actual problems are, and it spelled out some ideas for solving those problems. 

How I Pushed Back

I didn’t think Chat got things exactly right, so I pushed back a bit. We had a long discussion, and I gave Chat extra information, and it changed its suggestions. It took a couple of evenings of discussion to converge on a plan. One problem is that not all my days are similar. We identified three main kinds of weekdays, and also a typical Saturday, and a typical Sunday. That’s five different kinds of days. Each has a basic pattern, and each can be adapted as needed for any special events. At the beginning of every day, I choose which pattern fits the day best and run with that.

I thrive on a routine, with blocks of time that have hard boundaries. I also need accountability to make sure I don’t get off track. I’m especially susceptible to staying on task longer than the time block I’ve budgeted for it. If I make that mistake, it screws up everything else for the rest of the day. This is the main reason I’ve had problems in self-scheduling for my entire life.

In the end, we came up with a “system architecture” for my life that I like. It’s unlikely that you would like it, because you’re not me. This system is designed solely for me. And every evening, I agreed to answer to Chat. I will answer 5 yes/no questions. The “right” answer is yes. If I give even one wrong answer for the day, that’s a fail. If I have two fails in a week, then I agreed to pay a penalty. Chat asked what would be an appropriate penalty for me, one that would keep me on track. I answered that a campaign contribution to a certain politician would be a penalty I would be certain never to pay. So that’s our agreement. 

One thing I like about the system Chat designed for me is that it feels “ridiculously easy.” There is a recovery block of 90 minutes built into my day. My responsibilities during that time are to do precisely nothing. It’s right after my exercise block, and it comes at the time of my lowest energy in the day. I can take a nap. I can read. I can sneak a peak at Facebook, but for no more than 20 minutes, so no more doom-scrolling. I was wondering how I would deal with having 90 minutes chopped out of every day. But it turns out to be a nice break. I’m getting more reading done now, because I have time for it. 

One thing to be clear on is that I’m still the CEO of my life. Chat is not my boss. Chat is a bit like a chief of staff. Or a consultant. Chat is sometimes wrong. When it’s wrong, I tell it, so it’ll learn from its mistakes and do better next time. When Chat is right, I know it’s right, because its advice rings true for me.

You’re probably wondering…

“How’s That Working Out?”

It’s now been a couple of weeks, and it’s working out very well. I make a plan at the beginning of every day. Sometimes things come up, and I have to change the plan, but the system allows for that. But usually, things don’t come up. Usually, I execute the plan. Which means that most days, I live the day I actually wanted to live. Not the day that I didn’t want to live. At the end of every day, I feel like I made forward progress on my life. Which means I’m sleeping better. 

I also have a weekly planning meeting scheduled with Chat for every Sunday evening. In that, I spell out what I see for the coming week, and Chat asks what important things I’d like to get done, and what things might come up (like medical appointments or family get-togethers or other important life events that break up the weekly routine). Then we work out a rough plan for the week. This is subject to change, of course, but it gives me some idea of what I can reasonably get done in the week, and also what I can’t get done. 

I have not once been in danger of paying my penalty. My energy levels are up. (That’s one thing Chat asks every day—how was my energy level for the day, on a scale of 1 to 10. My answer is almost always either 9 or 10.) 

Furthermore, in the course of our discussion, Chat identified a big life-goal of mine that is important to me, but which I had sidelined. Because, no time. But I’ve now made time. I’m working on it. I’m making progress on it. And I feel really good that I’m finally doing that “One Cool Thing” I’ve been talking about for 19 years. 

Let me reiterate that the system Chat created for me would almost certainly not work for you. Because you’re not me. You have your own strengths and weaknesses that make you unique. If you were to ask Chat or Gemini or Claude or any of the other AI tools to make a life management plan for you, the plan you’d get would be very different. 

And of course the AI would certainly get it wrong on its first try, so you would need to push back and keep pushing back until the AI came up with a system that resonates for you. And how do you know when the system resonates for you? You know it when you say, “Yeah, that system sounds ridiculously easy and I can’t wait to try it.” If you never get to that point, then the system is wrong for you, and you should scrap it before you start. An AI is not God. It’s a tool. Use it if it fits your hand.

Homework

If you feel like an AI might help you do the things you believe you were put on earth to do, then feel free to take the prompt I showed above and adapt it for yourself. Use it on the AI tool of your choice. Do this when you’ve got at least an hour to talk. It may take two or three hours to work the process all the way through. Be ready to correct the AI’s first suggestion, and its second and third suggestions. Keep arguing until the AI gets it right. You want it to seem “ridiculously easy”. And then try it out, to see if it is. What have you got to lose?

Writing Through the Holidays

by Leave a Comment

in Production

It’s hard enough to stay motivated to keep working on your novel during normal times. When the holiday season rolls around, it’s even harder. How do you stay on the writing wagon through the holidays, so you don’t lose momentum?

The first step is to make sure you actually have momentum to begin with. I’ve blogged before about the importance of creating a writing habit, where you write on a regular schedule. But it’s hard to build a writing habit. 

