Sit down and write. There is no secret sauce here. You don't need to buy a fountain pen or ironic vintage typewriter or mechanical keyboard as a talisman, just start writing. If you aren't feeling...
Exemplary
Sit down and write.
There is no secret sauce here. You don't need to buy a fountain pen or ironic vintage typewriter or mechanical keyboard as a talisman, just start writing. If you aren't feeling it, you aren't feeling it bringing me to Bukowski's exhortation of, "Don't try." Don't expect to write the great American novel right off the bat. That takes time and patience.
Be grateful you did some writing today, even if it was an 'ask' post on a link aggregator-cum-social media site. You still wrote. You won today. It's exercise.
And if you can't write or find that passion again, that is okay too. You are still a human being with intrinsic value and you'll discover another pursuit to nourish your soul.
I have a similar issue as OP, been blogging over the years about tech topics on my personal blog, nothing substantial. What I have found is that as my writing skill got better and better over the...
I have a similar issue as OP, been blogging over the years about tech topics on my personal blog, nothing substantial. What I have found is that as my writing skill got better and better over the years, the writer's block also started creeping in more often and more nuanced with that. It's almost like the curse of perfectionism, the more you achieve greater your expectations and then you'll settle for nothing but the best. And if that best won't come for some reason, you won't feel like writing at all. As if there is an inherent force of chaos and entropy that makes good the enemy of best inside all of us?
No writer you love pulled that story out of their head perfect straight to the page. Writers work in drafts, plural, for a reason. I've talked with so many newbie, novice, budding (pick your word)...
Exemplary
No writer you love pulled that story out of their head perfect straight to the page. Writers work in drafts, plural, for a reason.
I've talked with so many newbie, novice, budding (pick your word) writers who have convinced themselves they have to birth whatever they want to work on from their mind in a final, finished, polished form. Which is not at all how it happens.
Writing is rewriting. There is no perfect. Writers don't finish stories; they let go and move on to the next one. One of the biggest issues I find with newbies is refusing to give themselves permission to get themselves and their work absolutely filthy with the mud of work-in-progress.
There's no one way. What you do is start writing. Then you keep writing. Get things out of your head, onto a page, where you can read them. So you can decide what's missing, what's not working, what needs help, what needs extra scenes for support. So you can work on it.
Passes. Drafts. One writer might start with plot. To them, they can't see what else is needed until they have the skeleton of how the blocking that holds up basic "Jane goes here, X happens, Four things will come about in reaction, Decision, Revelation, Big Finish" is in place.
Then they might go through and add characters to actually start breathing life into those framework events. Then they might spend several passes polishing those characters, looking for ways to make things funnier, or more tense, or whatever. Then they might actually start working out how some of those events actually happen (because before it was just "they fight" because she knew there was going to be a fight between Jane and Jill that needs to happen, but hadn't figured out the specifics of that yet).
And so on. And on. And on.
They'll keep doing drafts, making pass after pass through a slowly forming coherent story, until they get it to a point they're happier with. It might take a dozen drafts. It might take fifty. It takes as many as it takes. Every writer works differently, but none of them sit down and just spin out this tale of perfection in one pass, click send, and sit back in satisfaction as the world ooohs and aaaahs at their glory.
Writing is rewriting. Give yourself permission to have dirty hands and mud on your boots. Whatever the story is, it'll take shape a hell of a lot slower than newbies always seem to assume it will. That's normal. Start wherever you decide to, and keep writing.
As EL Doctorow opined:
Writing is like driving at night in the fog. You can only see as far as your headlights, but you can make the whole trip that way.
That's what the drafts are. Each draft is as far as the headlights of your mind reach. If you keep following where your mind takes you, eventually you get there. It takes as long as it takes, and no one can hold a whole story from scratch in their head. That's why you write it down. One draft at a time.
This is the best advice on here. Even as a young writer I see so many of my peers write these big long rough drafts all in a short amount of time without ever doing much to stop, reread, and...
This is the best advice on here. Even as a young writer I see so many of my peers write these big long rough drafts all in a short amount of time without ever doing much to stop, reread, and rewrite.
The only other thing I'd add onto your post is that writing is also not a solitary process. The myth of a writer secluding themselves in a forest cabin for months and emerging with a novel just isn't true. Sometimes we are not the best judge of what is and isn't working in our stories, and we need another person's opinion. One of the most beneficial things I ever did as a writer was joining a workshop. Not only did it encourage me to write as I had to bring something to the table to show, but it also gave me invaluable feedback from other writers who knew how to give constructive critique. They could tell me what sentences were awkward, what wasn't coming across the way I wanted it to, and even more importantly they could tell me what I was doing well and what aspects of the work they'd like to see more of. It's a forum that allows us to fix our weaknesses and enhance our strengths.
The quality of a workshop or group will obviously vary tremendously, but once you find one that you really gel with, then you can start outputting solid work. It also allows you to really join a community of writers which is at least great encouragement for me.
Can you get, like, bullet points down, or the roughest of rough drafts of that idea? Then chip away at it like marble? Refine/rephrase your ideas later just to see how they sound to you? I think...
Can you get, like, bullet points down, or the roughest of rough drafts of that idea? Then chip away at it like marble? Refine/rephrase your ideas later just to see how they sound to you? I think I'm an ok writer, but I know I'm a terrible artist. When I want to sketch something that I know is well beyond by skill to render, I just go for it anyway, and then tweak it again and again, erasing and resketching and getting closer to my intention. I think talking about difficult ideas or topics functions in the same way, and writing about them even more so.
I know how much of a rush just letting the words flow out of your pen/typing fingers can be, but sometimes the best way to sell your ideas might be to riff on them for a bit.
You have to show up. Everyday. Some days will be brutal, some will have you deluded into thinking you're a master genius. It doesn't matter. Just show up. 1 page everyday adds up faster than you'd...
You have to show up. Everyday. Some days will be brutal, some will have you deluded into thinking you're a master genius. It doesn't matter. Just show up. 1 page everyday adds up faster than you'd expect.
From your replies it sounds like what you might need to change is your expectation. Read some articles and interviews by other writers about their process; notice how it sometimes takes them...
From your replies it sounds like what you might need to change is your expectation. Read some articles and interviews by other writers about their process; notice how it sometimes takes them several drafts to end up with something they want to send to the world; see how they have to cut out enormous swathes of shit to unearth the gold underneath. Fully come to terms with the fact that as much as 9/10 words you write will end up on the cutting room floor. Now ask yourself: does it matter if the first draft is shit, if most of it is going to be discarded? No! Of course not. You just want to get it down, so that your second draft has some good bones to build around. You don’t have to worry if what you’ve written is exactly what you see in your head, because you’re going to go over it again and add those details later.
Writing everyday doesn’t work for me either. I don’t have the time. It means I write more slowly, but it also means I tend to write when I’m feeling inspired and I find I end up with a lot more usable words when I’m in that zone!
Edit: just saw your response elsewhere about this same recommendation. Leaving my thoughts here, maybe some one else finds it useful. I can recommend journaling. That helps me write something...
Edit: just saw your response elsewhere about this same recommendation. Leaving my thoughts here, maybe some one else finds it useful.
I can recommend journaling. That helps me write something consistently and has the added benefit of being good for mental health as a little piece of the earth where you can be as honest as you like. I prefer pen and paper, the tactile nature is also soothing. Plus it's low stakes, just write as much or as little as is on your mind, even if it's too say nothing happened today and there's nothing on your mind. Funny how once you commit that to paper something tends to pop up ha
Sorry for misunderstanding; I feel you sound a bit snippy about it but I think you have to expect a wide range of “well duh” type replies when you ask a simple question like this. Because what can...
Sorry for misunderstanding; I feel you sound a bit snippy about it but I think you have to expect a wide range of “well duh” type replies when you ask a simple question like this. Because what can be said, really, except “just do it”?
I was responding to this from another reply of yours “I have trouble articulating the ideas in my head into words that convey to someone else exactly what I felt visualizing it” which is really not an issue if you know you’re going to have several drafts.
I think this is either something you enjoy, and then it’s easy to do it, or it isn’t and then of course it’s hard. If it isn’t something you’re currently enjoying, why force it? If it is, what’s stopping you? Those, I think, are the only two questions to ponder to yourself. Try to dig deeper than just “I’m not doing it because I’m depressed” (or whatever the mental health problem is) but why that state is stopping you - is it fear of failure? Is it low motivation to do anything at all? Is it a negative view of your ability? Is it that other things sound more appealing? No need to tell me replies to these, just some things to think about. Perhaps digging into these will help you identify the problem, and that will bring you closer to a solution. Currently this is such a broad question that it’s quite hard to answer appropriately for you!
This is a topic that is close to my own interests. I am going to write a few things and possibly return later when I have more time. The books Atomic Habits and Procrastination by Jane Burka have...
This is a topic that is close to my own interests. I am going to write a few things and possibly return later when I have more time.
The books Atomic Habits and Procrastination by Jane Burka have taught me a lot about actually performing on goals that I have. They are worth reading in full, but one takeaway is that if you commit to ten or fifteen minutes of work that removes the friction about getting started and your work session will likely go longer because you are into it.
Big Magic Creative Living Beyond Fear by Gilbert has an unfortunate first chapter filled with questionable metatheory about what creativity is. But once you get past that, it is filled with stories about herself, her writer friends, famous authors including Herman Melville and Toni Morrison. She deals explicitly with psychological obstacles to writing and I love her book.
