A politicized hard fork is a black swan event and could have a serious effect on the value of one’s tokens, depending on which network will gain traction in the long run.
Yeahhhhhhh personally I wouldn’t want to invest or put money in a technology where a single developer could fork the network and instantly reduce the value of my tokens to zero.
As the community splits, developers often have to take a stance for one network or the other, which can result in a lack of necessary developer power. Miners also have to choose which network they continue supporting (read more: Part 2 - Institutional Economics & Governance of DAOs.).
Damn, this just does not happen in centralized currencies.
Technical protocol updates happen quite frequently, and don’t usually create too much controversy, especially when they involve minor technical upgrades.
This will happen with Ethereum, where proof-of-stake will drastically reduce the profitability of miners.
A split can occur when some nodes in the network continue to use the old protocol while all others use the new protocol.
Wow, this seems incredibly volatile and inefficient.
What a 51 percent attack cannot do is change existing transactions or fake transactions
You can change existing transactions by removing an old transaction and adding the altered old one.
In the ten-year history of Bitcoin, no manipulation by outside attackers has been successful.
This is probably because Bitcoin has yet to be adopted for anything useful other than wild speculation by investors. If companies’ business interests were tied up in Bitcoin, it’s highly likely that they would want to gain control of the network. If you can define what ‘reality’ is, that’s a huge amount of power, which is highly attractive to corporations (look at any monopoly in the US today).
Because of its computational intensity, the Bitcoin network is also very energy consuming.
And it’s killing the planet. Yay!
A rational economic actor would, therefore, refrain from cheating the system, as this would result in sunk costs of energy and infrastructure investment.
This is a huge assertion that needs evidence to support it. Hardly anyone under capitalism is ‘rational’.
The block reward gets reduced by 50 percent every 210,000 blocks, around every four years.
This causes the value of Bitcoin to spike, which decreases its usability as a currency. Bitcoin is incredibly volatile because of its highly limited supply and the sharp increase in demand, which decreases its usability as any sort of a currency.
(ii) it doesn’t pay to cheat because mining requires special-purpose computer hardware and consumes large amounts of power.
So it’s incredibly environmentally efficient and objectively worse than all existing centralized systems. Cool.
reward miners for adding truthful transaction blocks to the ledger
No. Proof of work rewards miners for being the first to mine a block correctly, not that it’s correct.
Public-private key infrastructure guarantees attack-resistant access control of one’s tokens
It decreases the likelihood of this happening, but it can’t guarantee it.
which made the economic cost of attacking the system disproportionate to the benefit of doing so
No. Benefits are subjective and not objective. It made the economic cost of attacking the network higher than it used to be.
For the first time in the history of distributed computing
Again, no.
Before the emergence of Bitcoin, it was believed to be impossible to achieve fault-tolerant and attack-resistant consensus among untrusted nodes in a P2P network.
All network participants have equal access to the same data in (almost) real time
No. Anyone who doesn’t have the storage space or network capacity to download the blockchain doesn’t have access to the data. The blockchain only becomes larger, which makes it increasingly likely that average users won’t have access to the data directly. They have to trust third-party institutions to provide the compute needed to explore the data on their own.
They provide a governance layer for the Internet.
???????????????
No they don’t??????????? The internet is not controlled by any centralized party (outside of maybe ICANN for domain names). It’s not possible for it to have any sort of a governance layer.
Only the owner of that address can request to send these tokens somewhere else.
Only the person with the private key.
due to the prohibitively large amount of computing power that would be required to do so, and taking into account extreme attack scenarios
This is not prohibitively expensive for large companies like Google, Amazon, Facebook, etc. who already have large amounts of compute and who have billions of dollars they could throw at this.
The consensus mechanism is designed to make it difficult to manipulate the ledger.
Difficult, but not impossible. This again just increases the likelihood that this won’t happen.
to make sure
To increase the likelihood. This is not possible to ensure.
Instead of a single trusted third party validating transactions through their servers with authority (single vote)
This is very much not how more centralized currencies work…
that all nodes in the network agree upon
That the majority of nodes agree on.
guaranteeing that each token is transferred only once
Again, tokens are not a thing here. Correct wording of this would be ‘this increases the likelihood that the same transaction would go through twice’.
Different people and institutions that do not trust each other share information without requiring a central administrator.
