megameter

You need a temporary space of 312GiB to "plot" which is RAM intensive(~4GB min, more if you parallelize) and disk I/O bottlenecked(serious Chia farmers put their temp on an SSD), but finished plots are 101.4GiB each and are just files accessed occasionally, so you can fit a dozen or so on a 2TB. If the drive goes, so do the plots, but the cost is just the time to replot so you probably don't need a backup strategy.

Plotting tends to be a bottleneck - hours to finish one in my experience, though that is with using the storage drive for the temp space. But even with 26 plots(2.5 TiB) I have already earned 4 Chia in a few weeks.

It's roughly similar to Brave's userbase, though Brave is opt-in. Figures from 2020 put both in the low tens of millions.

The analogy I use with NFT art is not with physical items but with banking. The token is just another form of credit issued by the artist. The relationship to the specific work is knowingly tenuous and entails intermediaries - and in this respect, it's still a tech demo - but the financial asset itself has meaning relative to the credit of the artist, given that the wallet that minted will itself have a persistent identity.

Recognition for previous work creates expectations for new work, even if the work itself is lost. Art has always operated that way.

But there are details that need sorting out. A proprietary URL link - or an IPFS link only hosted by one node - is too flimsy for long-term collectables. Use value is sorely needed to really flesh out the function of the tokens. Again, as it is, we have tech demos with some eyewatering valuations.

The kinds of tools like those in the article suffer from being outside of the "main loop of coding".

What is the main loop? Well, as it's defined since the rise of interactive computing, it's typing in the editor, compiling, then testing and debugging.

Thus we optimize for:

1. Typing code fast(or typing less code)

2. Compiling code fast(by doing less compilation)

3. Minimizing testing and debugging time(by narrowing our definitions of testing to that which is easy to automate).

The main loop of coding is not software development. It does not study the domain problem, define new benchmarks for success, untangle communication issues, or refine principles of development. It is a path to coder flow state, feature factoryism and shipping the org chart. These things sort of resemble development - a happy coder delivering features within a definite architecture does serve someone at some point - but also manifest dysfunction. Development failing is just shrugged off as bad programming or bad management.

Tools like a Whyline or Reflexion models are not in the main loop. They are an invitation to look carefully at whatever horror has been wrought and address it at a deep level. But that really needs a researching mindset, not a shipping one.

In practice, the available tools very slowly move in the direction of the suggestions of research. There are improvements, but they need projects that put together the pieces and make it a goal from the language design(the main interface) upwards.

I think there are two angles on Bitcoin that make some sense:

1. The true believers like labor theory of property a lot, and see having a mining rig as an extension of that(falliciously or no - there are plenty of critiques of labor theory of property itself). A bitcoin represents the "same amount" of effort relative to its value at a given moment in time.

2. As a lower margin use of energy premised on maintaining consensus, it has qualities that resemble perfect competition(in the economist's sense of that phrase) but actually go BEYOND perfect competition. While you can own a large part of the distribution or mining power, if you try to monopolize Bitcoin as a rentier you just end up forking it, as has been demonstrated by various events over the years. If people don't like your particular imposition of scarcity, they pull the rug in short order.

The second part is rather more important than the first, IMHO. Mining's meting out of energy usage as credit and the resulting price deflation is a way to get people on board with consensus and start engaging in speculative action, but there's a sense among the people interested in other cryptocurrency that that's just a starting point, and the development of distributed consensus tech itself is what makes this a "railroad" and not a "tulip".

"No rentiers" is really a pretty earthshaking concept, to the point where most of the people in the space can't see that light and look at the shadow puppets instead. But it does not seem to be refuted by anything I've encountered. While onramps and offramps to crypto are mostly controllable, this is still a huge upset to our models of what markets are or could be.

In the Melzer Philosophy Between the Lines book mentioned in the article, the author does a pretty throrough job of making the case that pre-modern philosophical writing defaulted to esoteric style, while in the modern era we default to exoteric.

He does this by pointing to the allusions and references they use and the clear contradictions within them - often inversions of meaning from the original sources that are alluded to. Socrates makes these errors, as does Machiavelli. If you assume the errors are mistakes, the work's apparent conclusions become more shaky, but if they are intentional puzzles, the work gains additional meaning. We often use puzzles as teaching devices and writing isn't really different in that respect.