Best of luck. When you finish, if you are willing, we would love to hear about your book. (Of course follow Tildes rules about self promotion)
Honestly, the bigger part of this is getting into the habit of sitting down at your desk for 15 minutes at the same time each morning. That should be your goal. Grab a cup of coffee and sit there....
Honestly, the bigger part of this is getting into the habit of sitting down at your desk for 15 minutes at the same time each morning. That should be your goal. Grab a cup of coffee and sit there. The rest will work itself out eventually, but putting pressure on yourself to crank out a book or produce x number of words or whatever places a huge burden and taints it. It's a walk in the woods, not a 5k run. Just be there for whatever happens and you will be posting snips of writing soon enough.
If you find yourself avoiding it once you've dedicated yourself to sitting at your desk, it's time to ask if you really want to even be writing. It's totally fine if the answer is no. There is plenty going on without crushing yourself under word debts.
Over the past year and a half, I've taken to Journaling a lot more and just generally writing my ideas down for the benefit of freeing up room in my mind. Sometimes these thoughts stray towards...
Over the past year and a half, I've taken to Journaling a lot more and just generally writing my ideas down for the benefit of freeing up room in my mind. Sometimes these thoughts stray towards fiction and I've been thinking about amalgamating them into some sort of story. I've been mostly using my phone for this purpose since it's always close at hand. You can even dictate to your phone to get the thoughts out.
I guess what I'm saying is that the opposite approach to "sit down and write" is to always be ready to write and just have the tools you need with you. You could at least get a thought you've been processing out of your head and look at it later for further analysis.
Maybe this isn't as effective for others as it is for me, but it's what has been working for me.
I second this. I carry around a small notebook and pen everywhere I go so that if I have a thought I want to break down I have a place to put it. I write theory so it's different from the kind of...
I second this. I carry around a small notebook and pen everywhere I go so that if I have a thought I want to break down I have a place to put it. I write theory so it's different from the kind of writing you do, but I think some of the same methods might still apply. I often use a tool I call 'mind mapping' which I start with a key idea and then draw lines from to break down further, which I could see being potentially helpful for world/character/sceen building. Also outlining has always been a great help for me in guiding where I will go next. And a final note, I find sitting down to edit when I can't get in the writing head space often helps me get into that head space and break back into writing. So even if I find my well seemingly dry for the chunk of time I've scheduled for writing, by spending that time at least editing and combining through what I've already done, I'm keeping myself in that space for the duration and allowing myself the greatest opportunity to keep going.
Just finish a small novel. Set a goal for 50k and start writing. 1k a day and in 50 days you'll have a shitty novel. This is how I started and I wrote a lot of shitty novels. The first 2 or 3 are...
Just finish a small novel. Set a goal for 50k and start writing. 1k a day and in 50 days you'll have a shitty novel.
This is how I started and I wrote a lot of shitty novels. The first 2 or 3 are sitting on a hard drive never to be read because they're so bad, but the 4th time around I was slowly starting to get somewhere. I now do KDP for a living and have written someone in the ballpark of 50 novels over various pen names, all in the span of about 7 years. I write pulp genre fiction so I'm not going for perfection or earth-stopping prose, just get a story down and do it fairly well.
I actually think I'm a half decent writer now. I write on the rat wheel of frequent publishing so I stick to one tight draft. My books would be even better off with multiple drafts, but I don't want to get better at writing one book, I want to get better at writing as many as possible. Only way to do that is write, publish, repeat. You might even make a job out of it.
Since nobody has recommended it so far, I highly suggest integrating an LLM like ChatGPT into your creative process. It's excellent for brainstorming ideas. It will always start with the most...
Since nobody has recommended it so far, I highly suggest integrating an LLM like ChatGPT into your creative process.
It's excellent for brainstorming ideas. It will always start with the most obvious and uninspired ideas, but if you ask for 20, there can be legitimately good ones in the middle.
If you're struggling with your plot, ask for advice.
If you literally have no idea what to do next (i.e. writer's block) go braindump what you have into the model and ask for possible directions.
If you have a section written, ask for critique. Then give the LLM a persona in your target audience and ask for another critique. Experiment with different personas for the "critic" to get feedback from different perspectives.
Before LLMs, the biggest challenge to learning to write creatively was a lack of people willing to give careful critiques. Reading thoughtfully takes a lot of time and concentration. With LLMs, this barrier is now gone, and the feedback loop for improvement is extremely tight.
The worst and least fulfilling way to use an LLM for creative writing is to just ask it to write for you. Without an extensive system of instruction and iteration (though I am working on a system for this), you will get boring but OK crap that will stifle your own voice.
However, LLMs can also be a tool to multiply your own creativity, turning written prose from brittle stone into malleable clay.
As a programmer, I'm starting to see actual written prose as compiled binaries. It's laborious to write them and they're fragile, brittle things to change. So instead I'm devoting my energy...
As a programmer, I'm starting to see actual written prose as compiled binaries. It's laborious to write them and they're fragile, brittle things to change. So instead I'm devoting my energy towards refining my worldbuilding documents—a proto source-code if you will.
With the right system of guided analysis and iteration, I think it will be eventually possible to use a worldbuilding "seed" to generate experiences ranging from customized stories to immersive movies and games.
I suspect that this is where entertainment is heading in the next 50 years. It's a level of abstraction higher than "writing" and it meshes well with VR.
Of course, LLMs in their current state are still very useful for actually writing, like I originally described.
I think to be a good writer you need to hone those skills yourself instead of relying on AI. It’s kind of like using PEDs in sports, to me. I certainly hope this isn’t the direction that writing...
I think to be a good writer you need to hone those skills yourself instead of relying on AI. It’s kind of like using PEDs in sports, to me. I certainly hope this isn’t the direction that writing is headed, and I don’t believe it is - I truly believe there will always be higher value from words written by people than by computers
Rather offended by this comment Not incumbent on you to care about the modern (now somewhat classical) edifice of formal (or informal) written prose—plenty of interest and value in concept, plot,...
Rather offended by this comment
Not incumbent on you to care about the modern (now somewhat classical) edifice of formal (or informal) written prose—plenty of interest and value in concept, plot, setting, etc.
Can communicate effectively without it (as this comment demonstrates)—maybe prose qua prose is overused (it is and it isn't, but that is a separate issue)—also with maps, computer structures&vis, etc.—but
You denigrate the art as purely mechanical only because you find interest in worlds and not words
Absurd
If you don't like words, why bother getting a computer to write them for you?
(Have I hidden fluff in these last few bullets, abusing the format and defeating my own argument? No; it is a concise conveyance of my emotional state. This bullet is fluff)
Comparison to compilers is apt, but not for the reason you think
Non-optimising compilers typically mechanical, simple, and boring; ok; but optimising compilers today have an interesting problem to solve yet are fairly primitive (humans can do much better despite memes claiming the opposite). Three things stand in their way:
Incompleteness theorem/rice's theorem/halting problem/etc. Mathematically proven nothing to be done about this; ok. Others more interesting
Too many states to explore. Not enough time to explore them all so need creativity to decide what to look at
Cost models—ie value judgments—very hard to determine what's faster than what, especially given time/space tradeoffs, heterogeneous arches, uarch lolfuckery, etc
Couldn't resist; expanding on this: Written interpersonal communication (convenient case study): Two main functions of fluff Content-free ingratiating ritual. Very valuable—similar to small talk...
maybe prose qua prose is overused (it is and it isn't, but that is a separate issue)
Couldn't resist; expanding on this:
Written interpersonal communication (convenient case study):
Two main functions of fluff
Content-free ingratiating ritual.
Very valuable—similar to small talk
Eg did I write the above comment under the fluffy-interpersonal regime I might have opened with a hedge (@kfwyre has a comment somewhere about these) like 'to be frank'
Redundancy
How do I ensure you don't misinterpret me?—I say the same thing multiple times, to ensure that the point gets across
Eg: the fluffy bullet in the above comment was completely redundant—but I did not want to be misinterpreted, so I included it anyway
1 is, if performed by a robot, completely useless—what is the point of forming an interpersonal relationship with a robot?
ESL sometimes find success with such things. Difficult to blame them—they don't know the rituals since they are not in their tribe—the blame lies with the native speakers for being exclusionary—but there is an affordance for the former anyway which is that not knowing the language is often cute and hence its own ingratiation
2 is also useless if performed by a robot: doesn't know what you were thinking if ambiguous; just makes something up to fill in the blanks
I sometimes make notes to myself in format like this. I already know how amazing I am and I am not communicating w/myself but creating mnemonics for memories
Corporate copy is filled with ingratiating fluff. It's dishonest. No relationship to be had with a corp
Internal corp emails often very terse and to the point—usually appropriately so, in context—the push for ever terser, conciser, clearer language comes from this crowd. Appropriate outside of context? Probably not
re: LLMs in storytelling and fiction prose writing
hotspots? identify and fix
99% quality? open question, new tool, nobody knows
bigger epic stories?
(translation and expansion)
Sorry for offending you!
In regards to the terse communication style, its effectiveness is questionable. Reading such an information-dense excerpt is incredibly taxing, since there's no room for error in interpretation. That may be more evident in my version above, since it will not be familiar to you.
This is a very interesting subject, and was a fun experiment! When I work on my worldbuilding, I try to make my outlines as dense as possible in order to reduce the token count. I spent some time today seeing what sort of "translations" into plain English I could generate from the above outline using airoboros-l2-70b-gpt4-2.0. (Ultimately, the results were good, but not good enough for me to paste into this post.)
