The real problem is not that they are buying all the hardware, but that they are buying it and keeping it in bloody warehouses without bringing it online. Because they haven't got DC capacity for it.
Has anyone else here been trying to rent some cloud gpus for reasonable money? I have. I have signed up to a dozen "providers" that advertise everything from tesla cards to b200 only to find stuff is either out of capacity or double/triple the price.
It requires skill to prompt the AI to do best possible work. I have senior colleagues that produce horrible slop and others that always produce great results. Managing the context, knowing when to stop the AI. Knowing what to question, what to "trust" in is a big deal.
Also knowing what models are good for what. There was time half a year ago when Google genuinely had a better model than everyone else. I used it for everything and 3 weeks later they lobotomise it (sorry "optimised") and I went back to opus...
I think Anthropic is like a drug dealer, giving us the sweet sweet drug for free (I don't think our $200 a month subscriptions even cover the electricity for our use) and the time to pay will come very soon...
I expect this subscription will cost $2k a month. Will it be a normal increase? Or will the "enshittify" existing models to the point you'll pay $2k to get fable 6 to do what opus 4.8 did fine in July 2026?
I wonder if August of this year will be remembered like that too. The very first "opus like" local AI model came out this August (Qwen 3.8 Flash Next). I've been running it locally since for real programming and I consider it pretty much the same as opus 4.6 in coding ability (it lacks a bit in the factual knowledge area). It even exceeds opus on some tasks.
I've tried every hyped "open source" model before and all including latest models bigger than 1T parameters are pretty much toys.
This is the first one that isn't. It can't be overstated how huge of a deal that is. No more reliance on Anthropic.
Running this model to do real work is still not cheap. I run it on a pc with 6 rtx3090s and 192GB of RAM (and I use 90gb of that ram for KV cache). The model is entirely on gpus. It runs at 55tok/s 1500 prefill for one user at a time, and about 35tok/s 650 prefill per user for 5 simultaneous users. It doesn't seem like much until one realises you manage your own kv cache. You ca leave your sessions in cache for as long as you want. You can save them and restore 200k sessions a week later in a dozen seconds.
What many people don't realise is that usage of those models skews extremely heavily towards input processing. My claude code usage is about 1.3B tokens input per week and only about 8M output. On claude code I get 80% cache. At home it's more like 95%.
If this progress keeps up, and we get a fable quality model in a year in under 200B to run at home... Those "frontier labs" will be renting all of their gpus per hour not to go bankrupt.
I am using qwen3.8-27B-UD-Q3_K_XL on a 5080 16gb card with 16gb of system ram at around 43-45tok/s with 65k context. I am using that model to set up containers on a proxmox server with two b70s and 96gb ram. The Q3 model implemented 10 different chat models, 2 image models, and 7 web based harnesses so I can compare. It can rewrite any part to improve it.
qwen3.8 is more than capable for software dev. The gemma models were terrible and fell apart during compaction. I think as long as the llm can test the results, you don't need anything being offered by a "frontier" model. The cloud AI is going to be used by people unable to run their own and they will eventually get squeezed on price.
The one tip I could give is compact before starting new steps or any action in the plan that is different than what was previously worked on. You want to manage what is in the context and don't want unnecessary work history details filling it up. You can always ask the llm to list the current plan, then compact after and do it more than once until you get the compaction <30%. This will be fixable by the harness that can choose better times to compact.
You don't want to start a new phase and have it compact a few minute after starting. This happening over and over again seems to potentially cause issues for long running sessions. Compaction slop that screws up what is in the context.
I can compare what I use at home vs paid models like astra at work and the difference is mostly meaningless.
Bingo, DeepSeek (v4.1) is horribly overhyped. In all my personal benchmarks it sits below Glm5.3 Flash. Waaaay below Qwen3.8-Flash-Next a model less than half it's size.
No, the only open weight model that really makes sense for me is Qwen3.8-Flash-Next, but it is mainly because I can run it locally with reasonable speed (prefill between 650-1400t/s generation between 22-50t/s depending on number of slots/users I configure).
This is the first model that truly competes with Opus 4.8. I'd say it may be better than Opus 4.6 on programming.
But it is very verbose when it comes to reasoning tokens. The more difficult the task the more verbose it is. Certain very hard tasks that take opus 4.8 400k tokens take Qwen3.8-Flash-Next 2M tokens... But it finishes them.