This is something that could happen before Bitcoin (example: Usenet). The only thing BTC did is increase the likelihood of that database’s integrity.
Unlike distributed databases, where data is distributed but managed and controlled by one single entity, blockchain networks allow for distributed control
This is not what a distributed database is. A distributed database is just a database that’s spread across multiple computers. Nothing about the definition says anything about organizational control.
Decentralized databases are a subset of distributed databases. So a blockchain is an example of a kind of distributed database because it’s not geographically stored in one place.
Instead, it manifests as an entry in the ledger that belongs to a blockchain address.
This is… incredibly misleading wording. Tokens aren’t stored anywhere in BTC because they’re derived values. If I have 0.5BTC in an account, there’s not ‘half a Bitcoin’ in my account. Instead, the transactions that add and subtract from my accounts’ value are summed up, which produces the derived value that’s my current balance. Bitcoin wallet balances are derived values, not a set of tokens that are added and subtracted from.
but which no single user controls
Misleading. Correct wording here is “that it’s unlikely that no single user can control”.
The hash value of a block therefore serves as a counterfeit protection that can be used to check the authenticity of a transaction on a ledger.
This is only true for old values and not for pending values. Pending values can be altered if significant power by one actor is acquired.
The Bitcoin protocol introduced a mechanism of making it expensive to copy digital values.
No. Bitcoin made it expensive to change its ledger of what happened and its representation of who has how much money. It made it expensive to trick the network and the network’s view of reality, not for it to be hard to ‘copy digital values’. This is misleading.
Before the emergence of Bitcoin, ideas around cryptographically secured P2P networks had been discussed in different evolutionary stages, mostly in theoretical papers, since the 1980s (read more: Annex - Origins of Bitcoin).
This has nothing to do with the internet and applications that are on the internet.
The way the Internet is designed today, one can spend the same value—issued as a digital file—multiple times, because digital information can be copied, and copies of that same digital file can be sent from one computer to multiple other computers at the same time.
No. TCP / IP / HTTP have no concept of what a ‘value’ is, they’re just ways to communicate / send information. ‘Value’ is an application logic concept, not something the Internet handles. The double-spending problem is something application developers have to solve in their own web apps, and it’s not a problem in all web apps. This point is incredibly misleading and false.
if distrust of centralized solutions is high enough to warrant current trade-offs in usability
And this is the key point. The issue is that users don't distrust centralized systems, and centralized systems are highly incentivized to keep things that way. When there are broad misuses of user data, users leave that platform or those platforms are regulated. Furthermore, platforms are able to use their massive profits to publish PR that significantly changes public opinion on these things. Look at Uber / Lyft and Prop 22 in California for an example of this happening.
In this case, the blockchain-client performs the functions of an HTTP client and a server, as all data is stored client-side
This is again misleading. There's lots of additional overhead here because the client has to wait for the network to reach consensus. It doesn't perform the 'server' functionalities, it performs part of the server functionalities.
At the time of writing this book such data is still, for the most part, stored on and managed by servers.
This is because downloading lots of small files from IPFS or any decentralized network is really, really slow. Google 'IPFS is slow' sometime or just try using it for anything.
HTTP and client-server interactions are the most efficient solution to this problem. Developers have no reason to switch frontends to decentralized alternatives other than for vague 'decentralization' handwaving.
blockchain network instead of a server
No, it communicates with individual computers on the blockchain network.
Decentralized Applications in the Web3
The graphic above is again incredibly misleading. Again, the current internet is not 'centralized', and you still have to communicate with individual servers in a decentralized cloud. You have to talk with individual computers at some point, you just have to talk with more of them. Maintaining connections to more computers is more inefficient than maintaining a single connection to a server.
Centralized systems have advantages and will likely prevail, at least for specific use cases.
The author has presented no evidence that users will ever want to mass-adopt decentralized apps.
Speed, performance, and usability are bottlenecks in the Web3 that will very likely be resolved over time, once the core components of the Web3 are up and running (read more: Annex - Scalability).
There is no evidence to suggest that this problem will be resolved anytime soon. Our current internet infrastructure was not designed to support single computers maintaining large amounts of connections to tons of different peers. Try using IPFS sometime: it's a bad experience because talking with lots of different peers and traversing over the DHT is slow and inefficient compared to asking a single endpoint for a value. There is no evidence to suggest that the speed offered by client-server interactions will ever be beaten by decentralized networks.