The density makes it harder for humans to understand. Does this reduce the comprehension of the LLM? Interesting question, but the answer from my testing today is "not by much" (at least for 70 billion parameter LLaMA 2).
When you take out the "fluff", you lose a number of things. First, you no longer have the in-context clues to the meaning of words: their definitions as well as how they relate to the flow of the thought. Additionally, you lose the natural transitions between thoughts. Finally, you present words in a way that is in an unfamiliar pattern. A familiar linguistic pattern is effortless (subconscious?) to decode, whereas unfamiliar structures require concentrated effort to parse.
"Redundancy" serves as error-correcting code. Muffled or misheard words in spoken language can still be interpreted correctly. I think this also helps with speed reading, since you can skip words while reducing the chance of missing an important detail.
I don't mean to belittle prose. It may be brittle, fragile, and (always, ultimately) flawed, but it can also be exceptional, awe-inspiring, beautiful, and artisanally crafted. It's also, up to this point, all we had, since computers couldn't actually manipulate abstract "ideas". Now they can (with caveats).
The heart of storytelling lies in concepts, worlds, plots, and characters—elements that can currently manifest as books, movies, TV shows, games, theatre, etc. But with the rise of generative AI, I think we will be seeing modified forms of these media that are interactive, immersive, adaptable, and customizable. I predict that the main energy of the next generation of creatives will be focused on inventing and fleshing out these new forms of media, though it will never entirely displace the classic forms.
Artists are not going to stop sketching, just because a computer will be able to create almost any detailed image imaginable. Writers are not going to stop writing, just because a computer can write an infinite number of well-structured stories! They will continue to be creative, and direct the computer to fulfill their creative vision, because the pure computer-generated versions will not communicate their vision well enough.
If you don't like words, why bother getting a computer to write them for you?
I'll admit I am not a lover of prose. I prefer to focus on characterization, conflict, and dialogue, having more experience with screenplays than books. That's why I am taking such an orthogonal approach in the first place!
Comparison to compilers is apt, but not for the reason you think
The comparison actually goes even deeper, and is more apt, but in my direction! While hand-written assembler has the potential to be faster than high-level code, it is severely limited in the complexity of computing problems you can tackle in assembler subroutines.
Once you no longer have to worry about memory addresses and registers, and can use functions, objects, garbage collection, etc, you can can build enormously complex programs. For example, LLMs could never have been produced if we were all limited to hand-written assembler. (Machine learning is all done in python, which is quite terrible from a performance perspective.)
In terms of performance, all you need to get 99%+ of the performance of a hand-written assembler program (caveat: not memory usage), is to do performance profiling to find the hotspots (usually in loops). In those hotspots, you can write human-optimized inline-assembly.
How does this compare to LLMs in writing? Perhaps a writing program uses LLMs to generate the base text, and then find the "hotspots". I imagine those spots will be the major points the plot turns on. In these spots, you "hand-optimize" the prose and dialogue with your unique human ingenuity and creativity: to make the most impact.
This is not a new pattern. In drawing and painting, it's a waste of time (and a mark of a beginner) to plaster detail all over the canvas. You want to use broad strokes, implied edges, and vague impressions for the majority of the space, and only detail the focal areas and focal points. Doing otherwise is a waste of your time and will generally be distracting to the viewer. That time is better spent on the composition, big-picture, and focal points.
Now, the question is, if LLMs will enable a jump in writing complexity similar to going from assembly to high-level languages, what might that look like in practice?
I think the low-hanging fruit, once the technical kinks are worked out and the context window of the LLMs are sufficiently expanded, is assisting authors in creating sprawling epics of massive complexity, while still being coherent, internally consistent, and well structured.
Caveat emptor! I did not write tersely or densely, and instead ended up writing quite a lot. Dear to my heart though the problem is, when I mentioned compilers I was not so much talking about the...
Exemplary
In regards to the terse communication style, its effectiveness is questionable
Caveat emptor!
I did not write tersely or densely, and instead ended up writing quite a lot.
Dear to my heart though the problem is, when I mentioned compilers I was not so much talking about the vagaries of code generation as high-level, structural, algorithmic knowledge and transformation. LLVM and its ilk are not even trying; for the non-strawman attempts, see things like computer algebra systems (ok, pretty much just mathematica), SMT/SAT solvers, DB query plan optimisers, and theorem provers.
A story may have moments of greater or lesser intensity (or emotion, or spectacle, or what-have-you), but that does not mean that the former are 'hotspots' in the sense of a poorly performing application. Recently, I was reading Gardner's The Midnight Reader, and, on a few occasions, had to go back and reread a paragraph or a few of them because I realised that it was perfect—every word perfectly placed, furthering its own goals as well as the goals of every structure around it. Not that the rest was poorly written—it was quite good—but it did not reach those heights. I despair of ever writing that well, but I don't think Gardner picked and chose those paragraphs to focus on. I don't think he could have. These were fairly mundane scenes; which mundanity is certainly rather important to the story's overall themes, but I did not notice anything particularly distinguishing those scene's content or placement with respect to the others (it is of course possible something went over my head). The muse speaks as she wills—you cannot control her—but that does not mean you are powerless, and you will much more clearly see the role of a word with respect to its mates if you have, you know, written the mates. You also cannot write that well without practising.
Or take Tolkien, whose writing has an incredibly consistent quality (last I reread Lord of the Rings was a few years ago, but I think I could only find minor fault with a couple of commas—and who am I to judge his comma choice, anyway?). This gives the text an intentional quality. But this intentionality does not just apply to the composition of the words; it applies to every aspect of the work. The best English teacher I ever had, in high school (alas, our personalities clashed, but hers was still probably the only worthwhile class I took in high school), was fond of saying: style is content. It is true: the stylistic ideals are and reflect the conceptual ideals; you cannot create a work with such a consistently intentional feel if you do not so intentionally choose every word. The language is also a draw in and of itself—analogous, more or less, to the death-by-a-thousand-cuts model of performance which has been credibly proposed as an alternative to the hotspot model—it is a joy to read. Nor do I think—regardless of whether or not his work is moderately offensive propaganda—that Tolkien's artistic vision was compromised, bogged down, or distracted by his focus on language—rather the opposite!
Ok, one more example. I read a comic book which had been translated into English—actually, I read two different translations, one of which was markedly better than the other. In the better translation, I marked one line as being the best in the whole work—it was subtle and a bit understated, but poignant and funny and, in particular, I felt, reflective of the translator's understanding of the character she was depicting and the cultures the character and the reader fit into. In the worse translation, it was filler. That's all it was: something to say to get to the next panel. The translator of the worse edition might have credibly claimed that the line in question was not very important, but that would have been more convincing had she written the comic herself and elided it or changed its role; under the circumstances, being that translating the text of a comic book forces certain constraints, I think her choice diminished the quality of the work. (Not to imply that that is in fact what she thinks; she might disagree that her choices were worse, or she might have been under time pressure...)
Drawing and painting are not really my media, and I do not claim to be able to speak about them authoritatively, but I will conjecture. Time is different from space; a painting is seen all at once. Hence, it is more like a concurrent system than a sequential one, and choosing an area to attack is more like finding bottlenecks than hotspots.
I am not a lover of prose. I prefer to focus on characterization, conflict, and dialogue, having more experience with screenplays than books.
And—this has been my sentiment from the beginning, if it was not clear—more power to you! I just don't see what you hope to gain from having robots write prose for you. I think that (for example) an interactive map showing the development of political (and e.g. geological, meteorological) events over time in some world could be rather interesting to explore.
I think the low-hanging fruit, once the technical kinks are worked out and the context window of the LLMs are sufficiently expanded, is assisting authors in creating sprawling epics of massive complexity, while still being coherent, internally consistent, and well structured
That sounds plausible to me. But it also sounds like a rather different application: running queries against an existing body of work. I'll link my previously expressed sentiments regarding grammarly and spell-checkers. I also don't think you would need such a tool to do such structuring, and it might be suboptimal for many aspects of the problem; I recall a demo from Lenat using Cyc to analyse Romeo and Juliet, and you could use much more primitive (or specialised) tools too.
In regards to the terse communication style, its effectiveness is questionable. Reading such an information-dense excerpt is incredibly taxing, since there's no room for error in interpretation.
It is taxing! I do not think that there an argument to be made for a monotonic reduction in comprehensibility, though. The tradeoffs are similar to those of APL—the denser your ideas, the more you can fit in your head at once.
You’re going to get some pushback for this but I think this kind of method is only going to get more popular and useful over time with Microsoft building “co-pilots” into windows and all their...
You’re going to get some pushback for this but I think this kind of method is only going to get more popular and useful over time with Microsoft building “co-pilots” into windows and all their office products.
I can empathize with the pushback TBH. Generative AI (specifically image generators) sent me into an existential crisis when it took off. Before that I was very focused on digital art. It makes it...
I can empathize with the pushback TBH. Generative AI (specifically image generators) sent me into an existential crisis when it took off. Before that I was very focused on digital art.
It makes it much harder to pick up my sketchbook nowadays.
When I do finally get back to character sketching and drawing (after I'm satisfied with my worldbuilding and writing) I plan to liberally abuse ControlNET+StableDiffusion to finish and transform my drawings instead of doing it manually. Maybe even train some (personal & private) qloras on my favorite artists to better analyze their styles for integration into my own sketching process.