And what you loose on the generation speed you get back on input caching you can keep on for weeks.
Bingo. They are scared s*itless of local models. Soon the raising interest rates will mean they can't just buy the entire world's gpu/ram/storage capacity and lock it away in a dark room for no one else to use them.
Once prices of hardware stop being bonkers many people will run local.
People often do those silly counts looking at local generation speeds of 100tok/s and saying you can only get 8M tokens a day and that is worth so little you'll never offset the local hardware cost.
But they are forgetting about two huge things. One, the split between input/output token use is huge in typical programming. I typically use 1.2-1.6B (as in Billion) input tokens and only 8M output a week. Out of that 75-80% of input is cached on Anthropic with their pretty inflexible short lived cache.
And here local AI shines. You can save your contexts to disk so you can go back to a session 3 weeks later and load it all from cache without having to prefill. If you have the RAM and you use models like Qwen4 (3.8 flash next) that can fit 6 to 14 262k contexts in 48GB of system ram as cache.
And the second thing is you can have 6 to 14 sessions that do 99% caching and you can run a lot more input for a long time in 48gb dedicated system ram. (the number varies a bit depending on the content of the context).
So you have RAM that caches "automatically" and you can share the cache between users. Or if you can remember to save to disk (server side, the client just sends a request to save/load). But this uses both RAM and flash storage. Two things that are horribly expensive now.
However when you have a local model it enables workloads that were simply completely impossible in the cloud.
I too am in the "it's done, it works well, don't mess with it" camp.
I've tested everything. Xfce, tiling window managers, kde. Eventually I settled on Gnome, but I incorporate few "tiling features". Such as keystroke changing workspaces on one monitor and the second monitor always showing the same few "always on" apps.
I always have my terminal with tmux full screen on workspace 1. My Web browser on no 2. My IDE on 4 and my password manager on 9. Then if I watch a video it's always on 3, if I do graphics work it's always on 5 and so on.
I use ctrl-super-number to switch workspaces. If I need to move current window it's ctrl-shift-super-number.
On a work windows mashine I use the exa t same keys thanks to AutoHotkey software. So I have the exact same work flow on Linux and Windows. The only difference is on Windows workspace 1 is WSL terminal.
Ive been working like this for years. Select copy and paste works fine with left/right mouse button.
I have all my widgets I need showing the temperatures, what my pc and network are doing.
While reading it, I kept wondering why he thinks there hasn't been much innovation in the Linux WM space. And amusingly, tiling managers are one of the shining points of Linux WMs. I can't get a good one to work on Windows or Mac OS the way I can with AwesomeWM. To me, those OS's are quite behind.
Stuff like compiz was first on Linux. Woefully underused - even on Linux!
Enlightenment was amazing in the old days.
> Why would I want to change anything?
You don't need to change anything, but others do! Suppose the world decided tiling WMs were the way to go, and you could no longer run Gnome the way you currently do. Would you be happy?
The ideal is to have more features/capabilities, while not breaking existing users like you and me.
Tiling window managers aren't new though, I went through my tiling phase 25 years ago (Ion). My Mac experience today is not uniformly better than it was 30 years ago and in some ways much worse. It's true that much of the infrastructure is way better now, we have pretty good multilingual support everywhere, protected memory, better color, much better networking, but as someone who often just wants to sit down and grind through the e-mail queue that experience has been getting worse for around a decade.
Tiling window managers predate Linux (and Mac and Windows). Not to diminish the work people are doing to make the desktop better, in general I think on the Linux side things have gotten pretty good, but it's hard to come up with very many widely used substantial UX changes. Comparing to say Windows or Mac drawing an arbitrary line at 1990, networked hypertext, desktop search... What else? The infrastructure underneath is all way more advanced but the overall experience is pretty similar.
Once again, I'm not talking about who invented them. I'm talking about the innovations in tiling managers in Linux WMs. I somehow doubt you'll find something as good as AwesomeWM in pre-Linux WMs.
In AwesomeWM, you don't have workspaces, you have tags. Each tag can have its own window layout. A given window can have multiple tags. As an example, I can have my terminal on tag 1, Firefox on tag 2, Emacs on tag 3. The tags appear like your typical workspaces.
Then I consult Firefox and need to type a command into the terminal. I can quickly assign tag 4 to both Firefox and the terminal. When I click on tag ("workspace" 4), I have both FF and the terminal tiled. If I go back to tag 1, it's just the terminal - or tag 2 and it's just Firefox.
