From Kristine Kathryn Rusch:
At the beginning of August, Patreon had what some termed a payment meltdown. Some creators couldn’t access their payments. Banks notified some Patreon backers that their payments were being flagged as fraudulent.
Patreon claimed that the problems weren’t one problem; they were two problems. For the creators, the problem was a payment partner of Patreon that wouldn’t let them cash out. For subscribers, the problem was a lot more arcane than I want to go into here.
The secondary problem for creators, though, was that the subscriber issue brought the subscription to the attention of the subscriber. The great thing about subscriptions is that once people subscribe, they tend to forget how much they paid for the subscription.
If that subscription in any way loses value in the mind of the subscriber, or if their financial situation has changed, or if they had completely forgotten the subscription existed, having that subscription brought to their attention makes them reevaluate it. When they do that, many cancel. In this latest kerfuffle, one creator claimed they lost 300 subscribers.
A friend on Facebook gleefully reported all of this, and mentioned this was why they avoided all outside platforms, choosing to go through their website and with the systems they had built. (And yes, I see the irony of a friend using a platform to claim that they never use platforms. Don’t go there.)
They did have a point about platforms, though. We’re moving to our own online store for a variety of reasons, but one is that it gives us a cushion should one of the bigger online retail platforms change its way of doing things.
Another friend complained about online store platforms like Shopify, and said that they preferred to design their own. Turns out when I looked at their site, they had defaulted to WooCommerce, which we had tried and didn’t like.
The first friend’s point about outside platforms caught me, though, and got me thinking. My initial gut reaction to both of these folks was that my website has had a lot more problems over the past 25 years than I’ve ever had with the platforms. The idea of depending solely on my website scares the bejeezus out of me.
Having only one point of contact for all of my work is too risky for me, even if I supposedly control that one point. I don’t control the platform on which my website rests. It’s also taken forever to get someone to help me rebuild the website here. Some of that is me stalling, but some of it is finding the right fit.
Still, outside platforms have their own issues. At Christmastime last year, Amazon caused a lot of turmoil by dropping its newspaper and magazine subscription service. Some magazines were invited into the Kindle magazine program, but others were not. As Neil Clarke of Clarkesworld wrote at the time:
Earnings from Amazon subscriptions provide a varying and sometimes significant portion of the revenue that these publications require to stay in business. If you don’t already know, genre magazines are subscription-driven, meaning that subscriptions make up the bulk of their income. Some people think advertising is a major source, but it actually represents a tiny fraction for us….
None of these magazines are entirely reliant on Amazon, but as the largest ebook retailer in the field, the cancelation of this program will hurt and in some cases, hurt badly. Badly enough to shutter a magazine? Maybe. It’s too soon to tell …
Platforms change all the time. They make decisions that have a huge impact on the people they partner with. Amazon’s change was a cost-cutting measure driven by severe layoffs in the fourth quarter of 2022 and into 2023. The Patreon problem had a lot to do with a platform they had hired to help them process payments.
PayPal also changed its fee structure in late 2022, closing a loophole that a lot of businesses used as well as upping some fees. I’m sure online store platforms like Shopify will change how they do things as well over time, and many of those changes will harm some group of their partners.
Sometimes these companies do things seemingly en masse. They’re not. (Well, some of them might, but mostly, no.) Generally, they’re responding to market conditions which were not favorable in the last half of 2022 to anything online.
. . . .
The online boom started roughly fifteen years ago, when I already had an established career. Frankly, it saved my novel career. I couldn’t sign the contracts from the major publishers any longer. I couldn’t take the rights grabs.
I was looking at a short-story-only career. At that point, I thought I could still bring in some money from royalties. That changed as well.
I had assiduously gotten my rights back for almost every book that was an original. I owned the books and could relicense them; I just wasn’t sure how to proceed.
Then the online revolution, from viable ebooks to print on demand to easy-to-produce audio to podcasting to video podcasting to video production—well, you were there. You know.
All of these things go on various platforms. I have too much product to put it all on my website or on websites controlled by me. I would never get to all of it.
Dean and I had to hire people to help us put our books and short stories up on the various platforms, otherwise we would have had to give up writing. That’s why we started WMG Publishing, so that we had people to handle the various platforms.
We have a good staff, but they’re horribly overworked. They can’t control everything, just like Dean and I couldn’t. We’re constantly researching and finding the best way to put our product out into the world. We do a lot of experimenting. That’s why we were on Kickstarter in 2012. It was an experiment—one that worked. We did a variety of experiments that did not work.
. . . .
I decided long ago to use other people’s platforms for a lot of the work that we do. We use Amazon and Barnes & Noble and D2D and Bookfunnel. We use Mailchimp and Kickstarter and Teachable and YouTube. I use Patreon. We now use Shopify. I’m sure I’m missing a lot of platforms that we use. I do know that there are many that we are investigating and some we used to use. There’s a lot we’ve tried and a lot we abandoned and a lot that we have learned to love.
But that doesn’t mean we’re going to use them forever.
Link to the rest at Kristine Kathryn Rusch
Tools exist for using…
…while they are useful.
As she said, once they lose effectiveness iy is better to move on than hope for change.