I apologize if this is not the case, but your reply sounds like you missed an important part of my comment. Even if you have no interest in generating ideas or prose with LLMs, getting immediate...
I apologize if this is not the case, but your reply sounds like you missed an important part of my comment. Even if you have no interest in generating ideas or prose with LLMs, getting immediate critique from various perspectives is invaluable:
If you have a section written, ask for critique. Then give the LLM a persona in your target audience and ask for another critique. Experiment with different personas for the "critic" to get feedback from different perspectives.
I suppose the immediacy of feedback is one advantage of using AI language models for critique, but I wouldn't say it's the best way to get critique at all. It depends on what kind of writing your...
I suppose the immediacy of feedback is one advantage of using AI language models for critique, but I wouldn't say it's the best way to get critique at all. It depends on what kind of writing your doing, but in almost any case there are existing entities that can do the same thing from a human perspective rather than from a perspective imitating what it thinks a human would say. If you write fiction, there are writers groups and workshops everywhere for every skill level which give you extremely pointed feedback on your story from real people with real-world perspectives, if you're a journalist it's probably a conversation you should have with your editor or coworkers, if you're a student it might be a conversation best had with your professor as any professor will accept early submissions for review before the final deadline.
There are plenty of situations where you could use something like ChatGPT to benefit you, maybe you're a first-time writer and you want to go into a real workshop with some idea of what to expect, but that's also something that could be accomplished by a lower-stakes online workshop, with the added benefit of being able to talk to real people and get relevant responses.
That's not to dismiss the usefulness of AI, but rather to say there are better ways to do the same thing!
There's really no value in this. A language model isn't a person. It's not going to give nuanced feedback to a piece of art. It's designed to predict the most likely next word in a sentence....
There's really no value in this. A language model isn't a person. It's not going to give nuanced feedback to a piece of art. It's designed to predict the most likely next word in a sentence. Please don't credit it with anything more.
I'm going to give some less prescriptive advice than others in here (though I'd echo the idea that the only way to write is to sit down and write): Start reading more. The more you read, the more...
I'm going to give some less prescriptive advice than others in here (though I'd echo the idea that the only way to write is to sit down and write):
Start reading more. The more you read, the more you'll feel inspired, the more you'll feel the pull to sit down and write.
I highly highlyhighly recommend Anne Lamott's Bird by Bird. I think it is the one "craft" book every writer needs to read. It's far less about how to write and far more about how to make writing a part of your life. I teach "Shitty First Drafts" in my freshman comp classes, because the most often-voiced concern new writers have is that their writing isn't up to par. But no writer spits out a perfect draft on their first try. They have to sit down and get the words onto page so that they can edit them into shape. Give yourself permission to write garbage. Don't let the elusiveness of perfection paralyze you and prevent you from even starting.
Another great philosophy Lamott advocates for is the one-inch picture frame. Focus on something small: Today you're just going to write the first page. Today you're just going to write the first sentence. Today you're just going to brainstorm some titles. Inevitably, you'll meet your benchmarks, and more likely than not, you'll feel pulled to keep going. And hey, now you've overproduced. Now you've got some work to edit.
Instead of telling yourself you want to write a novel this year, start small. Write a story in a month. Write a story in a week. Let yourself build up to your larger goals. No one wakes up and runs a marathon without building up the stamina for that distance.
This might sound silly but I started writing and posting erotic stories under a different name. It's easy to write, the expectations are low. I used to get stuck on a paticular word choice or a...
This might sound silly but I started writing and posting erotic stories under a different name. It's easy to write, the expectations are low. I used to get stuck on a paticular word choice or a bit of scene description. But since the stakes are so low I was a lot more comfortable just letting the words flow.
I'm getting a bit bored of it now, but it's got me back in the habit of writing and the mood to do some serious writing again.
What I've found useful to just get in the mindset of writing is "playing" solo journaling rpgs like Thousand Year Old Vampire. I found it really useful to have a more guided experience, like...
What I've found useful to just get in the mindset of writing is "playing" solo journaling rpgs like Thousand Year Old Vampire. I found it really useful to have a more guided experience, like writing prompts on steroids.
As someone that also deals with depression and the lack of motivation that comes with it, making writing fun was really useful to me and has allowed me to write more when I want to. Bonus points for the ideas I have from playing a journaling game inspiring my writing.
Hey, that's the same situation I was in literally days ago. I used to write a fair amount but rarely finished anything, then work got busy and I fell out of the habit. The method that I found most...
Hey, that's the same situation I was in literally days ago. I used to write a fair amount but rarely finished anything, then work got busy and I fell out of the habit. The method that I found most effective in the past was to set aside a specific amount of time for writing each day. It didn't matter how many words I put down on the page, but I had to focus on writing and nothing else.
Another trick I've read about and tried was called freewriting. What you do is take a piece of paper and write down anything that comes to your mind. It doesn't have to be coherent, the goal is to get into the right headspace for writing. Once you finish the page, move on to the project you're actually working on. This one was a little more hit-or-miss for me, but you could give it a shot too.
For the past three days I've scheduled myself 30 minutes to write. So far, it's been easier getting back into writing than I expected. That was nice.
I haven't written in a while, so who am I to preach? Still, I feel I have found a good approach, inspired mostly by the book Atomic Habits. My fundamental belief is that writing is easy; you...
I haven't written in a while, so who am I to preach? Still, I feel I have found a good approach, inspired mostly by the book Atomic Habits. My fundamental belief is that writing is easy; you literally just sit on your butt and go clickety-click on your keyboard. Consistent writers don't have any kind of superboosted willpower muscles, because willpower muscles is a myth, they don't actually exists. What they have are some habits which allows them to simply write without having to think much about it.
Schedule
Have a seperate writing cave with a computer you only use for writing. This is to help tell your brain that what happens in here is writing, so you don't really have to think about it. You just have go in there and sit you ass down in the chair, and you brain lazily go into writing mode because routine.
Set a timer and write for 2 hours in the morning at a fixed time, like 8:00 to 10:00. Two hours is short enough for you to write without burnout, and writing in the morning mean that you don't have the task waiting for you the entire day. But most importantly, writing at a specific time means that you don't have to think about when to write.
Seperate the task of writing (getting words down on the page) from any other writing-related task. (brainstorming, researching, editing, reading about writing, watching youtube vids about writing) Don't allow yourself to procrestinate with non-writing.
Don't have internet on your writing computer. If you need to research something, write it down on a piece of paper or in a notebook and look into it later. The classic authors simply didn't have the internet to distract them. They had a typewriter which could put letters on a piece of paper and that's it. No reason you should make it harder for yourself by having distractions constantly available.
Some General Advice
Create a playlist with the right mood for your story.
Have Google Docs or Zoho Workdrive on your mobile, so you constantly can type down your ideas.
Read books, so that you brain is reminded that this is a medium which matters to you. Also, where else would you steal your ideas?
Tips Regarding Writers Block
Consider aiming for short stories or flash fiction. Those are easier to finish before you run out of steam.
Don't be goal-oriented. Don't focus on finishing your novel. If you do, writing will feel like a failure, since you constantly haven't reached your goal yet. Instead, focus on your identity. You're an author. Now each day is a success. You have authored.
Write down what the core of your story is, why it matters to you. You will need to read this when you feel your story feel stale.
If you your brain keep worrying about being unsure about the story, set a fixed time of day where you are allowed to go on about it. At any other time, tell yourself that this is certainly a relevant topic, but you're not going to think about it right now. Then gently brush it away. Just because a thought pops into your head doesn't mean you have to give it your attention. (The idea of being in control of your own mind stems from Metacognitive Therapy (MCT))
If a scene in your story bothers you, sleep on it. Examine it gently or let it hang just outside your consciousness as you go to sleep. When you wake up, the solution will be crystal clear.
Brainstorm. Write 10 different versions and sort them by coolness. Use this for scenes, titles, names, bothersome paragraphs or what have you.
Or just allow the scene to suck, and move forward. You can alway go back and edit it later.
Or just grant yourself a break. But set an alert for the time you're going to get back into gear.
Try a different medium! Whether that's poetry when you usually do prose, sci-fi when you usually do grounded slice-of-life stuff, or (and this is what helps me) take your idea and do something...
Try a different medium! Whether that's poetry when you usually do prose, sci-fi when you usually do grounded slice-of-life stuff, or (and this is what helps me) take your idea and do something completely non-writing with it. I started making collages and assemblages a while back, and that helped with the book I'm writing in a huge way..... Now the book is illustrated, and nearly finished, and it's so freeing to want to work on my project, and have options to focus on if I'm not feeling writing that day.
I've never been one to "write everyday," "just sit down and do it!" ...fuck that. Doesn't work for me. Too much pressure, makes everything feel bigger and harder than it needs to. But even on my worst brain day, "I want to work" can mean anything from writing to digging through books for collage pieces, to digital graphic design, to resin (I'm layering the collages with resin to add more depth). Or even just reading about the subject matter, that counts too.
And regardless of the medium, make intentionally bad art. Write a terrible story, draw something your 6 y/o cousin wouldn't even be proud of, take a block of wood and whittle the shittiest poking stick you've ever seen. Have fun with being creative in general, and the writing will come.
it's tough to do so when you no longer feel life's worth living. I think people here are glossing over this. If you are actually struggling with existential despair and or clinical depression,...
it's tough to do so when you no longer feel life's worth living.