If you want to think in terms of workspaces, it's the equivalent of having your window in multiple workspaces.
But even better: Instead of assigning tags 4 to both of them, I can just say "Let's temporarily combine tags 1, 2 and 3", and I'll have a (transient) view with all 3 windows, tiled. Very handy if I just need to do it for a short period of time - no need to untag 4 from all of them.
This way you can quickly do the "union" and "difference" of different tags as needed - really quickly. Say I've got both FF and Emacs in one tag. Then I need to quickly run a terminal command and copy/paste the output. I just say "Hey, let's do a union with tag 1, run the command, and undo the union."
Once you get used to it, you miss it in any window manager that doesn't have it.
no, you were among that day's lucky 10k when someone told about tiling GUI, and that day was the date it was invented in your mind. and it's not. awesomewm itself is like 20 years old and it was a competitor to other 10 years old tiling WMs at the start of that project.
>Stuff like compiz was first on Linux. Woefully underused - even on Linux!
Compiz/compositing I think kinda was us going down on the wrong path. Generally it meant every app needed a texture (more like 2/3) to draw into, and compositing itself has increased latency.
We blinded ourselves with flashy effects only to give up on frugal resource use and display latency (as well as power use).
This whole thing is only just being fixed with Display Planes on Wayland, which are kinda poorly supported right now, requiring the right combination of HW/compositor/DE
For the last year or so I have used Niri window manager exclusively. It's out there, even by Linux tiling window manager standards, and it's the most delightful window management experience I've ever had. I'm guessing there isn't anything remotely like it on Windows or MacOS.
As you say, there is incredible variety and innovation to be found in Linux desktops. The post would make more sense if it mainly concerned itself with the UX stagnation on MacOS IMO.
That's how I felt before I discovered tiling window managers. Now my flow is very similar to yours, but since I was wrong before, I can't rule out the possibility that I am still wrong and there might be something better out there. For example, Plan 9 had some interesting UX ideas like making the window manager infinitely nestable and mountable over the network. It also had the neat behavior in acme where to execute a command, you middle-click on some regular text. In the end, I'm not a fan of Plan 9's UI because of the reliance on the mouse, but I think it shows there's still room for novel innovation out there.
What's so interesting about tiling managers? I've been using i3. But my usage is so simple (I guess) that I have just a few windows maximized in tab mode, usually just two: browser and terminal. I rarely need to split vertically. To be fair I like i3 simplicity. But I'm not really "wooed" by i3. I may be missing something?
For me it the epiphany was that I either want one full screen window, or several windows side by side, I never want one window to obscure another window. And with our modern wide monitors I almost always want several windows side by side. Why do I have to line them up by hand?
The downside is that there are many poorly programmed applications that are unable to adjust themselves as the window manager requires(calling you out steam, or it just may be me but steam was not happy with a tiling wm), and model dialogs, which always suck, suck even more when tiling. But most tiling wms's support a float mode for just that occasion.
Tiles were a common early metaphor, And I think tiles are generally more useful but floats looked cooler in the ad copy and so that is what ended up as the dominant desktop metaphor.
On the subject of one window partially obscuring another window the only tech that makes that halfway tolerable is "don't raise on click" and "point to focus" This lets you look at the upper window while manipulating the lower window. two features not found on the major desktop systems. X11 has it's warts, and is very weird. But it is miles ahead in actual usability over it's peers.
Depends on your screen size. On my laptop, I follow pretty much the same pattern you do, so I actually stopped bothering with XMonad.
But on my desktop with two 27" monitors, I find tiling incredibly useful. It's really nice to be able to quickly arrange a mix of Emacs windows, browsers and random other programs to use all that space. If I need to closely read through some documentation while also looking at code, I can quickly open a browser next to the code window. Later, when I want to focus on just code, I can send the browser to the other monitor or to a different workspace altogether. And I can juggle "side" stuff (chat windows, OBS, music, whatever) just as easily.
The key thing is that XMonad with some lightly customized layouts gives me the operations to do all of this quickly without needing to think about it. It only takes a couple of keystrokes to get to the state that I want at any time and, after a pretty short time, those keystrokes became muscle memory. Using a mouse to drag things around is more intuitive and flexible, but it requires shifting my attention far more than just using a single keyboard command. (And thanks to ADHD, that kind of distraction can be a real drag on my mental energy!)
