Turniper
5d ago
A couple thousand words away from finishing my second novel (Xianxia series called Between Beast and Buddha). Looks like it's gonna clock in at just shy of 200k words. A bit meandering in places, and darker in tone than the first. I'm a little too close to the work to have an objective opinion on it yet, but I'm mostly satisfied, it's a necessary part of the story I set out to tell. Reader reception has been mostly good, but I'm not sure how the ending will land.
I really should get back to editing the first one. It would be nice to finally officially be able to say I have a book out there, instead of explaining web serials every time someone asks me about it at a party.
Turniper
22d ago
In no particular order:
Yes, I know, people hate it when I respond to advice like this. But it's part of how I was raised — to automatically find flaws in and try my best to shoot down any proposed plan, because anything that can go wrong will go wrong, so you must spend every waking moment considering every possible point of failure and taking measures to address it, always have a Plan B for when Plan A fails, and a Plan C for when Plan B fails, and a Plan D…. My first instinct whenever anyone suggests anything is to poke holes in it, find every possible point of failure, and respond with every reason why it won't work.
Man, you can respond to things however you want. But I do think you should try some. If you're already so worried about doom, what could failing really cost you?
Who pays for that? And through what platform? Don't all the payment processors and crowdfunding/donation platforms/Patreon clones pretty much forbid NSFW? And I'd need to somehow obtain skill with AI art generation, enough to be competitive with all the other people out there with far more skill and experience. (Plus, I'm hardly the person to know what constitutes desirability or quality in pornographic output, given my own unfamiliarity with and lack of consumption of porn.)
Subscribestar and Patreon. Patreon's rules are actually pretty lax, basically no underage, no rape, and no real people. Not sure the other platforms even go that far. There are people clearing tens of thousands a month just reselling curated image sets. Probably won't be a thing for more than a year or two, but they're making money now. This is a 'skill' that didn't even exist two years ago, you're not competing with ML engineers, you're competing with random degenerates who asked someone on discord what model they used and how to set it up.
Can you be specific what you mean by "repackaging" here, and is that something people actually pay for? Plus, it's my youngest brother who's the artistic one in the family.
You say you sell images suitable to be book covers. You don't say those images are largely AI. A lot of books come out every year. The same people who were previously willing to pay 200 bucks for some stock images photoshoped on top of each other (Go browse some romantasy covers on Amazon) are now perfectly happy to pay for AI ones. Fiverr is the starting marketplace, but anyone who gets real business just makes their own portfolio site eventually.
I talked to the insurance company recruiters at the Job Fair a couple of times, discussing my math skills, and that stuff all gets done outside of Alaska. Plus, I lack actuarial training (and whatever degree/certification) and such.
Actuaries are a very small percentage of finance. Remote is also very available these days. That being said, a 4 year degree (Literally any major) is very important for these jobs, and if you don't have one, it's probably not a good path.
actually making book covers
(Doing art. Print shops are a dying breed, but people still commission covers)
Edit: I guess you might say my "dream job" is "prophet of doom," but nobody pays for that.
PEOPLE LITERALLY PAY FOR THIS. Seriously. People love the evening news, they love their various youtuber naysayers, and all those AI Doomer substacks. If you actually think you're good at it, go put your money where your mouth is and start seriously seeking a broader platform instead of contenting yourself to be a replyguy on niche reddit offshoots. It doesn't matter how good your content is, if it isn't top-level on a large site, it will never gather an audience.
Turniper
25d ago
You just want a large list? Data entry, content moderation, sales, repackaging AI generated images into book covers, actually making book covers, teaching english via zoom, customer support, all the ancillary shit software companies need that isn't code like project management, basically 70% of finance and insurance, streaming, selling curated AI generated porn. I'm sure there's more but that's all I can think of immediately without googling.
Turniper
26d ago
We've had this conversation before, like 20 months ago, and you detailed your many limitations, though their specifics escape me now. I still maintain that if you can write with this clarity and frequency on the motte you are theoretically capable of earning income through intellectual work, you just need to find a way to cross the gulf between it impacting your benefits and it being enough to live on. There are huge numbers of jobs that can be done entirely via interacting with a computer or phone these days. I personally have made money writing software, tutoring high schoolers, and accepting donations for fiction. LLMs are changing many of those niches, but at least for the moment near as many opportunities are being created as are being destroyed, though that may not hold forever. But for the moment, if you want freedom, you'll need to earn income. That's the way the world works. If this 'slaughter' you cannot speak of arrives, well, that might change as some nation begins accepting US asylum applications. But until then, not a single nation is interested in acquiring dependents with no ties to their citizenry.
Turniper
29d ago
Design isn't the bottleneck though. We have designed fabs before, most of the hard part is solved. It's actually building these massive facilities, supplying them with water and energy, keeping them at pressure, staffing them with technicians, and troubleshooting the million small issues that can eat into yield ratios that is hard. It would take some world changing advancements in robotics or AI just leapfrogging our current understanding of how to build chips for them to make a dent in the difficulty of the process.
It is currently 142,448 words, pending a little bit of churn from ongoing edits.
More options
Context Copy link