I think people here are glossing over this. If you are actually struggling with existential despair and or clinical depression, professional help is something I would recommend trying. Sometimes depression is brain chemicals. Other times it is overwhelming grief. Other times it is about loneliness or having been battered by abusive relationships or work places. But based on your replies here, you need to deal with your depression to move forward.
I'm coming to this very late. I'm going to apologize that I've only read a couple of the replies that seem very good but I wanted to add in a few things that work for me. I tend to write in waves....
I'm coming to this very late. I'm going to apologize that I've only read a couple of the replies that seem very good but I wanted to add in a few things that work for me.
I tend to write in waves. Super productive highs and do nothing or fuck around elsewhere lows. Seems like you're in a low point. The best thing that tends to get me back into the groove is listening to songs I've picked out for scenes/characters. My normal playlists consist of a lot of beat heavy music which for the most part doesn't jive with the slice of life/soap opera story I'm writing. Listening to things that remind me of characters/scenes makes me think about them and how I want to get them on paper.
Oh, not sure if anyone else said this or not but you do not have to write linearly. If there is a scene or character in your head yelling to get out, write that. You can always go back to the point you were working on before plus it gets that first terrible version out of your head to be improved later.
So yeah, don't try to force it. Nudge it a bit with some music (or something else that makes you think about it) and see how it goes.
Good luck!
your fellow writer toiling through a series they won't finish before they die lol
Big same. Totally get that as well. My whole project started as a webcomic, but I can't draw and the folks who I wanted to do it are out of my price range AND have their own long ass stories going...
Random music is where a lot of the character and story beats come from for me. Scenes just come to me sometimes (often from listening to a song)...
Big same.
...but it can be tough to get them down because they play in my head like a film.
Totally get that as well. My whole project started as a webcomic, but I can't draw and the folks who I wanted to do it are out of my price range AND have their own long ass stories going on. So, it was a Thanos 'fine, I'll do it myself' moment.
I want the reader to have that same experience.
Still agree. But its fucking hard to get it on paper how I see it in my head. lol
One of the characters really sticks in my mind and it's difficult to push them aside even for a time to put more attention on the others.
Yep. My spotify discover weekly has essentially become 'what synthwave am I adding to UCSO this week?' when that isnt even my big project. It takes EFFORT to swap back over to the main story music.
I guess being that they're essentially a modern take on the siren myth it makes sense that it's hard to put them out of your mind.
Yo, THAT is an excellent analogy!!!!
See, you've got what it takes to do this. Just need to push your brain out of its own way sometimes. lol
Do you share your stuff anywhere? I'm still pretty new to this (4 years or so in) but I'd be willing to take a look and share thoughts.
I got myself back into doing it by setting up two things: A consistent schedule, and less limitations. On the first point, just resolving to regularly do it will go a long way. It doesn't need to...
I got myself back into doing it by setting up two things: A consistent schedule, and less limitations.
On the first point, just resolving to regularly do it will go a long way. It doesn't need to be a long time, either. For me, it's 30 minutes every other day. Regardless of how I feel, or whether I have any ideas, I will sit down and write something out. A lot of the time it's just a stream of consciousness, or something akin to a journal entry, but occasionally it comes together real well and with that I'm apt to continue. I keep these things all in one place and refer back sometimes.
On the second point, I let go of really any expectations to produce. I am not writing a book. I am not writing a story. I'm not even writing a scene necessarily. The point is to just do it in some form or other, and if it takes the shape of those things, cool. Forget the audience, don't even think of the audience, just do the thing and see what that is, has been my mindset with that.
What about writing some nonfiction? Then you're writing, but you don't need to be creative in the same way. Write a walkthrough for a video game you like, or idk, a guide to getting set up on...
What about writing some nonfiction? Then you're writing, but you don't need to be creative in the same way. Write a walkthrough for a video game you like, or idk, a guide to getting set up on Tildes or the Fediverse. Maybe it'll help with the habit enough that you'll start feeling the spark again, and get back into fiction straightaway.
I'm not sure I have much advice for the OP. I've written a few novels, and self-published them through Amazon, but mostly for my own satisfaction. The motivation for my writing comes from within....
I'm not sure I have much advice for the OP. I've written a few novels, and self-published them through Amazon, but mostly for my own satisfaction. The motivation for my writing comes from within. My first novel was an exploration of a recurring dream, where one character kept showing up. It was simple curiosity, and I had to find out all that I could about him. I could picture the first scene, but I didn't know why he was there or what the setting was. So, I started typing.
Some writers build outlines, and require structure before they can write. I'm just the opposite. I have to explore as I write, and I'm in a constant state of wonder, as new scenes and new characters appear from my imagination.
Determine your process and dive in. The first few drafts will be garbage. None of us are Mozart in Amadeus, and it won't appear on the page as perfection from above. Don't get discouraged in these first few rounds of the fight in other words.
OK. It looks like you know your own process well enough. What you're lacking is a spark. I'd suggest keeping your mind open to events around you. Keep reading and observing people. If you find...
OK. It looks like you know your own process well enough. What you're lacking is a spark. I'd suggest keeping your mind open to events around you. Keep reading and observing people. If you find something interesting, build an outline for a story in your mind about it. You don't have to actually write it down. It may not be compelling enough for that, but just working through it in your mind will help.
When you find something that is truly worth your attention, put it on paper. Again, you don't have to plan an opus. You're trying to get back on the horse, not win a race. Little steps are what you need at first.
I agree with some of the others in here, that writing every day is not helpful for many. It becomes mechanical, tedious and counter-productive. What you need are subjects that you care about. If you are indifferent to your subject, how are you going to make anyone else care about it?
Sit down and write.
There is no secret sauce here. You don't need to buy a fountain pen or ironic vintage typewriter or mechanical keyboard as a talisman, just start writing. If you aren't feeling it, you aren't feeling it bringing me to Bukowski's exhortation of, "Don't try." Don't expect to write the great American novel right off the bat. That takes time and patience.
Be grateful you did some writing today, even if it was an 'ask' post on a link aggregator-cum-social media site. You still wrote. You won today. It's exercise.
And if you can't write or find that passion again, that is okay too. You are still a human being with intrinsic value and you'll discover another pursuit to nourish your soul.
I have a similar issue as OP, been blogging over the years about tech topics on my personal blog, nothing substantial. What I have found is that as my writing skill got better and better over the years, the writer's block also started creeping in more often and more nuanced with that. It's almost like the curse of perfectionism, the more you achieve greater your expectations and then you'll settle for nothing but the best. And if that best won't come for some reason, you won't feel like writing at all. As if there is an inherent force of chaos and entropy that makes good the enemy of best inside all of us?
No writer you love pulled that story out of their head perfect straight to the page. Writers work in drafts, plural, for a reason.
I've talked with so many newbie, novice, budding (pick your word) writers who have convinced themselves they have to birth whatever they want to work on from their mind in a final, finished, polished form. Which is not at all how it happens.
Writing is rewriting. There is no perfect. Writers don't finish stories; they let go and move on to the next one. One of the biggest issues I find with newbies is refusing to give themselves permission to get themselves and their work absolutely filthy with the mud of work-in-progress.
There's no one way. What you do is start writing. Then you keep writing. Get things out of your head, onto a page, where you can read them. So you can decide what's missing, what's not working, what needs help, what needs extra scenes for support. So you can work on it.
Passes. Drafts. One writer might start with plot. To them, they can't see what else is needed until they have the skeleton of how the blocking that holds up basic "Jane goes here, X happens, Four things will come about in reaction, Decision, Revelation, Big Finish" is in place.
Then they might go through and add characters to actually start breathing life into those framework events. Then they might spend several passes polishing those characters, looking for ways to make things funnier, or more tense, or whatever. Then they might actually start working out how some of those events actually happen (because before it was just "they fight" because she knew there was going to be a fight between Jane and Jill that needs to happen, but hadn't figured out the specifics of that yet).
And so on. And on. And on.
They'll keep doing drafts, making pass after pass through a slowly forming coherent story, until they get it to a point they're happier with. It might take a dozen drafts. It might take fifty. It takes as many as it takes. Every writer works differently, but none of them sit down and just spin out this tale of perfection in one pass, click send, and sit back in satisfaction as the world ooohs and aaaahs at their glory.
Writing is rewriting. Give yourself permission to have dirty hands and mud on your boots. Whatever the story is, it'll take shape a hell of a lot slower than newbies always seem to assume it will. That's normal. Start wherever you decide to, and keep writing.
As EL Doctorow opined:
That's what the drafts are. Each draft is as far as the headlights of your mind reach. If you keep following where your mind takes you, eventually you get there. It takes as long as it takes, and no one can hold a whole story from scratch in their head. That's why you write it down. One draft at a time.
Keep writing.
This is the best advice on here. Even as a young writer I see so many of my peers write these big long rough drafts all in a short amount of time without ever doing much to stop, reread, and rewrite.
The only other thing I'd add onto your post is that writing is also not a solitary process. The myth of a writer secluding themselves in a forest cabin for months and emerging with a novel just isn't true. Sometimes we are not the best judge of what is and isn't working in our stories, and we need another person's opinion. One of the most beneficial things I ever did as a writer was joining a workshop. Not only did it encourage me to write as I had to bring something to the table to show, but it also gave me invaluable feedback from other writers who knew how to give constructive critique. They could tell me what sentences were awkward, what wasn't coming across the way I wanted it to, and even more importantly they could tell me what I was doing well and what aspects of the work they'd like to see more of. It's a forum that allows us to fix our weaknesses and enhance our strengths.