Fully decentralized solutions, such as IPFS
IPFS is implemented and can be used today.
so that the network stays attack resistant
Reward tokens don't ensure attack resistance, they reduce the likelihood of it happening.
The transition from “client-server Web” to the “decentralized Web” will, therefore, be gradual rather than radical.
The author has presented absolutely zero evidence that there will ever be such a transition, or that users would want such a transition.
have varying quality in terms of speed and reliability
Most all of these are incredibly slow, and the protocols / networks that do have speed have a larger amount of centralization.
Other Web3 Protocols
Couldn't comment on it, but the graphic above is incredibly misleading. The current internet doesn't have a single point of failure, it's not a 'single server' that has control over data (again, see federated networks), and P2P networks do not provide no unique point of failure. That's not something they can guarantee. They reduce the likelihood of that happening. Those are two very, very different things.
It serves as a distributed accounting machine recording all token transactions and performing computation.
Again, not what a blockchain network is. Blockchains do not give you arbitrary computation power.
A blockchain network is simply the processor for decentralized applications that operate on top of the Web3
This is not what a blockchain network is. A blockchain network manages a blockchain, which is an immutable, decentralized data structure. Certain types of decentralized networks build arbitrary data processing on top of this (like Ethereum), but that's not a 'blockchain'.
Apart from computation we need file storage, messaging, identities, external data (oracles) and many other decentralized services
Yes, this is the case. 'Web3' folks have to throw out a large amount of the decades of research and engineering that have gone into the current internet. They do this for incredibly arbitrary reasons that no average user wants or is asking for.
If I asked my grandma if she wanted to use a Facebook alternative that was slower, less efficient, and cost money, she would look at me like I'm crazy. The average user likes the current internet, and there's no reason to suggest that they'd want to move over to a 'Web3' based internet. If users cared about their data, we'd see significantly more data privacy legislation than we see now, or large amounts of users leaving platforms in droves. We see none of that.
Decentralized applications can manage some or all of their content and logic by a blockchain network or other distributed ledger.
The author fails to note that this is orders of magnitude slower and less efficient than any sort of centralized protocol or application (like a database) that we have today. Even in networks where majority consensus is not needed, the amount of network latency, compute power, and storage needed to record a single transaction is indescribably high compared to the amount needed to do that in a comparable centralized system.
The Web3 architecture leverages the collectively maintained universal state for decentralized computing.
Again, blockchains do not ensure unique state, that it's collectively maintained, or that that state is completely free from modification.
The Bitcoin white paper of 2008 initiated a new form of public infrastructure where the state of all Bitcoin tokens are collectively maintained.
No. Bitcoin mining is done by larger groups and / or organizations. It's maintained by multiple organizations, yes, but this statement is misleading. It's not controlled by everyday average Joes, but by larger companies with profit incentives and who, if they grew to a large enough size, could control everything on the network.
As a result of this, users are paying for services with their private data.
Yes, this is now what we observe. Finally something true in this paragraph.
Business models have, therefore, developed around targeted advertising that builds on the data sets collected, which provide “state” for these platforms.
I have zero clue what the author thinks 'state' is, but holy fuck please stop using buzzwords you don't understand.
What followed was targeted advertizing based on user behavior and the commodification of private data.
What followed was first untargeted advertising, then better untargeted advertising, then targeted advertising. Again, it's more profitable to do targeted advertising.
Therefore, many of these Web2 platforms needed to find alternative ways to profit from the free services they provided, and this alternative was advertising.
This is a false depiction of what happened. Advertising won out over all other forms of monetization because it's what users preferred. Companies didn't choose it because they wanted to create a specific kind of internet, they chose it because it was more profitable because people liked it more.
customers were often not willing to pay for online content with a recurring subscription fee, and micropayments are still not feasible, in most cases
Consumers don't like paying for information or online content just like they don't like paying for healthcare or taxes. People don't like paying for stuff because it makes them poorer.
Furthermore, since the early Internet was created around the idea of free information
This is false. The internet was created to solve a communication problem, not an access problem.