It's not really different from reading for literary subtext, it just applies that toolbox more widely. I believe it's more useful for writing than for reading because esotericism makes it easy to encode something legible only to intended recipients. I would not expect to parse new wisdom from the utterings of a CEO, as in article's example, until I become familiar with corporate culture in general.

The way my mom did it(and how I cooked it last night) does not even use a real roux: after dumping the pasta water, stir in chopped butter and then shredded cheese directly. The pasta's hot so everything melts in moments. This makes a non-creamy mac, but it's equally edible in my estimate.

I think this is more a case of the laptop being the 21st century guitar, than any instrument being "obsoleted". The trends in popular music have followed a min-maxing of physical logistics, loudness of sound, and timbral flexibility relative to the era and setting:

The railroad preceded the rise of the large, Sousa-style marching band, which also made use of the loudest instruments of the era(brass). After electric amplication set in you got an era of radio crooners and smaller-piece popular groups, which ultimately gravitated towards the now familar guitar-keys-drums arrangements by midcentury as the nascent rock band could fit that gear on a bus and get a wide timbral variety with use of amplification and effects without relearning the instruments. The guitar in particular has a good ergonomic profile while also being amenable to amplification and effects, hence the stereotype of the "travelling guitarist" coincides with its central role in large arena rock productions where a three or four-piece plays to an audience of tens or hundreds of thousands.

The laptop fits in with the trend by being even smaller, more portable, more flexible. It fills in nearly every need for back-up, so a solo act can sound as big as desired. And it can be used in its own right as a performance instrument, though computer keyboards aren't tremendously expressive interfaces in the way guitar strings are. (For that, there is always someone selling some kind of new controller interface idea. Though they have a habit of not sticking...)

Yep. In the terms I use, it's another instance of "the medium is the message", and if the medium is an edifice of mechanisms to configure the medium - well, you have trouble communicating a message. So in an actual workflow, most of the DAW will go unused at any moment, for the same kinds of reasons that most of Excel is unused most of the time. The process will traverse and pingpong various paths that loop outside of the software. The tools don't make the rules, but they suggest a path of least resistance.

It's an old problem within software itself and not really addressed until the individual puts their foot down and says, "This. This is the medium, these are the restrictions, this is the way I structure everything." At that point it's possible to define perfect software, but we're too fickle to really get there.

My "desk" is actually four stations that I switch between.

1. Gas arm mounted monitor over the floor for lying supinated.

2. Floor desk with a backrest.

3. Rocking chair with a lap desk.

4. Prone using a laptop.

I also switch between various wireless peripherals as I switch between stations.

Of the four I'd say the floor desk is actually the best long term one because it's easy to shift position when working wirelessly, but the desk and mounting solution I have isn't quite tall enough to hold my gaze forward. Still, it's fine for a day.

I don't see this going in the direction of a heavily enforced response, although some places may talk big. While a collective ban is rationally sensible, nations are made of individuals and every person will do a calculation of personal benefit relative to national interest. Crypto is valuable as a refuge, so a recurring story is currently playing out in destabilized nations all over the world where crypto is used to escape the inflated local currency: the government gets wishy-washy about what it's going to do with the stuff, with different officials taking wildly different stances according to their own self-interest. They talk of a ban, and then of national cryptocurrency, and then of nationalized mining. Everyone is looking for safe harbor but pulls in a different direction.

Focusing on controlling onramps and offramps to cryptocurrency, like controlling goods import/export, is a relatively straightforward compromise to this internal dilemma and allows the government to be a player in the space without much direct oversight, even if some folks are slipping through the cracks. A sloppy, haphazard enforcement is likely to prevail.

And there's another way to reassert state control within that framework. If the bad cryptocriminals come for your stuff, the government can bail you out - if you play by their rules. The loss compensation mechanism is a simple plan to mitigate dangers, since it's easy for governments to create and redistribute credit internally and doing so can build consent.

Submitting to this framework does mean that the government is truly in competition with decentralisation to provide a better, more trustworthy service to handle central credit, and has to massively step up security efforts in the process. Everyone started "asleep at the wheel" on this, with individual tech firms all building out their own insecure house-of-cards fiefdoms and relying on the intellectual property law framework to keep them up. But everyone knows it's flawed - of course you can copy, what's important is if you can get credit, and that part is also changing with the crypto sector - assignment of credit is the whole thing of NFTs in a nutshell. The nation that can grasp this the quickest and turn it into a coherent part of their economic framework will cruise ahead.