And of that interesting stuff, how much can't be had just by configuring the window tiling shortcuts in floating window managers to have the best of both worlds?
I've used dwm for years and recently switched to niri and really, really like it.
I'm with you, that tiling vertically is not useful. Niri aligns better with my innate mental model: just keep adding new panes to the right offscreen that I can scroll through, then I can swipe up and down for more lanes to do collections of different tasks.
I feel much more productive doing read/write tasks, like viewing tickets + code or PDFs + tickets and so-on
In usability studies, it’s not uncommon for users to report they are faster at doing things one way, but actually timing them they were faster doing those things a different way, which they thought was slower.
So I don’t know about your particular set up. But maybe a usability researcher will come up with a way of working that’s more productive or efficient than what you’re doing now.
If you've settled on Gnome then you don't have to change anything. Gnome will change it for you. Regardless of you. I really miss Gtk3/4 File->open dialogs having the ability to take pasted filepaths without erroring out.
But I will hand it to Gnome: they're the only wayland protocol desktop that supports accessibilty. That's huge. So even if I hate how it behaves right now while mostly sighted Gnome DE may be the only option for me in the future. And I guess I won't have to look at it when my retinas finish falling off.
> I too am in the "it's done, it works well, don't mess with it" camp.
As am I.
One thing I would like to change is that on Linux I can paste text by selecting it and clicking the middle mouse button to paste. This is better than using ^C ^V because I don't have to get off the mousing and onto the keyboard. But unfortunately on some webshites the MMB doesn't work.
Does it even work on Wayland? I always thought this was a feature of X. I really like it too but the inconsistent support seems to catch me more and more often these days.
Yeah, it's an X thing. But on X, the highlight/middle-click clipboard is separate from the ctrl-c/ctrl-v clipboard, so you can hold two things at once when moving text around. IIRC this was just a historical quirk rather than any intentional design, and I think in Wayland they duplicated the controls but also combined them into a single clipboard.
TBH IMO the Gnome is pretty much done software except that some of the desktop Idioms they can borrow from Apple (like swipe three fingers to switch the workspace) if they're not patented and they could be in settings if someone wants to opt into those otherwise yes.
Gnome is "done software" already and I like it very much.
From somebody who has been through this 3 times before: the moment gnome becomes “done software” is the moment they decide they need to throw out everything and start fresh.
I think it's a combination of it's done to a degree and comfort + habit.
Let's say there are areas to improve still, but introducing anything new requires investment of time to learn it, breaking habits and giving discomfort. Even if it eventually starts paying off, it can cause a push back naturally. So it's really hard to do it right, may be only very gradually? Surely not like Gnome tried to do it.
Some direct comparison with such idea could be let's say a suboptimal keyboard layout and more optimal ones that are barely used.
I personally stick with KDE exactly because "it just works well enough" and its flexibility at the same time.
Same... ish. 2D grid of desktops. Can add/remove them with a hotkey, but I normally have a 3x5 grid. Hyper+Arrows to navigate them. Hyper-s to stick/unstick a window, which also works for moving a window to a different desktop. Hyper-g to group/ungroup windows into a single shared frame with tabs. Hyper+Alt+Arrows to move a window, Hyper+Shift+Arrows to resize, Hyper+PgUp/PgDn to raise/lower, Hyper+Tab to change tabs in the current frame, etc. Lots of other keys, mostly Hyper+something.
Each desktop has a project or something, a related group of windows which I leave open until I'm done with it. Some desktops have remained more or less the same since the 1900s since they've been relevant and useful for that long.
Over time, my desktop hasn't gotten fancier... it has gotten simpler. Bells and whistles are inversely proportional to useful life span. Similarly, over time I've relied less and less on big, full-featured, specialized applications and databases... and more on plain text files, a text editor, short scripts, clean filesystem organization, and a whole lotta terminals.
I have a very similar setup on kde. I used to run the tilling window manager i3, but since using more manufacturing software (kicad, freecad, slicers, etc), I wanted more traditional window management. kde with custom shorcuts gives me enough tilling management. I also like asign a sticky topic to numbered desktops.
I predict when a xterm comes along that behaves like tmux, ie makes all its panes available to all login sessions and does the reverse - makes your ssh sessions available to your favourite xterm and other ssh sessions, you will change.
