Scaling is Layering

in #tauchain6 years ago (edited)

Size matters. Some people object that it does not matter, but has meaning. But meaning always matters, so it is the same. 

The bigger problems one solves, the bigger the gains. Big problems require big solutions. We live in a big universe and our very survival is to deal with bigger and bigger problems, which require bigger and bigger solutions to cope. 

But nevertheless to build big is hard so we naturally prefer to create small things which can grow. Small from point of view both of understandable and affordable to build. So best fit are small solutions, cheap and easy to make which scale out or unfold or unleash into big means to address big problems. Scaling is everything. 

Scaling.[1] Scalable! Scalability !! 

The root-word 'scale' possesses marvelous riches of meaning in English language [2] with lots of poetics inside.: 

[3] snake skin epidermals - wisdom, memory, protection, rejuvenation, regeneration, eternity... 

  hen to pan (ἓν τὸ πᾶν), "the all is one" 

[4] warrior armour - security, defense, power, strength. 

[5] weighting scales - device to measure mass, unit, measure, account. 

all very Blockchainy wording without any shadow of doubt. 

The scalability issues could be grokked [6] with the following anecdote: 

Bunch of workers on a construction site and a huge log. The onsite manager commands a few of them to lift and move it. They try and object ''Too heavy!''. The manager adds more and more workers, until they shout back again: ''Too short!''. 

A few real examples, the first two - bad and the last three excellent: 

[a] I won't name this 'crypto' just will say it is named after a mythical element of the universe, according to the prescientific gnostic [7] imaginations. It's core 'value proposition is to shovel meaningful computation into a thread of computation which very value proposition is to be as random, meaningless and unidirectional (hard to do, easy to prove) as possibly possible - the blockchain. The theoretically most expensive form of computation. Visualize: cars and airplanes made of gold and diamonds burning most expensive perfumes. Or mass production of electricity by raising trillions of cats and hiring trillions of people to pet them with grid of pure gold wires to discharge and collect the electrostatics. If they have chosen the original Satoshi blockchain [8] for their 'experiments' - where the futility of such attempt would become instantly clear and would die out outright due to impending unbearable cost - will of course be more fair way to do, and would've spared dozens of billions of dollars to the Mankind, but logically they preferred a 'controlled' blockchain of their own. In a sense that the guys with vested interest into it have the power to hand-drive, stop, restart and vivisect it. The only use of this 'blockchain supercomputer' is ... tokenomics by Layering. Why it was at all necessary for a blockchain advertised as so good as to do all the general computation, to be made so hairy and bushy with layered tokens??

[b] Another trio of chaps, won't mention names again, were really at awe with Satoshi's creation, so much that they not just liked, but wanted it and decided to have it. For themselves. All of it. And rebelled and forked out and provided 'scaling' errrmm ... uhhh... solution. By increasing the blocksize. Something which Satoshi meditated on, extensively discussed with his disciples and not occasionally decided to put breaks on. [9] Very recently the crypto news headlines said that the blocksize increase solution providers are eyeing ... Layering. Which they furiously were advocating that blocksize increase makes unnecessary. Cause it is the solution, isn't it? Or maybe it just was. And is not anymore? Well, I'd say that all the aka 'alts' [10] - to provide a rejuvenated clone of Bitcoin tweeked here and there to provide momentary ease of difficulty and transaction fees - suffer from one and a same problem - traveling back in time does not tell you the future. 

[c] Lets jump half a century back in time. It is 1960es. The very making of internet. Computers are already here and scaled up in numbers so their networking to become a problem/juice worth the solution/squeeze. The birth of TCP/IP [11] and the report of the very makers of it. Of the solution for the network scaling. Enjoy the ancient wisdom: 

Initially, the TCP managed both datagram transmissions and routing, but as the protocol grew, other researchers recommended a division of functionality into protocol layers. Advocates included Johnatan Postel of the University of Southern California's Information Sciences Institute, who edited the Request for Comments (RFCs), the technical and strategic document series that has both documented and catalyzed Internet development. Postel stated, "We are screwing up in our design of Internet protocols by violating the principle of layering." Encapsulation of different mechanisms was intended to create an environment where the upper layers could access only what was needed from the lower layers. A monolithic design would be inflexible and lead to scalability issues. The Transmission Control Program was split into two distinct protocols, the Transmission Control Protocol and the Internet Protocol. 

The layering made the Internet as we know it. By the simple trick of just one node needed to permit another. Unstoppable inclusivity! 

[d] The Mastercoin / Omni Layer [12]: 

«A common analogy that is used to describe the relation of the Omni Layer to bitcoin is that of HTTP to TCP/IP: HTTP, like the Omni Layer, is the application layer to the more fundamental transport and internet layer of TCP/IP, like bitcoin». 

[e] The Lightning network (LN) [13]: 

The Lightning Network is a "second layer" payment protocol that operates on top of a blockchain (most commonly Bitcoin).  

Satoshi spoke on 'payment' channels in his masterpiece. Foreseeing the way to scale. 