The quality of a workshop or group will obviously vary tremendously, but once you find one that you really gel with, then you can start outputting solid work. It also allows you to really join a community of writers which is at least great encouragement for me.
Can you get, like, bullet points down, or the roughest of rough drafts of that idea? Then chip away at it like marble? Refine/rephrase your ideas later just to see how they sound to you? I think I'm an ok writer, but I know I'm a terrible artist. When I want to sketch something that I know is well beyond by skill to render, I just go for it anyway, and then tweak it again and again, erasing and resketching and getting closer to my intention. I think talking about difficult ideas or topics functions in the same way, and writing about them even more so.
I know how much of a rush just letting the words flow out of your pen/typing fingers can be, but sometimes the best way to sell your ideas might be to riff on them for a bit.
You have to show up. Everyday. Some days will be brutal, some will have you deluded into thinking you're a master genius. It doesn't matter. Just show up. 1 page everyday adds up faster than you'd expect.
From your replies it sounds like what you might need to change is your expectation. Read some articles and interviews by other writers about their process; notice how it sometimes takes them several drafts to end up with something they want to send to the world; see how they have to cut out enormous swathes of shit to unearth the gold underneath. Fully come to terms with the fact that as much as 9/10 words you write will end up on the cutting room floor. Now ask yourself: does it matter if the first draft is shit, if most of it is going to be discarded? No! Of course not. You just want to get it down, so that your second draft has some good bones to build around. You don’t have to worry if what you’ve written is exactly what you see in your head, because you’re going to go over it again and add those details later.
Writing everyday doesn’t work for me either. I don’t have the time. It means I write more slowly, but it also means I tend to write when I’m feeling inspired and I find I end up with a lot more usable words when I’m in that zone!
Edit: just saw your response elsewhere about this same recommendation. Leaving my thoughts here, maybe some one else finds it useful.
I can recommend journaling. That helps me write something consistently and has the added benefit of being good for mental health as a little piece of the earth where you can be as honest as you like. I prefer pen and paper, the tactile nature is also soothing. Plus it's low stakes, just write as much or as little as is on your mind, even if it's too say nothing happened today and there's nothing on your mind. Funny how once you commit that to paper something tends to pop up ha
Sorry for misunderstanding; I feel you sound a bit snippy about it but I think you have to expect a wide range of “well duh” type replies when you ask a simple question like this. Because what can be said, really, except “just do it”?
I was responding to this from another reply of yours “I have trouble articulating the ideas in my head into words that convey to someone else exactly what I felt visualizing it” which is really not an issue if you know you’re going to have several drafts.
I think this is either something you enjoy, and then it’s easy to do it, or it isn’t and then of course it’s hard. If it isn’t something you’re currently enjoying, why force it? If it is, what’s stopping you? Those, I think, are the only two questions to ponder to yourself. Try to dig deeper than just “I’m not doing it because I’m depressed” (or whatever the mental health problem is) but why that state is stopping you - is it fear of failure? Is it low motivation to do anything at all? Is it a negative view of your ability? Is it that other things sound more appealing? No need to tell me replies to these, just some things to think about. Perhaps digging into these will help you identify the problem, and that will bring you closer to a solution. Currently this is such a broad question that it’s quite hard to answer appropriately for you!
This is a topic that is close to my own interests. I am going to write a few things and possibly return later when I have more time.
The books Atomic Habits and Procrastination by Jane Burka have taught me a lot about actually performing on goals that I have. They are worth reading in full, but one takeaway is that if you commit to ten or fifteen minutes of work that removes the friction about getting started and your work session will likely go longer because you are into it.
Big Magic Creative Living Beyond Fear by Gilbert has an unfortunate first chapter filled with questionable metatheory about what creativity is. But once you get past that, it is filled with stories about herself, her writer friends, famous authors including Herman Melville and Toni Morrison. She deals explicitly with psychological obstacles to writing and I love her book.
Best of luck. When you finish, if you are willing, we would love to hear about your book. (Of course follow Tildes rules about self promotion)
Honestly, the bigger part of this is getting into the habit of sitting down at your desk for 15 minutes at the same time each morning. That should be your goal. Grab a cup of coffee and sit there. The rest will work itself out eventually, but putting pressure on yourself to crank out a book or produce x number of words or whatever places a huge burden and taints it. It's a walk in the woods, not a 5k run. Just be there for whatever happens and you will be posting snips of writing soon enough.
If you find yourself avoiding it once you've dedicated yourself to sitting at your desk, it's time to ask if you really want to even be writing. It's totally fine if the answer is no. There is plenty going on without crushing yourself under word debts.
Best of luck to you.
Not at all telling you how to live your life. When you mention a dream you want to realize, I translate that to goal. But best of luck.
Over the past year and a half, I've taken to Journaling a lot more and just generally writing my ideas down for the benefit of freeing up room in my mind. Sometimes these thoughts stray towards fiction and I've been thinking about amalgamating them into some sort of story. I've been mostly using my phone for this purpose since it's always close at hand. You can even dictate to your phone to get the thoughts out.
I guess what I'm saying is that the opposite approach to "sit down and write" is to always be ready to write and just have the tools you need with you. You could at least get a thought you've been processing out of your head and look at it later for further analysis.
Maybe this isn't as effective for others as it is for me, but it's what has been working for me.
I second this. I carry around a small notebook and pen everywhere I go so that if I have a thought I want to break down I have a place to put it. I write theory so it's different from the kind of writing you do, but I think some of the same methods might still apply. I often use a tool I call 'mind mapping' which I start with a key idea and then draw lines from to break down further, which I could see being potentially helpful for world/character/sceen building. Also outlining has always been a great help for me in guiding where I will go next. And a final note, I find sitting down to edit when I can't get in the writing head space often helps me get into that head space and break back into writing. So even if I find my well seemingly dry for the chunk of time I've scheduled for writing, by spending that time at least editing and combining through what I've already done, I'm keeping myself in that space for the duration and allowing myself the greatest opportunity to keep going.
Just finish a small novel. Set a goal for 50k and start writing. 1k a day and in 50 days you'll have a shitty novel.
This is how I started and I wrote a lot of shitty novels. The first 2 or 3 are sitting on a hard drive never to be read because they're so bad, but the 4th time around I was slowly starting to get somewhere. I now do KDP for a living and have written someone in the ballpark of 50 novels over various pen names, all in the span of about 7 years. I write pulp genre fiction so I'm not going for perfection or earth-stopping prose, just get a story down and do it fairly well.
I actually think I'm a half decent writer now. I write on the rat wheel of frequent publishing so I stick to one tight draft. My books would be even better off with multiple drafts, but I don't want to get better at writing one book, I want to get better at writing as many as possible. Only way to do that is write, publish, repeat. You might even make a job out of it.
Since nobody has recommended it so far, I highly suggest integrating an LLM like ChatGPT into your creative process.
It's excellent for brainstorming ideas. It will always start with the most obvious and uninspired ideas, but if you ask for 20, there can be legitimately good ones in the middle.
If you're struggling with your plot, ask for advice.
If you literally have no idea what to do next (i.e. writer's block) go braindump what you have into the model and ask for possible directions.
If you have a section written, ask for critique. Then give the LLM a persona in your target audience and ask for another critique. Experiment with different personas for the "critic" to get feedback from different perspectives.
Before LLMs, the biggest challenge to learning to write creatively was a lack of people willing to give careful critiques. Reading thoughtfully takes a lot of time and concentration. With LLMs, this barrier is now gone, and the feedback loop for improvement is extremely tight.
The worst and least fulfilling way to use an LLM for creative writing is to just ask it to write for you. Without an extensive system of instruction and iteration (though I am working on a system for this), you will get boring but OK crap that will stifle your own voice.
However, LLMs can also be a tool to multiply your own creativity, turning written prose from brittle stone into malleable clay.
Different strokes I guess because I personally hate the idea of using AI in any creative capacity!
As a programmer, I'm starting to see actual written prose as compiled binaries. It's laborious to write them and they're fragile, brittle things to change. So instead I'm devoting my energy towards refining my worldbuilding documents—a proto source-code if you will.
With the right system of guided analysis and iteration, I think it will be eventually possible to use a worldbuilding "seed" to generate experiences ranging from customized stories to immersive movies and games.
I suspect that this is where entertainment is heading in the next 50 years. It's a level of abstraction higher than "writing" and it meshes well with VR.
Of course, LLMs in their current state are still very useful for actually writing, like I originally described.
I think to be a good writer you need to hone those skills yourself instead of relying on AI. It’s kind of like using PEDs in sports, to me. I certainly hope this isn’t the direction that writing is headed, and I don’t believe it is - I truly believe there will always be higher value from words written by people than by computers
Couldn't resist; expanding on this:
Written interpersonal communication (convenient case study):
Two main functions of fluff
How do I ensure you don't misinterpret me?—I say the same thing multiple times, to ensure that the point gets across
1 is, if performed by a robot, completely useless—what is the point of forming an interpersonal relationship with a robot?
2 is also useless if performed by a robot: doesn't know what you were thinking if ambiguous; just makes something up to fill in the blanks
I sometimes make notes to myself in format like this. I already know how amazing I am and I am not communicating w/myself but creating mnemonics for memories
Corporate copy is filled with ingratiating fluff. It's dishonest. No relationship to be had with a corp
Internal corp emails often very terse and to the point—usually appropriately so, in context—the push for ever terser, conciser, clearer language comes from this crowd. Appropriate outside of context? Probably not
(translation and expansion)
Sorry for offending you!