The only thing I'm missing with the workspaces or tiling WM is slide in Windows for cross cutting concerns...
Want a terminal on the exact screen/workspace without minimizing windows or switching workspaces?
Want to drag a file from a file manager into an open Window?
Want to lookup something in your contacts or calendar?
I thought of something like quake-mode [1]/ guake as global overlay triggered by a shortcut like super+t for terminal etc. Would be a nice Addition to cosmic with tiling i think. Did not find the time for a PR yet.
Spectrwm(rather obscure, but I like it) has a free window mode for things like this. where free windows are in all workspaces. Define some hotkeys for the special app invocation and perhaps add a protected region to keep normal windows out.
Multiple desktops(spaces) each with one window of browser (Zen) and one window of cmux and an Emacs frame. One desktop for one task.
CMux has good native AI Agent notifications. So it works well in that capacity too.
Sometimes I wish there is some automatic tiling, but there is nothing good in Mac for that (I tried a couple of 3rd party software, but didn't work out)
I also agree that the desktop is pretty much perfected at this point. I'm not saying UX researchers shouldn't try things - by all means, try new things. But don't expect that most of the ideas will land, because the paradigm we have really is quite good.
The Windows 95-7 paradigm was good. After that it went downhill. OSX was already downhill (probably just due to different conventions) and Linux can't stop digging deeper.
the Gnome that came with Debian Trixie is certainly worse than my stable Linux previous desktop, including eye-candy desktop effects coupled with remarkable fails on simple GUI features that have been stable for thirty years.
Well the "non comparability" to gold has to do with the fact you need a functioning network to spend (good luck verifying a key by hand with a calculator). You can spend gold even when you're transacting with the last person on earth.
However the value is as much as people agree to value it and for a typical person both have little utility. Maybe BTC has even more utility because it facilitates remote transfers of value very easily.
So as long as the network exists there is intristic value in BTC. I believe more than one can say about gold.
Still, a good portfolio will contain both gold (in small coins likely as a kind of "war hedge") and BTC as a kind of hyperinflation hedge.
You're kind of repeating my point about non-equivalent of gold. Yes I agree in case of civilisation collapse btc will not be useful and gold may very well be.
But on the hyperinflation I disagree. Where did we see purchasing g power of btc to drop when currencies experience high inflation?
We have a plethora of opposite examples in various countries, and one very good global example. During covid when money printing was running full steam BTC skyrocketed and kept most of it's value.
There was a panic induced crash initially, but that recovered very quickly. Mind that sadly btc is used for speculation mostly, not as a currency so it is very prone to various panic and euphoria events. Still if you want to use it as a currency (to transfer or store wealth) independently of the banking system it is not bad.
Having said that I'd never put more than 1% of my worth in btc mainly because it's a bit archaic as far as crypto goes.
My local coffee shop only takes long government currency.
My bank already allows me to keep my bank account money in stocks and sells units whenever I spend. (We have no capital gains tax here, so this is less insane than it sounds.)
Deflation is a good thing, it rewards delayed gratification.
Those evil Keynesians have convinced the world a little bit of inflation is good. It isn’t. Losing purchasing power on your money is a bug.
Nothing wrong with putting money under a mattress for 100y if the value of money is not evaporating.
For most of human history the money was stable. It’s the disasters of 20th century wars that eroded the value, and 21st century lack of monetary discipline that keeps driving it down now.
It's nice when I do it. Not so nice when everyone else does it. If sitting on the money has better returns than running a supermarket, why run a supermarket? Any investment has to beat deflation. Why hire people? In fact maybe I should fire everyone to hold on to more capital and spend as little as possible...
> If sitting on the money has better returns than running a supermarket, why run a supermarket?
First of all, because not everyone starts with inherited wealth. Also because ideally running a supermarket should give you more money even in a deflationary world. Worst thing is that you gain less money on day N+100 vs day N, but it does not mean you lose money or stop gaining it.
> Also because ideally running a supermarket should give you more money even in a deflationary world.
Running a super market involves owning physical goods for some period of time. With deflation, the price you can sell those goods for drops while you are holding them. In fact most economic activity involves paying for inputs (labor, materials, etc) and then later getting paid for your outputs. Deflation directly impacts profitability and can cause losses.
Since deflation causes demand to drop as economic actors wisely choose to start hoarding currency and buying less, this causes a feedback loop where deflation can spiral.