An estimate of the power of LN layering [14].: 

''The bitcoin devs accept that eventually larger block sizes will be needed. The current transaction rate isn't going to cut it if people all over the world actually start using bitcoin daily. They estimate that eventually, if everyone in the world uses bitcoin and makes 2 transactions a day, but uses the lightning network, a 133mb blocksize will be needed. Without the lightning network, something like a 200gb (GIGABYTE) size PER BLOCK would be needed to accommodate that much usage.'' 

Layering upscales it with orders of magnitude of higher efficiency. 

If Bitcoin is the 'first layer' and Omni and Lightning are 'second layer', I see which one is the 'Zeroth Layer' and also foresee [15] the inevitability of the merger or 'Amalgamation' of all second layers over all blockchains, so the user will be able to transact everything into anything to anybody, without to know or care which chain is in use ... I have special nicknames for these and will go back to these topics in series of future posts. 

Enough of examples I reckon. 

The Postel's sacred Principle of Layering comes from the implementation levels paradigm. 

or Abstraction layering [16]: 

''separations of concerns to facilitate interoperability and platform independence''

With other words - delegate the task to that layer of the system which does the particular job best. We can generalize this into The Scaling Commandment. Only one enough: 

''Thou shalt not jam it all into a single layer!''


The Layer Cake architecture is literally ubiquitous across the Universe.: biology, semantics, informatics ... 

It seems that it is if not the only, at least THE way to scale. 

Maybe, someday, we the Humanity, upscaled by Tauchain will discover more powerful than Layering ways to Scale, but it is all we have for now. 

Scaling is a problem. Scaling must be scalable, too. 

Metascale from here to Eternity.

References: 

  • [16] - https://en.wikipedia.org/wiki/Abstraction_layer

  • 18qSKUUTAGw1uL53simrSiZ6pJpfxKACvj for research support. Thanks. 

    Copyright © 2018 Georgi Karov. All rights reserved.


    Sort:  

    Nice writeup @karov. Scaling is indeed layering. As a big supporter of both Steem and Tau, I'd add that the adoption and usability of Steem (https://blocktivity.info/ & devlog) is showing there may be many DPOS-like layers supporting various currencies coming up to grow like mushrooms and perhaps on the security side of things, match hard-to-scale POW protocols, at least energetically. With Steem devs working on Smart Media Tokens (the generation of more Steem-like currencies for different communities), I took this "layering" into the extreme by exploring a short essay a while ago here.

    Tauchain isn't out yet, but I expect it to be paving the way for scaling discussion and consensus-formation while making software / hardware development deeply accessible to the world's population as time goes on, including improving user/application discovery. Can't wait to see how both of my favourite projects play out together in the future!

    PS: I got a very similar tattoo with the head of snake meeting tail, but its shape twisted into ∞ :)

    Hey, Kevin. Valuable comments, thank-you! The blocktivity and devlog links are very useful. Network activity / traffic are, yes, maybe the best weight-scaling of the Metcalfe's law 'coefficient'. Measure of true organic value. It sounds exactly inline with the Warren Buffett's investment main criterion - scan and detect the undervalued and buy... :)

    As Ohad said Dan's DPOS is a soft-decentralization form. I again will refer to: http://www.truthcoin.info/blog/pow-cheapest/ of Aug 2015 by Paul Sztorc, which reveals that non-PoW , hybrid to fiat/trust systems are not 'energetically' more efficient if we include in the account all the energy budget incl. the societal one which is exploited by the blockchain.

    Layer mashrooming is very precise poetics, Rhyming with another form of layer scaling.: https://en.wikipedia.org/wiki/Eternal_inflation of A Vilenkin :) ... ''Both Linde and Guth, however, continued to support the inflationary theory and the multiverse. Guth declared:
    <It's hard to build models of inflation that don't lead to a multiverse.>'' with Vilenkin-Garriga's https://arxiv.org/abs/gr-qc/0102010 ...

    On Tauchain empowering the world's population by 'deep accessibility' -- true, I expect Tau as an unintended byproduct to be also the ultimate teaching machine - you'll see those suspicions of mine expressed in the next article to roll out in few days.

    I also can't wait to see Tau in another aspect: It will be the tool to assess thoroughly the Cryptoverse and to put it upright. Providing us with powerfull tools to grok and distinguish and to boost value in various consensi constructs. We'll be able to see instantly the worth in every proposal and automatically to funnel the dev and impl means into it.

    Your 'everybody - coin' article is a WOW! for me. I have an 'embryo' of an article - will gestate in few weeks with working title ''Selfcoin'' - very much inline, and I definitely will refer to yours and part of it will be exegesis of your one. :)

    Tatoos: I have a hands opening an eye one - which inspires me to find and use such an image for my Tau Edu upcoming post. :) Ouroboros is truly infinite because it grows as it devours itself. The First law of thermodynamics :P :P

    What a nice article! How it is possible?

    Thank you for the nice words! Few dozens blueprints in the pipeline ...

    Wow wow, outstanding post. How you did all this work.

    Thx! It is years of accumulated impression, which come out now as those posts :)

    Comment removed

    A very eloquent summary of things past and to come. I look forward to the next one!

    Thanks, Steemtruck. They (the themes) more or less converge in ... meaning. Like multiple storylines :)

    Btw, ''an environment where the upper layers could access only what was needed from the lower layers'' we observe also in physics->chemistry(i.e. electronics)->genetics->biology->neurology->psychology ...