Writing itself is hard. Creating a writing habit is even harder. But there’s a secret trick I learned a long time ago for creating a habit, and that same trick helps you maintain through the holidays. It’s called…

“Ridiculously Easy”

If you want to create a habit, start out ridiculously easy. As an example, suppose you wanted to create a habit of working out every day. First write down what your ultimate goal is. Maybe you write “I want to do 50 pushups every day.” Then set a time every day when you’re going to work out, maybe 4 PM. Create an alarm on your phone for 4 PM, with the notation “Do pushups.”

Then on the first day, when your alarm goes off, DO NOT do 50 pushups. 

Just do 1. 

That’s right, one measly pushup on your first day. And repeat that for the entire first week. Don’t cheat and do more!

It sounds stupid. It sounds like this couldn’t possibly do you any good. And it’s true that doing 1 pushup is probably not going to stretch you physically very much. But the action of working out every day for a week will start to build mental muscle memory. You’re training your neurons to think differently about yourself. 

After a solid week of doing 1 pushup every day, bump that up to 2 pushups. And continue that for another week. By the end of two weeks, your mental muscle memory will be a bit stronger. You are now officially half a month in on your workout routine. It’s becoming a regular part of your life. It’s what you do. You are now officially Someone Who Works Out Every Day. 

After two weeks, if 2 pushups is easy, increase it to 3. But keep it ridiculously easy for several weeks more. Make it so easy that you’re actually embarrassed because it’s “too easy.” That’s the point. During the habit-forming process, you want to be looking forward to it and longing to do more. You are building a mindset that “I can’t wait for my workout every day.”

Over a few months, you can ramp things up to the point where you start to stress yourself physically. At this point, your habit is created, and it’s a very positive habit. You look forward to it. You wish you could do it more often. You actually like working out. Building that habit is very hard mental work. That’s why you keep the physical part easy, until the habit is fully built. 

Your Writing Habit

You can build a writing habit exactly the same way. Write down your ultimate goal. It might be “I want to work on my novel for an honest hour every day, five days per week.” Then set a time every day when you’re going to write. Create a daily alarm on your phone to remind you at the appointed time. 

Then when the alarm goes off, work on your novel for exactly 1 minute. No more than that. Seriously. Just 1 minute. And do that for a week. Again, you want this to be so embarrassingly easy that you could do it in your sleep. And over the course of that week, your brain will start playing wicked tricks on you. It’ll start bringing up ideas for your novel while you’re in the shower. Or driving. Or mashing the potatoes. These are the traditional times when all good ideas come to authors or scientists or any other creative types of people. 

But discipline yourself. You must be firm. You are not allowed to work more than your allotted time, no matter how much you want to.

See what that does? It starts to drive you crazy. You start to REALLY want to write. The idea of procrastinating during your one precious minute per day is now laughable. You can’t afford to waste a second, because as soon as your minute is up, you have to stop. Be strong and STOP WRITING when your time is up. I mean it! Build your habit slowly.

Then after a week, you can open the floodgates. Now you are allowed to write for a whole 5 minutes per day, but not one second more. 

Ramp up like that over a month or two, and you’ll discover yourself with a writing habit that won’t quit. You’ll wake up at 3 AM with ideas that you dictate into your phone so you won’t lose them. 

When The Holidays Roll Around

 So let’s assume you’ve built yourself a writing habit, and now it’s that time of year again. The holidays are coming. Relatives start popping in and out. Parties start happening. There may be shopping to do, meals to make, guests to entertain. 

And you’re terrified that your carefully cultivated writing habit is going to disappear, just like it did last year. So what do you do? How do you prevent that? 

Be proactive. Set a date when your Holiday Hiatus officially begins. On that date, you can still write. But you only get 1 minute per day. Just one minute, and then you absolutely positively must stop. No cheating. No writing for 2 minutes. You must be firm. 1 minute per day. 

Until the Holiday Hiatus is over. 

What does this do?

It maintains your writing habit. Your pace has slowed down, sure. But your habit is alive and hungry and screaming to be let loose again. 

Once your Holiday Hiatus is over, ramp up again. If your habit has a long history of many months, then ramp up fast. Do 5 minutes for a few days, then 10 for a few, and keep going until you’re back to your regular schedule. If your habit is less firmly grounded, then ramp up slower. 

Do this and you’ll never fall off the wagon. The wagon will slow down, and that’s OK. Because you’ll never fall off, so you’ll never have to go through the pain of getting back on again. 

The Anthropic Class-Action Lawsuit

by 1 Comment

in Publishing

Anthropic is an AI company, the creator of the AI tool Claude, which is similar to ChatGPT. If you’re a published author, then this matters to you, because Anthropic has recently settled a class-action lawsuit filed by lawyers on behalf of authors.

Standard disclaimer: I am not a lawyer, and nothing I say here should be construed as legal advice. The case is complicated, and I can’t claim to fully understand it all. So this my best attempt to explain the situation in simple terms.