Similarly, inflation causes demand to increase since holding currency is unwise and it is better to spend or invest that currency than hold onto it.
These two patterns mean that the neutral state (no inflation or deflation) is unstable as any deviation above or below starts a feedback loop until things fall apart. This is the boom and bust economic cycle that modern monetary management is supposed to ameliorate.
Given that you want economic growth, the best solution is to try to stabilize around a small fixed amount of inflation. Arguing for the end of inflation is arguing for the end of economic growth.
My family needs food though. Perhaps a supermarket is a poor example. I think people might purchase fewer luxury goods in a deflationary world which TBH I'm not sure is a bad thing.
To put it another way, the model you're presenting reads well in an economics textbook and I'm sure is exactly how we justify our MMT social policy but it doesn't fully account for actual human behavior. I'll buy necessities (house/food/water/electric/communication) even in a deflationary economy.
When was the last time you went to the grocery store and thought "I better buy this milk today because my money will have less spending power tomorrow"?
An economy driven only by spending on bare essentials would be worse than Great Depression level malaise and stagnation. Where are you imagining growth would come from?
Time is money. Emotion. You think someone's not going to buy a sports car because it will be cheaper in 10 years? That's 10 years without a sports car.
They might, however, ask whether they need a car with 30% profit margins or one with closer to 10%.
I don't think the grand social goal is to maximize economic growth. It's a stable healthy society.
Right now it really just feels like people spend on whatever the hell they want to without any thought whatsoever. Consume. Yes good, consuuuume. I'd trade slower economic growth for more responsible spending...
> First of all, because not everyone starts with inherited wealth.
So how are you going to build the supermarket?
> Also because ideally running a supermarket should give you more money even in a deflationary world.
If it needs to give you more money than just saving the investment (which it should, you need to be rewarded for the risk or you would just save the money), obviously the profit margin has to be higher than it currently is, which would increase prices.
> First of all, because not everyone starts with inherited wealth.
So then you need an investment; you're going to have to return a multiple of the deflation rate since the risk of your supermarket shutting down is probably higher than the currency changing course.
But a deflationary system rewards inherited wealth. It's a pyramid scheme where the person at the top splits their big piles up into smaller piles, selling them to newer people, who then sell their smaller piles to newer people...
So you'd be working for 0.000000000000000001 coins per day at the amazon warehouse, while Bezos has 500000 coins because he was born with them. There would never be a way for you to get 500000 coins, because there are only 20m coins in existence.
The easiest thing for people to do in that situation is just create another Bitcoin. The person that has the 500K original Bitcoin is free to trade with themselves. This is why Bitcoin is much more of a social network than an algorithm.
The point of money is not to reward delayed gratification. The point of money is to efficiently tabulate human preferences, and deflation directly counteracts this by introducing potentially unbounded latency at every step. That's why it destroys economies, as it has throughout history.
The problems with using physical gold as currency are very well known. When population would increase, or when someone would hoard it, it would cause deflation. Likewise, when a new deposit of gold was found, it would cause inflation.
This is, in part, why there were expeditions to find gold.
This is a almost entirely oversimplified take on the Great Depression to the point of meaninglessness
For a decade before Black Thursday,there had been many things that were signs that the economy was having trouble even if the "Roaring Twenties" made it seem like everything was fine.
IMO the largest issue was that American farm sector was teetering on the edge because of the dramatic drop in crop prices. This deflation screwed over farmers who mechanized with lots of debt, which because of said deflation, became impossible to pay off.
The fed also implemented rate hikes to curb speculation right before 1929 which froze up credit contributing to deflation
The problem of the Great Depression was NOT the stock market crash, it arguably wasn't even the real start, just the most "spectacular" one. The problem was that with the entire economy deflating, it caused a massive downward spiral that the Fed did not really have the tools to fix, because of Gold Standard and lack of legal authorization.
This was why the Govt went to extreme lengths to try and figure out how to raise prices, which is why you get programs to pay farmers to NOT grow food, and mass killings of pigs and cows and other farm animals, even as the farmers who raised those lifestock went hungry.
So no, speculation was not the problem, it just sparked the key issue of the fact that the economy was deflation uncontrolled, but was just hidden.
This is a complex topic and I think you have done a good job of summarizing the main issues. To add a little context:
>...that the Fed did not really have the tools to fix, because of Gold Standard and lack of legal authorization.