Relevant quote from Robert Taylor:
For each of these three terminals, I had three different sets of user commands. So if I was talking online with someone at S.D.C. and I wanted to talk to someone I knew at Berkeley or M.I.T. about this, I had to get up from the S.D.C. terminal, go over and log into the other terminal and get in touch with them.... I said, oh man, it's obvious what to do: If you have these three terminals, there ought to be one terminal that goes anywhere you want to go where you have interactive computing. That idea is the ARPAnet.
Web2 platforms contributed to a re-centralization of economic decision making, R&D decision making, and subsequently, to an enormous concentration of power around these platform providers
Web2 platforms, not web2 protocols. The protocols had absolutely nothing to do with this. IP, TCP, and HTTP do not care about power or control, that is a social construct.
However, wealth was mostly accumulated by the companies offering the services, and less by the general public contributing content and value to those services
This is how unregulated markets work. Wealth accumulates in the hands of a few, and the rest are left with scraps. We've seen this in nearly every unregulated market we've ever observed. There's no reason to suggest that decentralized alternatives wouldn't do the same.
While session cookies provide better usability, these cookies are created and controlled by a service provider, such as Google, Amazon, Facebook, your bank, your university, etc., whose role is to provide and manage the state of their user.
........this is how applications work. This is how any application has to work.
which meant that we had to resubmit our user information every time we were using a website
You still have to do this with session cookies, this is just a more efficient way of doing this. Session cookies are shorthand for the data that needs to be sent every time.
we had no browsing history, no favorite sites saved, and no auto-complete
This is ridiculously false.
For many reasons, centralized data storage became the mainstream form of data storage and management.
The main reasons are convenience and cost. Centralization is faster, easier, and cheaper for users. Decentralization is not.
Data could be stored centrally, or decentrally
And this is where the logical fallacy is. Nothing about the web's current protocols dictate that data must be stored centrally. Look at Mastodon or any other federated platform that doesn't store data centrally. It is false to say that the current internet is centralized: it's not, there's nothing in the protocols to make it that way. The reason why several large corporations have gained more control over it is because that's how capitalism works.
While trying to manipulate data on a server resembles breaking into a house, where security is provided by a fence and an alarm system, the Web3 is designed in a way that you would need to break into multiple houses around the globe simultaneously, which each have their own fence and alarm system
This is very misleading wording. The issue is not manipulation of data, it is access to the data. Google, Facebook, Amazon, etc. do not make money off of you by manipulating your data. They don't make money by changing your browser history or your search preferences or what you've viewed. That would remove their profitabilty.
Rather, they make money by reading your data and transforming it into something more useful. Like an oil refinery, they take something in a raw form (data) and extract something useful out of it (your likes and dislikes, preferences, triggers, etc.). They then use this derived data to try to sell you things.
Data in decentralized protocols is completely public. Anyone has access to the entire history of the blockchain, that's how it has to work. The data you send, therefore, has to be encrypted. The same is not true of the current internet: nobody has access to your Google search history except for you and Google, and it's very likely encrypted.
The analogy here would be "Trying to access data on a server resembles trying to steal someone's bank account number by breaking into their house through a complex security system, finding the number, and de-scrambling it. Trying to access data in Web3 is similar to the criminal having to do almost none of that, because your bank account number would be publicly accessible, but in a scrambled form. They would just have to un-scramble it."
This is possible but prohibitively expensive.
This is exactly the same sort of thing people said about private companies using our data to manipulate us, or third-parties using advertising platforms to manipulate elections. 'Prohibitively expensive' does not matter when large corporations have a huge incentive to access and manipulate your data.
in the absence of intermediaries
This is false: these protocols just make it harder to spot the intermediaries. Every cryptocurrency currently must have its value transferred back to a currency with actual value (eg. the dollar). This process must take place through a centralized institution (or intermediary), the currency exchange. Cryptocurrencies are an intermediary currency because their uses as a currency currently (hah) are very limited. Because of this, these centralized currency exchanges are effectively intermediaries for otherwise decentralized transactions.
providing a unique set of data—a universal state layer—that is collectively managed by all nodes in the network
This is false: decentralized protocols don't provide a unique state layer. There are infinitely many state layers that can exist. Anyone, at any time, can fork any particular blockchain and create a new state layer. There are multiple competing state layers currently.