In regards to the terse communication style, its effectiveness is questionable. Reading such an information-dense excerpt is incredibly taxing, since there's no room for error in interpretation. That may be more evident in my version above, since it will not be familiar to you.
This is a very interesting subject, and was a fun experiment! When I work on my worldbuilding, I try to make my outlines as dense as possible in order to reduce the token count. I spent some time today seeing what sort of "translations" into plain English I could generate from the above outline using airoboros-l2-70b-gpt4-2.0. (Ultimately, the results were good, but not good enough for me to paste into this post.)
The density makes it harder for humans to understand. Does this reduce the comprehension of the LLM? Interesting question, but the answer from my testing today is "not by much" (at least for 70 billion parameter LLaMA 2).
When you take out the "fluff", you lose a number of things. First, you no longer have the in-context clues to the meaning of words: their definitions as well as how they relate to the flow of the thought. Additionally, you lose the natural transitions between thoughts. Finally, you present words in a way that is in an unfamiliar pattern. A familiar linguistic pattern is effortless (subconscious?) to decode, whereas unfamiliar structures require concentrated effort to parse.
"Redundancy" serves as error-correcting code. Muffled or misheard words in spoken language can still be interpreted correctly. I think this also helps with speed reading, since you can skip words while reducing the chance of missing an important detail.
I don't mean to belittle prose. It may be brittle, fragile, and (always, ultimately) flawed, but it can also be exceptional, awe-inspiring, beautiful, and artisanally crafted. It's also, up to this point, all we had, since computers couldn't actually manipulate abstract "ideas". Now they can (with caveats).
The heart of storytelling lies in concepts, worlds, plots, and characters—elements that can currently manifest as books, movies, TV shows, games, theatre, etc. But with the rise of generative AI, I think we will be seeing modified forms of these media that are interactive, immersive, adaptable, and customizable. I predict that the main energy of the next generation of creatives will be focused on inventing and fleshing out these new forms of media, though it will never entirely displace the classic forms.
Artists are not going to stop sketching, just because a computer will be able to create almost any detailed image imaginable. Writers are not going to stop writing, just because a computer can write an infinite number of well-structured stories! They will continue to be creative, and direct the computer to fulfill their creative vision, because the pure computer-generated versions will not communicate their vision well enough.
I'll admit I am not a lover of prose. I prefer to focus on characterization, conflict, and dialogue, having more experience with screenplays than books. That's why I am taking such an orthogonal approach in the first place!
The comparison actually goes even deeper, and is more apt, but in my direction! While hand-written assembler has the potential to be faster than high-level code, it is severely limited in the complexity of computing problems you can tackle in assembler subroutines.
Once you no longer have to worry about memory addresses and registers, and can use functions, objects, garbage collection, etc, you can can build enormously complex programs. For example, LLMs could never have been produced if we were all limited to hand-written assembler. (Machine learning is all done in python, which is quite terrible from a performance perspective.)
In terms of performance, all you need to get 99%+ of the performance of a hand-written assembler program (caveat: not memory usage), is to do performance profiling to find the hotspots (usually in loops). In those hotspots, you can write human-optimized inline-assembly.
How does this compare to LLMs in writing? Perhaps a writing program uses LLMs to generate the base text, and then find the "hotspots". I imagine those spots will be the major points the plot turns on. In these spots, you "hand-optimize" the prose and dialogue with your unique human ingenuity and creativity: to make the most impact.
This is not a new pattern. In drawing and painting, it's a waste of time (and a mark of a beginner) to plaster detail all over the canvas. You want to use broad strokes, implied edges, and vague impressions for the majority of the space, and only detail the focal areas and focal points. Doing otherwise is a waste of your time and will generally be distracting to the viewer. That time is better spent on the composition, big-picture, and focal points.
Now, the question is, if LLMs will enable a jump in writing complexity similar to going from assembly to high-level languages, what might that look like in practice?
I think the low-hanging fruit, once the technical kinks are worked out and the context window of the LLMs are sufficiently expanded, is assisting authors in creating sprawling epics of massive complexity, while still being coherent, internally consistent, and well structured.
Caveat emptor!
I did not write tersely or densely, and instead ended up writing quite a lot.
Dear to my heart though the problem is, when I mentioned compilers I was not so much talking about the vagaries of code generation as high-level, structural, algorithmic knowledge and transformation. LLVM and its ilk are not even trying; for the non-strawman attempts, see things like computer algebra systems (ok, pretty much just mathematica), SMT/SAT solvers, DB query plan optimisers, and theorem provers.
A story may have moments of greater or lesser intensity (or emotion, or spectacle, or what-have-you), but that does not mean that the former are 'hotspots' in the sense of a poorly performing application. Recently, I was reading Gardner's The Midnight Reader, and, on a few occasions, had to go back and reread a paragraph or a few of them because I realised that it was perfect—every word perfectly placed, furthering its own goals as well as the goals of every structure around it. Not that the rest was poorly written—it was quite good—but it did not reach those heights. I despair of ever writing that well, but I don't think Gardner picked and chose those paragraphs to focus on. I don't think he could have. These were fairly mundane scenes; which mundanity is certainly rather important to the story's overall themes, but I did not notice anything particularly distinguishing those scene's content or placement with respect to the others (it is of course possible something went over my head). The muse speaks as she wills—you cannot control her—but that does not mean you are powerless, and you will much more clearly see the role of a word with respect to its mates if you have, you know, written the mates. You also cannot write that well without practising.
Or take Tolkien, whose writing has an incredibly consistent quality (last I reread Lord of the Rings was a few years ago, but I think I could only find minor fault with a couple of commas—and who am I to judge his comma choice, anyway?). This gives the text an intentional quality. But this intentionality does not just apply to the composition of the words; it applies to every aspect of the work. The best English teacher I ever had, in high school (alas, our personalities clashed, but hers was still probably the only worthwhile class I took in high school), was fond of saying: style is content. It is true: the stylistic ideals are and reflect the conceptual ideals; you cannot create a work with such a consistently intentional feel if you do not so intentionally choose every word. The language is also a draw in and of itself—analogous, more or less, to the death-by-a-thousand-cuts model of performance which has been credibly proposed as an alternative to the hotspot model—it is a joy to read. Nor do I think—regardless of whether or not his work is moderately offensive propaganda—that Tolkien's artistic vision was compromised, bogged down, or distracted by his focus on language—rather the opposite!
Ok, one more example. I read a comic book which had been translated into English—actually, I read two different translations, one of which was markedly better than the other. In the better translation, I marked one line as being the best in the whole work—it was subtle and a bit understated, but poignant and funny and, in particular, I felt, reflective of the translator's understanding of the character she was depicting and the cultures the character and the reader fit into. In the worse translation, it was filler. That's all it was: something to say to get to the next panel. The translator of the worse edition might have credibly claimed that the line in question was not very important, but that would have been more convincing had she written the comic herself and elided it or changed its role; under the circumstances, being that translating the text of a comic book forces certain constraints, I think her choice diminished the quality of the work. (Not to imply that that is in fact what she thinks; she might disagree that her choices were worse, or she might have been under time pressure...)
Drawing and painting are not really my media, and I do not claim to be able to speak about them authoritatively, but I will conjecture. Time is different from space; a painting is seen all at once. Hence, it is more like a concurrent system than a sequential one, and choosing an area to attack is more like finding bottlenecks than hotspots.
And—this has been my sentiment from the beginning, if it was not clear—more power to you! I just don't see what you hope to gain from having robots write prose for you. I think that (for example) an interactive map showing the development of political (and e.g. geological, meteorological) events over time in some world could be rather interesting to explore.
That sounds plausible to me. But it also sounds like a rather different application: running queries against an existing body of work. I'll link my previously expressed sentiments regarding grammarly and spell-checkers. I also don't think you would need such a tool to do such structuring, and it might be suboptimal for many aspects of the problem; I recall a demo from Lenat using Cyc to analyse Romeo and Juliet, and you could use much more primitive (or specialised) tools too.
It is taxing! I do not think that there an argument to be made for a monotonic reduction in comprehensibility, though. The tradeoffs are similar to those of APL—the denser your ideas, the more you can fit in your head at once.
You’re going to get some pushback for this but I think this kind of method is only going to get more popular and useful over time with Microsoft building “co-pilots” into windows and all their office products.
I can empathize with the pushback TBH. Generative AI (specifically image generators) sent me into an existential crisis when it took off. Before that I was very focused on digital art.
It makes it much harder to pick up my sketchbook nowadays.
When I do finally get back to character sketching and drawing (after I'm satisfied with my worldbuilding and writing) I plan to liberally abuse ControlNET+StableDiffusion to finish and transform my drawings instead of doing it manually. Maybe even train some (personal & private) qloras on my favorite artists to better analyze their styles for integration into my own sketching process.
I apologize if this is not the case, but your reply sounds like you missed an important part of my comment. Even if you have no interest in generating ideas or prose with LLMs, getting immediate critique from various perspectives is invaluable:
I suppose the immediacy of feedback is one advantage of using AI language models for critique, but I wouldn't say it's the best way to get critique at all. It depends on what kind of writing your doing, but in almost any case there are existing entities that can do the same thing from a human perspective rather than from a perspective imitating what it thinks a human would say. If you write fiction, there are writers groups and workshops everywhere for every skill level which give you extremely pointed feedback on your story from real people with real-world perspectives, if you're a journalist it's probably a conversation you should have with your editor or coworkers, if you're a student it might be a conversation best had with your professor as any professor will accept early submissions for review before the final deadline.