This was just bad policy by the Fed. The Fed had the legal authority to be the lender of last resort and could have prevented the bank failures. Many explanations have been given over the years as to why the Fed didn't provide liquidity. Because the Fed failed to supply emergency liquidity, the U.S. money supply plummeted by nearly 30% over the next couple of years, which essentially turned what likely would have been a recession into the Great Depression.
This is not to say the gold standard was not a problem. During the 1930's, leaving the gold standard was one of the few good moves done to help the economy by the federal government.
The Japanese economy is not, in any sense, destroyed. It doesn't get the infinite exponential growth unhinged economists want, but life on the ground is stable, wealth inequality is low, cost of living is low, average quality of life is very high. It is the perfect counterexample to the doctrine of chasing line go up.
The idea that you can put away an amount of money under your bed that buys 1,000 loaves of bread or one GPU, leave it there for decades, and then have it buy exactly the same number of loaves of bread or GPUs is a fantasy. You can hold onto the shiny rock but you cannot stop the world rotating around you and changing all its relative prices.
> For most of human history the money was stable
Achieved by a combination of restrictions on trade, price stability laws, occasional crippling shortages, and quietly shaving bits off old coins. A much poorer world.
This would be more convincing if it wasn't from a site trying to sell me gold. Do people really believe that the mechanization of clothing production in the industrial era has made no difference to "real" prices?
I think deflation-based economy could produce some interesting capital-allocation environemnt. Investment offering a 2% real return becomes unattractive if cash itself earns 2% real purchasing-power yearly. You could argue this raises the hurdle rate for investment and eliminates low-quality projects. And the counterargument is exactly the same: it raises the hurdle rate for investment and therefore some potentially good projects would never receive funding. And thats probably where the intellectually interesting argument really lives, rather than in inflation good deflation bad
Wildly inaccurate, thanks to forgery and coin shaving - sometimes even governments officially reduced the silver or gold content to make more money out of their coin reserves. Even when proto-banks began issuing letters of credit, the quasi-fiat letters were subject to loss of confidence.
However, the availability and quasi-fungibility of other silver/gold currencies meant that if you didn't trust Edward's penny, you could use a Dutch penning instead. That provided an alternate path to dampen inflation, as long as the dominant currency was coinage.
But it was equally hard to buy a pig or a new suit with silver pennies by the 20th century. Bank notes, even when theoretically backed by exchange for their value in precious metals (the Gold Standard), were even easier to forge, and suffered from "loss of faith" inflation (runs on banks meaning they couldn't practically be exchanged for 14 pounds of silver pennies).
The right thing would be to have 0 change in the value of money as long as the right amount of money exists.
The right maount of money is the amount of money we as normal humans need to work with (buying and selling stuff).
Inflation and deflation are results of too much money or too little money in comparision to the production capability of a society.
If i save today for my retirement and money gets less valuable when i'm retired, i have to give more 'saved' capacity back to get the real capacity (people taking care of me) and if i have more value, the others have to do more for me.
Controlling this is 'work' from experts and is not solved by bitcoin btw.
The normal term would be “velocity” of money, btw. Its a key consideration in addition to total supply whenever you need to evaluate inflation or manias.
Why does everyone assume that we're the ones keeping money under the mattress, not the ones who would have been paid by money otherwise not spent? All transactions have two sides, no?
> Deflation is a good thing, it rewards delayed gratification.
"Delayed gratification" is also provided by investments producing returns. An economy with lots of investors will outperform one where people stuff their cash into their mattress, and deflation makes it very hard for potential investments to beat that strategy.
> For most of human history the money was stable.
[citation needed]
The Spanish empire was driven to collapse by hyperinflation. Even in the US, there were financial collapses in the 19th and 18th century. Bank runs have been a thing for as long as banks have: https://en.wikipedia.org/wiki/Bank_run
Your premise is based on faulty assumptions. The existence of credit itself is what causes monetary instability, and without credit the world would look very different.
Money is destroyed when a loan is paid back. Private credit does not expand the monetary supply permanently. Only the state can increase the money supply.
Your understanding of monetary theory is somewhere between 110 and 5,000 years off. Furness had a pretty cogent explanation of a monetary system without central authority or functional currency about 100 years ago with the Yap. They even managed to have bouts of inflation without the concept of a bank or state.
credit does provide a kind of flexibility that is sometimes needed, though. However, predatory lending, and the endless stacking of recursive loans, and government money printers are a massive stability issue that we're running into globally, and have (as you say) run into multiple times, historically.