The art for Indy 3 in EGA definitely shows signs of dither being desired but not used, which I always found made it feel sparse and unfinished, even in VGA. Mind, I was coming to this from the standpoint of really getting into the Lucasarts adventures after LOOM. It's likely that on this game they were stretched for space regardless, since it spans a lot of scenes.

Mix of price, compatibility, dev support. Atari was behaving like a consumer electronics company and left the developers starved for information on the computer, while Commodore shipped every unit with a very comprehensive manual. The C64 was a clean break from predecessors in most respects and the VIC II was more straightfoward to program for typical raster graphics(even if it missed some of the neat features of Antic and CTIA/GTIA), while the Atari had a legacy install base with lower memory capacity. So, by the mid 80's the Commodore had become the defacto cheap game machine, in absence of new console competition. It was never a great applications computer, in fact(I/O limits derived from the low cost architecture meant it did less than an Apple II for many development and business tasks) but it hit a certain sweet spot of being a great first computer.

Atari after the Warner acquisition was a hugely dysfunctional company - there are a lot of proposed alternate histories where different projects got prioritized and execution was better, that would have given them a stronger position in the computer market. As it is, they still put out some pretty cool hardware.

EGA definitely has a charm to it. The main limitation is that the viable color ramps are pretty awkward and only slightly improved with dithering:

* Grey values * Grey-blue * Grey-yellow * Grey-green * Yellow-red * Yellow-brown * Yellow-grey * Blue-brown * Blue-green * Blue-purple

With this setup, you often have to live with a harsh transition between extreme color saturation and no saturation. You can see the result in the way Ferrari's backgrounds are formed - usually only one or two color ramps plus dither, ignoring a lot of the palette, heavy shadows and rim lighting to indicate the color without a hue shift.

It's actually really great if you are going for a warm-cool contrast because you can run lengthy ramps of reds and blues.

The medium that is changing with NFT is one of credit and incentives, not aesthetics. NFT art isn't differently made art, it's differently positioned art. It's like jumping from movies to television. And it's similar in a nominal sense with respect to current platforms, but there's a definite sense of "we have not really done everything this medium can offer yet". The phase where it's a gimmick will pass.

Bitcoin mining has also become the lifeline for the likes of Iran and Venezuela, where the power is priced cheaper in local money.

Iran has been moving towards an official legal framework for Bitcoin.

The electricity argument loses weight when you look at the world allocation of GPUs. Although mining has cut into supply, they're also being used for gaming, workstation, and cloud compute tasks. One way or another all of that chip supply would be sold, and then it is a case of benchmarking the social value, load and efficiency of the different uses to some utilitarian optimum.

Much easier to cut off the framing early and declare "all crypto is bad and has no value", which is what the contra side is leaning towards. It's easily embraced by authoritarian-left ideologues, since their inclination is to reject market solutions anyway.

My suspicion is that we'll all be surprised in eighteen months. The "boom" will be gone, but an unalterable trend will have taken hold regardless, and the platforms will have largely moved on to proof of stake, closing the environmental argument.

The thing of cryptocurrency is that at the end of the day, all you have is a file of a few hundred gigabytes that somehow represents hundreds of billions of dollars of market value. That value means that an absolutely huge number of eyeballs are scrutinizing it, looking for a way to protect their investment. They can't "turn off" the system and make sweeping changes. The hypotheses of anything-goes "assassination markets" have not panned out - crypto has become a sector with a civic outlook.

It's also transactional if they put the work behind a centralized DRM system, yet somehow we are not pushing back against that, which to me suggests a lack of imagination.

All depends on the specific incentive models. If an NFT holder aims to increase speculative value, they turn into vocal champions of the work. That drives a great deal of the cryptocurrency space to date - there's no particular reason why it will fail for art.

And this was previously done in a very exclusive way, in elite venues, but now it's possible to throw open the doors and make it like eBay - without even needing the seller ratings.

It's how this is, at its root, about the concept of credit, rather than the concept of property.

If I grab something out of your hands and you say, "that's mine!" That would be property.

But if I put up a painting and you ask, "who made that?" That would be credit.