There are plenty of situations where you could use something like ChatGPT to benefit you, maybe you're a first-time writer and you want to go into a real workshop with some idea of what to expect, but that's also something that could be accomplished by a lower-stakes online workshop, with the added benefit of being able to talk to real people and get relevant responses.
That's not to dismiss the usefulness of AI, but rather to say there are better ways to do the same thing!
There's really no value in this. A language model isn't a person. It's not going to give nuanced feedback to a piece of art. It's designed to predict the most likely next word in a sentence. Please don't credit it with anything more.
I'm going to give some less prescriptive advice than others in here (though I'd echo the idea that the only way to write is to sit down and write):
Start reading more. The more you read, the more you'll feel inspired, the more you'll feel the pull to sit down and write.
I highly highly highly recommend Anne Lamott's Bird by Bird. I think it is the one "craft" book every writer needs to read. It's far less about how to write and far more about how to make writing a part of your life. I teach "Shitty First Drafts" in my freshman comp classes, because the most often-voiced concern new writers have is that their writing isn't up to par. But no writer spits out a perfect draft on their first try. They have to sit down and get the words onto page so that they can edit them into shape. Give yourself permission to write garbage. Don't let the elusiveness of perfection paralyze you and prevent you from even starting.
Another great philosophy Lamott advocates for is the one-inch picture frame. Focus on something small: Today you're just going to write the first page. Today you're just going to write the first sentence. Today you're just going to brainstorm some titles. Inevitably, you'll meet your benchmarks, and more likely than not, you'll feel pulled to keep going. And hey, now you've overproduced. Now you've got some work to edit.
Instead of telling yourself you want to write a novel this year, start small. Write a story in a month. Write a story in a week. Let yourself build up to your larger goals. No one wakes up and runs a marathon without building up the stamina for that distance.
This might sound silly but I started writing and posting erotic stories under a different name. It's easy to write, the expectations are low. I used to get stuck on a paticular word choice or a bit of scene description. But since the stakes are so low I was a lot more comfortable just letting the words flow.
I'm getting a bit bored of it now, but it's got me back in the habit of writing and the mood to do some serious writing again.
What I've found useful to just get in the mindset of writing is "playing" solo journaling rpgs like Thousand Year Old Vampire. I found it really useful to have a more guided experience, like writing prompts on steroids.
As someone that also deals with depression and the lack of motivation that comes with it, making writing fun was really useful to me and has allowed me to write more when I want to. Bonus points for the ideas I have from playing a journaling game inspiring my writing.
Hey, that's the same situation I was in literally days ago. I used to write a fair amount but rarely finished anything, then work got busy and I fell out of the habit. The method that I found most effective in the past was to set aside a specific amount of time for writing each day. It didn't matter how many words I put down on the page, but I had to focus on writing and nothing else.
Another trick I've read about and tried was called freewriting. What you do is take a piece of paper and write down anything that comes to your mind. It doesn't have to be coherent, the goal is to get into the right headspace for writing. Once you finish the page, move on to the project you're actually working on. This one was a little more hit-or-miss for me, but you could give it a shot too.
For the past three days I've scheduled myself 30 minutes to write. So far, it's been easier getting back into writing than I expected. That was nice.
I haven't written in a while, so who am I to preach? Still, I feel I have found a good approach, inspired mostly by the book Atomic Habits. My fundamental belief is that writing is easy; you literally just sit on your butt and go clickety-click on your keyboard. Consistent writers don't have any kind of superboosted willpower muscles, because willpower muscles is a myth, they don't actually exists. What they have are some habits which allows them to simply write without having to think much about it.
Schedule
Some General Advice
Tips Regarding Writers Block
Try a different medium! Whether that's poetry when you usually do prose, sci-fi when you usually do grounded slice-of-life stuff, or (and this is what helps me) take your idea and do something completely non-writing with it. I started making collages and assemblages a while back, and that helped with the book I'm writing in a huge way..... Now the book is illustrated, and nearly finished, and it's so freeing to want to work on my project, and have options to focus on if I'm not feeling writing that day.
I've never been one to "write everyday," "just sit down and do it!" ...fuck that. Doesn't work for me. Too much pressure, makes everything feel bigger and harder than it needs to. But even on my worst brain day, "I want to work" can mean anything from writing to digging through books for collage pieces, to digital graphic design, to resin (I'm layering the collages with resin to add more depth). Or even just reading about the subject matter, that counts too.
And regardless of the medium, make intentionally bad art. Write a terrible story, draw something your 6 y/o cousin wouldn't even be proud of, take a block of wood and whittle the shittiest poking stick you've ever seen. Have fun with being creative in general, and the writing will come.
it's tough to do so when you no longer feel life's worth living.
I think people here are glossing over this. If you are actually struggling with existential despair and or clinical depression, professional help is something I would recommend trying. Sometimes depression is brain chemicals. Other times it is overwhelming grief. Other times it is about loneliness or having been battered by abusive relationships or work places. But based on your replies here, you need to deal with your depression to move forward.
I wish you the best. It's hard to do.
Try picking up Stephen King's "On Writing". Fantastic book and very inspiring imo.
Highly recommend On Writing. Helped me get out of a writing slump a few times.
I'm coming to this very late. I'm going to apologize that I've only read a couple of the replies that seem very good but I wanted to add in a few things that work for me.
I tend to write in waves. Super productive highs and do nothing or fuck around elsewhere lows. Seems like you're in a low point. The best thing that tends to get me back into the groove is listening to songs I've picked out for scenes/characters. My normal playlists consist of a lot of beat heavy music which for the most part doesn't jive with the slice of life/soap opera story I'm writing. Listening to things that remind me of characters/scenes makes me think about them and how I want to get them on paper.
Oh, not sure if anyone else said this or not but you do not have to write linearly. If there is a scene or character in your head yelling to get out, write that. You can always go back to the point you were working on before plus it gets that first terrible version out of your head to be improved later.
So yeah, don't try to force it. Nudge it a bit with some music (or something else that makes you think about it) and see how it goes.
Good luck!
Big same.
Totally get that as well. My whole project started as a webcomic, but I can't draw and the folks who I wanted to do it are out of my price range AND have their own long ass stories going on. So, it was a Thanos 'fine, I'll do it myself' moment.
Still agree. But its fucking hard to get it on paper how I see it in my head. lol
Yep. My spotify discover weekly has essentially become 'what synthwave am I adding to UCSO this week?' when that isnt even my big project. It takes EFFORT to swap back over to the main story music.
Yo, THAT is an excellent analogy!!!!
See, you've got what it takes to do this. Just need to push your brain out of its own way sometimes. lol
Do you share your stuff anywhere? I'm still pretty new to this (4 years or so in) but I'd be willing to take a look and share thoughts.
I got myself back into doing it by setting up two things: A consistent schedule, and less limitations.
On the first point, just resolving to regularly do it will go a long way. It doesn't need to be a long time, either. For me, it's 30 minutes every other day. Regardless of how I feel, or whether I have any ideas, I will sit down and write something out. A lot of the time it's just a stream of consciousness, or something akin to a journal entry, but occasionally it comes together real well and with that I'm apt to continue. I keep these things all in one place and refer back sometimes.
On the second point, I let go of really any expectations to produce. I am not writing a book. I am not writing a story. I'm not even writing a scene necessarily. The point is to just do it in some form or other, and if it takes the shape of those things, cool. Forget the audience, don't even think of the audience, just do the thing and see what that is, has been my mindset with that.
What about writing some nonfiction? Then you're writing, but you don't need to be creative in the same way. Write a walkthrough for a video game you like, or idk, a guide to getting set up on Tildes or the Fediverse. Maybe it'll help with the habit enough that you'll start feeling the spark again, and get back into fiction straightaway.
I'm not sure I have much advice for the OP. I've written a few novels, and self-published them through Amazon, but mostly for my own satisfaction. The motivation for my writing comes from within. My first novel was an exploration of a recurring dream, where one character kept showing up. It was simple curiosity, and I had to find out all that I could about him. I could picture the first scene, but I didn't know why he was there or what the setting was. So, I started typing.
Some writers build outlines, and require structure before they can write. I'm just the opposite. I have to explore as I write, and I'm in a constant state of wonder, as new scenes and new characters appear from my imagination.
Determine your process and dive in. The first few drafts will be garbage. None of us are Mozart in Amadeus, and it won't appear on the page as perfection from above. Don't get discouraged in these first few rounds of the fight in other words.
OK. It looks like you know your own process well enough. What you're lacking is a spark. I'd suggest keeping your mind open to events around you. Keep reading and observing people. If you find something interesting, build an outline for a story in your mind about it. You don't have to actually write it down. It may not be compelling enough for that, but just working through it in your mind will help.
When you find something that is truly worth your attention, put it on paper. Again, you don't have to plan an opus. You're trying to get back on the horse, not win a race. Little steps are what you need at first.
I agree with some of the others in here, that writing every day is not helpful for many. It becomes mechanical, tedious and counter-productive. What you need are subjects that you care about. If you are indifferent to your subject, how are you going to make anyone else care about it?