My thought on this would be a dynamicaly stable currency. estimate debt and transaction activity, and the more debt and more liquid activity there is, the more deflationary currency should be. the less debt there is, and the less of a percentage of the money is actually in-use, the more inflationary the currency should be. this, though, is fairly off-the-cuff.
If you have a brilliant technical solution that requires throwing out all conventional economics, you don't have a brilliant technical solution. Bitcoin is rotten to its core and every excuse you make for it proves the point.
>For most of human history the money was stable.
Absolutely ridiculous. People have been counterfeiting and debasing money for as long as there has been money.
That only makes sense if money is a durable good destroyed by use. But money is improved by use and lost when put under a mattress. In economics terms, MV=PQ, and your proposal sets V low, which harms Q (goods available for sale)
No, no. The issuer of your money is really, really happy when you don't use the money. Because that means they can issue more money, without causing inflation to spike.
..and, you think that covers both individual and collective good?
..balance in all things. Neither being completely stingy, individually, nor being excessively spendy will benefit us, individually or collectively. ..but there are times for either.
I wonder if there's a way to quantify that and put a variable on the conditions, and have an inflationary/deflationary currencynthat is dynamically stable depending on conditions.
..i mean, individually, most people will eventually spend, if they have much saved and it benefits them to do so. but occasionally, we do need a kick in the pants. whenever the economynis in gridlock, that's the time for inflation. ..but when people are spending excessively, it's a time for deflation, which discourages taking on debt, and pushes the economy towards real wealth. rewarding long-term thinkers is valuable, and has a very broad effect on society.
This can be said about many conflicts between the individual and society, though. In many ways we are prevented from just taking what we want and “keep us in shape” because if everyone did the same it would be a problem.
Hahahaha, oh my. You think the world was some idyll halcyon pre bretton woods? My man Enmentrna is going to come back and declare a jubilee for your great revelation. When has any historic monetary system been “stable” for an appreciable amount of time. Debasement is a very literal ancient word and problem.
Even your straw man 20th century cut off is hilarious where you just kind of forget about 1873? Or maybe that decade is your shining example of the benefits of deflation. So much joy and global prosperity the peasants just forgot how to eat. Its cool, Bismarks destruction of the bimetallic system really helped usher in that age of stabikity from the international gold standards.
The idea that deflation is built into Bitcoin is exactly equivalent to saying "the real value of bitcoin will always increase" which is an absurd premise.
Bitcoin is deflationary only in a hybrid Keynsian - Austrian worldview. In the Keynsian worldview it cannot by definition be deflationary because that would mean that the value is always increasing which is just kind of a mad thing to believe. In the Austrian worldview it is not deflationary because the amount of Bitcoin is always increasing by design. Only if you accept the Austrian framing of "deflation is when you decrease the money supply" together with the Keynsian framing of "money supply is measured in real terms not nominal" do you arrive at the idea that it could be deflationary, and there are exactly zero economists who believe both of these things.
I suspect it was a deliberate strategy to create scarcity, allowing the original creators to massively cash out. If you make an inflationary distributed currency, it may work better but it's a bit harder to get rich on it.
During much of the industrial revolution, gold also rose in real price. But people still did business in gold standard countries.
(Hint: the gold might be under a mattress or in a vault, but you can still an almost arbitrary amount of gold denominated debts and loans and deposits.)
There is zero evidence that deflation has any effect on spending.
At the micro level, the change in price is too small for every day purchases. Would you starve yourself for one day because the pizza will be one cent cheaper tomorrow?
At the macro level, every interest rate will be adjusted based on the base inflation/deflation rate, so the net effect is zero. Banks will offer a higher profit rate for their savings account to entice people to deposit their money in the bank instead of their mattress.
It's not silly, it harnesses some of the mechanics behind ponzi schemes to encourage viral spread. Early entrants are incentivized to evangelize it to newer ones
An even better idea would have been to integrate it as an extension for C2PA, which already has an ecosystem developed around it. The capture portion of the ecosystem has lots of problems, and this is one potential improvement.
Has anyone else here been trying to rent some cloud gpus for reasonable money? I have. I have signed up to a dozen "providers" that advertise everything from tesla cards to b200 only to find stuff is either out of capacity or double/triple the price.
reply