There is no scarcity of blockchain space - you can download Bitcoin right now and use less space than the newest "Call of Duty" collection. The property is very easy to acquire.

But then if you ask "who made Call of Duty", you are forced to point to the system of IP law, which may or may not credit the individual creators properly.

So, what NFTs are trading is a particular form of credit. This tears up some economic assumptions around what credit is "worth". A popular blockchain is one that allocates credit accurately and fairly.

Some notes:

1. Drawing as a skill does not live in the muscle memory(although it is possible to encode symbolic elements in muscle memory) but in the conversion of eyesight to concrete measurements, followed by a reproduction of those measurements on the page. If you really "get" it, the medium will not matter.

2. Drawing on the Right Side of the Brain does get you the starting point of this - by giving you simple techniques to decode what your eye sees - but it does have a gap when you aim for cartoon construction and drawing from imagination - at that point, maintaining proportion as you transform the object in your head becomes the challenge.

3. A simple way of getting past the gap is to focus on graphic design and typography. Simple ratios of 1:2, 1:3, 1:5, over and over, subdividing a space or being translated to an extrusion. If your eye can see simple ratios, construction of the form proceeds by referencing the ratios and mapping them to primitive shapes with the intended relationships. Sometimes you need a lot of construction lines, other times you can see it immediately. The way in which you structure things directly affects the results.

4. Perspective is sweaty work - while you can estimate how a form looks with some rotation, it gets very hard to compute what's going to happen as you make more extreme changes with foreshortening. Professional artists are always looking for ways to cheat around attempting this computation directly, and many fall into a trope of stock scenes and poses adjusted or interpolated into each other, since those are the ones they know well. Making the scene really lively and distinctive always comes back to good use of reference.

Assumed is the model of production: bespoke assets for each game.

This is untenable going forward if the industry continues to have a rising cost curve. Raph Koster's discussion of game cost is a good overview of what's happening and the challenges that result from it.

The step towards an NFT market begins with the creation of viable transfer mediums - assets not optimized towards particular games but with a reasonable degree of retargetability. The market for modding and independent asset creation then writes to that format and some games start to subscribe to it, enacting a "Multiverse". At that point opting in to an NFT market and allowing trade across games becomes a viable way to offer new experiences at lower cost, and the tokens themselves are meaningful for the same reasons that existing microtransaction tokens are - use value in-game, ability to "flex", but probably not an in-game advantage(unless we go full "Barcode Battler" and manufacture a randomized meaning from the token).

This isn't hypothetical - ESPA/Digitalax is an example of an attempt to make this market. It's pre-launch with a lot of unanswered elements on the technology side of things - I doubt the format that they currently propose - but the concept otherwise makes sense, and it's possible to pivot, aim lower, and not go for universal formats right off the bat.

It should be noted that should you have the right kinds of connections, bitcoin mining itself lets you launder. The miner is given the dirty money as capital, energy subsidies, etc, then creates clean bitcoin, which appears in some anonymous wallet and is then sold.

So it's not just that the black market has incentive to use it, there's incentive to make it and even to prefer it over alternatives, because the operation's excess is a good front.

The actual conclusion I think is most relevant is the nilhilistic one: "None of these tokens really matter!" The actual place of decentralized ledgers is as one of several tools for designing autonomous, non-coercive communities, and when designed as such most of the overhead of proving disappears(e.g. DAG coins). Everything else is an attempt to extract rent into the coercive economy, which in due course will be trivially subverted by forking the chain or starting a new one.

I think the key difference is exactly what you're getting at - there isn't a physical scarcity to be had here, so a digital token primarily expresses speculative value by default. Speculation always comes and goes with the hype of a bubble.

Lasting value beyond speculation appears when there is use value. But how do you apply a use value to a token? Well, you can't do it by locking down the nominally encoded asset - that's just a sticker pasted on it. These things do not gatekeep well. The use value comes from someone declaring that the bearer of the token is now entitled to something - a song, a poem, a noble rank. Anyone can do that for any token at any time. That is a thing that a unique, uncensored digital asset could be very relevant towards.

So the interesting part has not really come yet.

It's a coding strategy intended to optimize around console refresh targets first and then "do whatever" for PC build.