(“Hope is sitting around doing nothing, waiting for somebody else to solve your problem.” Geoff Johns, CA 2008)
Today Microsoft is holding a dog and pony show on what they are are using LLM for, such as an integrated LLM Copilot app for Bing, Windows, and Microsoft 365/Office.
Already we got a hint of things to come for audio. Audiobooks are going to have a TTS moment real soon:
https://techcommunity.microsoft.com/t5/azure-ai-services-blog/introducing-super-realistic-ai-voices-optimized-for/ba-p/3933744
Check those samples.
Those, and other optimized voices are available to any customer (read: software developer) working on AZURE. As noted:
“Microsoft offers over 400 neural voices covering more than 140 languages and locales. With these Text-to-Speech voices, you can quickly add read-aloud functionality for a more accessible app design or give a voice to chatbots to provide a richer conversational experience to your users. In addition, with the Custom Neural Voice capability, you can easily create a brand voice for your business.”
Microsoft is not alone in this. Amazon is, too. And if Google isn’t, they’ll soon have a Me2 project.
This points in two directions:
1- The most obvious being an LLM based audiobook creation app: set the desired voice parameters (gender, nationality, personality, vocal range, narrrative tone, etc), feed it a document, receive an audiobook.
2- Less obvious, way better TTS ereader apps that might render audiobook editions niche, if not quite obsolete.
Note both of these are narrated audiobooks, not *dramatized* audiobooks. Those will take a bit more work to define the voices for the entire cast. Although, “AI” has the potential to use generic voices for spear carriers and extras. Maybe the entire cast???
The tech is practically made to order for corporate tradpubd, if they ever notice, and if the economics work out, even more for indies.
Again: chatbots are the least of what’s coming.
I keep some track of the gaming scene, even though I don’t play any of them. (Starfield, eventually – but apparently I’ll need a new graphics card, which isn’t currently in the budget.)
Some of the things being shown off, like the latest Unreal Engine, make me think. I predict that before the end of this decade, it will be possible to create a near realistic animated movie of an entire novel for less than $10K (2023 dollars, who knows how much more Bidenflation is going to cost us).
AI will take care of all of the currently pesky details – walking, running, climbing, matching mouth to speech, etc. I’m hoping the old brain hasn’t completely calcified by then, so that I can learn at least some of it.
I expect to be on STARFIELD until DREAD WOLF comes out and that might be 2025.
Which is fine by me.
Digital forge summed it up best: SKYRIM MEETS MASS EFFECT.
Yeahhh!
As to playing it:
1- Are you wedded to playing on PC? The reason I got on XBOX was MORROWIND and getting a console was cheaper than the minimal graphics card required. And I got to play on the living room TV. FYI, in a few weeks weeks the white XBOX Series S should hit holiday sale prices of $249 or less. Last year it hit $229 and in this week’s FTC leak, MS admitted the thing is designed to sell profitably for $200. (Some folks are buying them for retrogaming because they have an official DEVELOPER mode where they work run UWP Windows apps and net apps.) My sister has one and it handles video at 4k and most games at native 1440 with optional upscale to 4k. All glitch free.
2- Alternately, I tried the cloud version on my last gen OneX anf it ran smooth, albeit upscaled from 720P native.
When you get to it you’ll find it rolls up all the mechanics of all previous Bethesda rpgs into a millieu not unlike Asimov’s 50 worlds era. Or maybe THE EXPANSE might be a good comparo. Good company anyway.
Much fun, few bugs, no crashes 60 hours in.
About your animated production costs, I asume you’re including some hefty hardware to run the “AI” local?
Well, hardware would be amortized over several projects – many, if you were doing it as a business.
The price/power points are still coming down – and I’m allowing six years here. Intel just recently announced a processor with optical internal paths, and (IIRC) 128 cores. Not commercial yet, but in six years it may be “old tech.”
Database storage – SSDs are coming down in price/capacity too. A few hundred terabytes will hold quite a large one (the database, once loaded, is pretty much fixed, so you won’t get “SSD wear” on them).
One constraint that might be an issue – keeping the durn things cool! My quite modest system puts out enough warmth that I don’t need to heat my office except on the very coldest of winter days. In Tucson, mind you, not Minnesota. A top of the line gaming system would probably let me get by even there; they already have to be water cooled. My hypothetical animation setup in 2030 might need to have an integrated 1,000 BTU A/C unit in it…
Oh, on the video card – I can justify an eventual upgrade as a business expense, as I do a fair amount of graphics (and will be doing much more). The IRS would not look kindly on depreciating an XBOX.
You’ve seen this, right?
https://www.cnet.com/tech/computing/three-future-generations-of-intel-processors-will-outclass-apple-ceo-says/
The road forward includes trillion transistor chips, with built in ML acceleration (which even the XBOX SS has).
Windows12 will actually require an ML accelerator chip.
That should be fun considering the mess that was Win11 requiring a security chip.
The key takeaway:
Always ensure that you have an effective alternative to any third party’s black box upon which your business reputation relies. Anything relating to “payment clearance” is a black box (as a matter of both law and practice; do you really know what steps an electronic payment goes through from point A to point Z, and who is responsible for any failure in there, and under what conditions and in what timeframe? I thought not).
It’s the business corollary of the computer geek’s understanding of backups: The time you will most need a backup copy of critical data (whether code or anything else) is when it’s either (a) inaccessible or (b) not yet made.