Even inside those state layers, uniqueness is not guaranteed either, because there are infinitely many different representations / protocols for state management.
Blockchains guarantee immutability and certain kinds of uniqueness within the blockchain itself. They can't, and will never be able to, guarantee that other blockchains will exist, and most can't guarantee that other arbitrary representations of 'ownership' won't exist inside the system itself.
they also dictate all the rules and they control the data of their users
Yes, this is how capitalism works. Larger private corporations remove the power of individual consumers. This is not new.
However, the Internet we use today is still predominantly built on the idea of the stand-alone computer
Not correct. It's built on the idea of connecting to a single access point, which is very different from 'a stand-alone computer'. Tech companies distribute data across lots of computers and lots of zones, but you still have to access your data through them. You connect to different computers / servers each time you interact with a web app, but those are always managed by the same institution.
Can I trust those people and institutions that store and manage my data against any form of corruption—internally or externally, on purpose or by accident?
Decentralized protocols can never guarantee that corruption won't exist, just that the likelihood of it happening will decrease. No protocol or network or algorithm can get rid of corruption.
Centralized data structures not only raise issues of security, privacy and control of personal data, but also produce many inefficiencies along the supply chain of goods and services.
Decentralized protocols will always be less efficient than any sort of centralized protocols that exist. Even on smaller, 'greener' networks, there are still more efficient centralized protocols that can undercut them in terms of efficiency. The problem of trust introduces overhead that no centralized protocol has to deal with.
Every time we interact over the Internet, copies of our data get sent to the server of a service provider, and every time that happens, we lose control over our data.
Yes, this is how any protocol has to work. No protocol can ensure that nobody else has access to your data except you. The only way you can ensure that you have sole control over a given series of bits and bytes is by never exposing it to the public internet. As soon as you send those bits and bytes to someone else in any sort of readable form, encrypted or otherwise, they can do with that whatever they want.
No decentralized protocol can solve this problem. Even if you do encrypt, at some point you have to hand that series of bits and bytes off to some other computer that is beyond your control. Decentralized protocols can incentivize those other computers to act in a privacy-respecting way, but they have no way of guaranteeing that.
nor do we have a native value settlement layer
I would not view this as a 'problem' of the existing web. Never have I wanted a 'native value settlement' layer to exist in a protocol. The thing everybody likes about the web is that it is free to access, and adding a value settlement layer removes that.
We do not control our data
This is wrong. We have control over the creation of our data and who we give it to. If you don't like that Facebook uses your data to show you stuff that makes you angry, you can stop using Facebook (ie. not giving them more data). If you don't like that Google tracks you on the web, you can delete your Google account and install a browser extension that removes ads and stops tracking. You have the power to, at any time, stop tech companies from accessing new data and / or using your data to profit off of you.
What we do have an issue with is control of how our data is used. We have no say in the terms and conditions or privacy policies of tech companies. As an individual, you have no bargaining power in the transaction between larger tech companies.
However, we can change policies through collective action. If we boycott companies, companies change their policies. The issue is that most people simply don't have an issue with how tech companies use their data, even after being informed of the facts. So that doesn't happen.
The need of a constantly expanding market for its products chases the bourgeoisie over the entire surface of the globe. It must nestle everywhere, settle everywhere, establish connexions everywhere.
Colonialism is a product of the bourgeoisie expanding to maintain their power!
It has resolved personal worth into exchange value, and in place of the numberless indefeasible chartered freedoms, has set up that single, unconscionable freedom — Free Trade. In one word, for exploitation, veiled by religious and political illusions, it has substituted naked, shameless, direct, brutal exploitation.
Marx's view here seems to be that capitalism is in a lot of ways a lot worse than feudalism. Feudalism at least had subtlety: people under it at least had worth, some sort of decent level of understanding that they were human. Under capitalism, however, people are only valued at their market value. People are no more than products under capitalism: they are not valued for who they are, their humanity, or anything other than the amount of capital they can bring a business.
Feudalism still had tons of exploitation and was still terrible, and Marx doesn't try to distance himself from that fact. While he does say that capitalism is worse in that its exploitation is "naked, shameless, direct, brutal", it's still just that: exploitation. Both of these systems are unjust and exploitative, they just differ in levels of how exploitative they are and on what they base the value of the people living and working within those systems.