Generally, a AAA console game will target 30hz or 60hz. Therefore the timing loop is built to serve updates at a steady pace of 30 or 60, with limiting if it goes faster. Many game engines also interpolate animations separately from the rest of the gameplay, allowing them to float at different refresh rates. Many game engines further will also decouple AI routine tick rates from physics to spread out the load. Many game engines now also interleave updates and kick off the next frame before the first is complete, using clever buffering and multithreaded job code. All told, timing in games is one of those hazard zones where the dependencies are both numerous and invisible.

When you bring this piece of intricate engineering over to PC and crank up the numbers, you hit edge cases. Things break. It's usually possible to rework the design to get better framerate independence, but doing do would be invasive - you'd be changing assumptions that are now baked into the content. It isn't just fixing one routine.

Bitcoin(and by extension the rest of the cryptocurrency space) is an enabler for speculation - in the specific case of BTC, on the price action of the asset as more is mined, and on the cost of mining itself. Speculation's action - which is always crude when looking up close - has the effect of building a market where there once was nothing, and as BTC has grown crowded speculative activity has proceeded towards projects promising(if not necessarily delivering) additional features and social value.

One of the major limitations of a pure debits-and-credits system like Taler is that it has limited speculative interest. It can't be gamed for returns on investment, so it can't initiate the boom-and-bust cycle.

The important thing I see in the NFT space is the potential for "speculation all the way down" to occur. You can make tokens for the research paper you just wrote, own one and sell the others. The owners who bought the token have a strong incentive to increase the value of their investment so they create derivative work - new research, inventions, etc. They, too, issue NFTs. On and on this process goes until you reach the finished good or service, at which point the speculative value of everything that went into it is realized. Wouldn't you like to own the ideas of an Einstein? If the original author was a crackpot, of course, the value chain breaks down somewhere in the middle and the token is forgotten. But on the whole, this would be a way of measuring credit that wasn't previously feasible.

In contrast our existing system is an "exploitation" economy instead of a "speculation" economy; the momentary value has high correspondence to enforcement of your property rights through the state and industrial apparatus(laws, marketing, distribution, etc.), and if you cannot afford enforcement you may well never see a dime, while the process of enforcement itself constricts efficient usage. Not everything makes sense as an NFT, but many things could.

I would suggest approaching this in two structured phases - "breadth" and "depth".

In the breadth phase, you instruct the kid to study the topic of interest at a distance, collect information, attend events casually and give some reports explaining what they like. Make the thing of "pursuing an interest" just a little bit academic and intentional on their part. With a lot of topics they'll have their fill and loosen their grip pretty quickly, and when that happens, allow them to go on to the next thing.

If they can't shut up about it, that's when you go towards the depth phase and push them towards a more intensive effort, to take the class, read the book, join the club. Set modest goals that still take a committment, and indicate that it's very likely that some kids will be ahead or pick up the material faster. They should still report how things go and get your feedback so that you can spot issues, or teach them how to seek out good feedback where you lack competence. But there's a definite thing here of getting them to see the struggle itself as something rewarding, not the outcome like "being the best" or "making a career". Because if they get a feel for that, they will reach adulthood with some sense of balance and intention to what they pursue and why and a sense of their strengths and weaknesses.

As an informally trained writer of compilers, I believe parsing could be reduced to just these concepts:

1. Greedy matching

2. Minimal matching

3. Tokenization

Regular expressions and implementation of a subset of their operators are a good way to introduce and discuss each, and recursive descent is the elaboration on that, the graduation to a customized mechanism. The last is to generalize it all to a form of constraint logic and introduce other forms of backtracking(e.g. pathfinding algorithms). Discussion of types follows from discussion of constraints and explains "smartness" in compilers, so it might be the last thing if I were teaching the course.

What I really think is at issue with our vast number of grammars is just the reliance on text as the interface. If we discuss parsing as a thing applicable to arbitrary data streams we can start asking "well, what if we don't serialize it to characters, but to some other token format?" That is way more interesting today, since we seem to have dispensed with the premise of natural language being the way to program computers, that really got the whole parsing thing started.

NFTs are not easy to understand, but sit within the intersection of these two questions:

1. What is the value of art?

2. What is the value of money?

When we look for analogues, they're of the industrial era, things about intellectual property, branding, merchandise etc. This is a way of directly speculating on an idea. Is it a good idea? If so, then other people will want it, right? Then, the token must have value. There are no gatekeepers or enforcers to say otherwise.

We have barely begun to grasp what that means.