The plaintiffs alleged that Anthropic trained its AI model on books downloaded from two pirate sites, Library Genesis and Pirate Library Mirror. Anthropic denies the claims, but it has agreed to pay out $1.5 billion to authors. The estimated number of books alleged to be used for training Anthropic’s AI is about half a million.

$3000 Per Book

If you’re doing the math here, that works out to a settlement payout of $3000 per pirated book. Some of that money will go to the class-action lawyers. (That’s why they filed the lawsuit.) The rest of the money will go to whoever holds the rights to the pirated books—authors, publishers, and possibly others.

If you published a book that was pirated and used to train Anthropic’s AI model, then you are due for a payout, and it’s fairly easy to register your claim. Details on that are below, but first a few comments. There are a few cases to distinguish:

  • If you self-published the book and if you’re the sole author, then the whole payout goes to you. 
  • If you self-published a book and if you have some co-authors, then the payout will be split between you, and the group of you have to agree on the split.
  • If you published a book with a publisher and if the book has gone out of print and the rights have reverted back to you, then the whole payout goes to you. (If you had co-authors, then the payout is split between you.)
  • If you published a book with a publisher and if the book is still in print (so that the publisher still owns the right to publish), then the payout is split between you and the publisher, and the two of you have to agree on the split.

Obviously, there are a lot of possible variations on these, but it all comes down to one thing. Anyone who has a right to a payout will get paid if they file a claim and agree on a split with the others who are due money for the same book.

How To File Your Claim

There are three steps to the process of filing your claim. 

  1. If you’re curious, you can learn all about the lawsuit at this website: https://www.anthropiccopyrightsettlement.com
  2. Find out which of your books are eligible for a payout. There is a search tool on the settlement website that will tell you which of your books are eligible. And it will give you almost all the information you need to file the claim—author names, title of the book, ISBN/ASIN, copyright number, and whether the book is for “education”. I recommend that you copy this info and paste it into a spreadsheet. It’ll make the final step much easier. The search tool is here: https://secure.anthropiccopyrightsettlement.com/lookup/
  3. File a claim for each book. You can either enter info one book at a time, or you can download a spreadsheet and fill it in and then upload the spreadsheet. The page to file your claim is here: https://secure.anthropiccopyrightsettlement.com

Please note that not all your books may be covered in the settlement. If your book was published very recently, it wasn’t used to train Anthropic’s AI model. If your book was not on the pirate websites, then it’s not covered by this settlement. 

But it takes very little time to find out which of your books might be due for a payout and to file your claim. For what it’s worth, let me tell you my own experience with filing a claim.

Less Than An Hour’s Work

I hate doing paperwork, but I’ve now filed my claim, and it took me less than an hour. I have published 12 books in my writing career, and I figured that even one payout would be worth my time. I have author friends who’ve published dozens of books. A few authors I know have published over a hundred! That’s potentially a lot of dinero. But if you don’t file a claim, you won’t get paid.

I entered my name in the search tool and got 8 results back. The results were displayed in a table. I selected all the rows in the table with my mouse and then copied them and pasted the results into a spreadsheet. 

That left 4 of my titles that were apparently not covered. I entered the title of each of the missing 4 books into the search tool. 3 of these gave no results—they’re not covered by the settlement. But the 4th title returned a result. I copied that and pasted it into my spreadsheet. 

This gave me 9 titles that I had to file a claim for.

1 of these titles is still in print, so I have to share the payout with the publisher. 

1 of these titles was written with a coauthor and the publisher reverted the rights to us, so I have to share the payout with my coauthor but not the publisher. 

7 of the titles were written solely by me, and I either self-published them or the original publisher reverted the rights to me, so the payout is all mine. 

Armed with my spreadsheet, I went to the page to file my claim. 

The first obstacle was that the page asked me for the “Unique ID” provided in the Settlement Notice. I didn’t have this handy, but there was a large button that says “I don’t have a Unique ID” so I clicked that button.

This took me to a page that asked me to fill in the information for each book. The form asked for pretty much everything in my spreadsheet. It also gave me an option to download a spreadsheet to fill in the information. I downloaded it, but the columns were in a different order than my spreadsheet, and there was other infomation to be filled in. I decided just to enter the info manually into the website, since I had only 9 books. 

The whole process went pretty quick. I had two books that required split payouts. The form wants the address and contact information for the publisher or the coauthor. I had to look these up. The one publisher I had to share with has merged with another publisher and moved to a new location, but I found their contact info online. On the form, I requested a 50% split with my publisher. I also requested a 50% split for the book I wrote with a coauthor. 

Looking Forward

I would not be surprised if we see more class-action suits like this in the future. The major AI companies have trained their models on an enormous amount of information which they did not pay for, and for which they had no permission. 

The current lawsuit awarded damages only for pirated books. As I understand it, the judge didn’t award any sort of payout for the ongoing use of authors’ intellectual property. The payout seems to be only for copyright violation. I think this is pretty lame. It would be fairer to pay authors licensing fees for use of their intellectual property. I note that Joe Konrath recently blogged about this, and he’s not a bit happy either. 

Privacy Policy Cookie Policy
Terms and Conditions
>