Editors Note: This article comes fromPolkaWorld（ID：gh_6c4c2038ddba）, reprinted by Odaily with authorization.
Editors Note: This article comes from
, reprinted by Odaily with authorization.
Note: This article was originally published on gavofyork (Gavin Wood)s blog Insights into a Modern World on April 17, 2014, when Gavin was still serving as the co-founder and CTO of Ethereum, and Polkadot had not yet been founded. In this blog, Gavin comprehensively explained what the Web 3.0 era should look like in his mind, and the four components that make up Web 3.0. This age-old article conveys Gavins thoughts at the time, and now he is leading Polkadot to turn these ideas into reality step by step. As we move into the future, we see a growing need for zero-trust interactive systems. Long before Snowden, we realized that entrusting information to arbitrary entities on the Internet was fraught with dangers. In the wake of Snowden, however, this view has clearly fallen into the hands of those who believe that large organizations and governments often seek to expand and outstrip their own power. So we realized that entrusting information to organizations is often a fundamentally broken model. These organizations dont mess with our data simply because the effort to do so outweighs the expected benefits. Given that the models they like to adopt require them to have as much data on people as possible, realists will realize that changing information misuse is incalculably difficult. Protocols and technologies on the web, and even on the Internet, acted as a great technology preview, with SMTP, FTP, HTTP(S), PHP, HTML, Javascript these major tools all contributing to the cloud-based applications such as Google Drive, Facebook, and Twitter, not to mention countless others such as gaming, shopping, banking, and dating software. In the future, however, many of these protocols and technologies will have to be redesigned based on our new understanding of the interplay between society and technology. Web 3.0, or what could be called the post-Snowden Web, is a reimagining of all the things we already use the Web with, but with a fundamentally different interaction model between the parties. We publish information that we think can be made public. For information that we believe has reached a consensus, it will be placed in the consensus ledger. Information we consider private is kept confidential and will never be disclosed. Communication always takes place over encrypted channels and only with anonymous identities as endpoints. Never include anything traceable (such as an IP address). In short, since we cannot reasonably trust any government or organization, we design the system to mathematically enforce our previous assumptions. The post-Snowden Web consists of four components: static content publishing, dynamic messaging, trustless transactions, and integrated user interfaces. First of all, we already have a lot of things: a decentralized, encrypted information distribution system. All of these operations take some short inherent address (hashing in technical terms) of some information and return the information itself after some time. You can submit new information to it. Once downloaded, since the address is inherent, we can guarantee it is the correct information. This static publishing system does most of HTTP(S) and all of FTP. There are many implementations of this technology, but the most handy example is Bit Torrent. Every time you click on a Bit Torrents magnet link, all youre really doing is telling your client to download the data with an inherent address (hash) equal to that value. In Web 3.0, this part of the technology is used to publish and download any (potentially large) static information that we are happy to share. Just like with Bit Torrents, we can incentivize others to maintain and share this information, but combined with other parts of Web 3.0, we can make this process more efficient and precise. Since the incentive framework is inherent to the protocol, we implement DDoS protection anyway (at this level) by design. Is this benefit not bad? The second part of Web 3.0 is the identity-based anonymous underlying messaging system. It is used to realize the communication between people on the network. It uses strong cryptography to ensure adequate security for messages. They can be encrypted using an identitys public key to ensure that only that identity can decode them. It can be signed with the senders private key to ensure that it really came from the sender and provide the recipient with secure proof of receipt of the communication. Shared secrets can provide opportunities for secure communication, including between groups, without proof of receipt. Since each of them provides the final message logistics, there is no need to use transport protocol level addresses. What was once a user or port and IP address are now hashed together. Messages will have a time-to-live, thereby disambiguating as much as possible between published messages that are active and immediate signaling messages that are expected to be transmitted in the network as quickly as possible. So there is a trade-off between latency and longevity. The actual physical routing will take place through an adaptive network system employing game theory. Each peer tries to assert that other peers incoming messages are valuable, thereby maximizing their value to other peers. Peers whose information is of no value will disconnect and take their place by connecting with some other possibly unknown (or possibly secondary) peer. To make it more useful to the peer, messages with some specific properties will be requested, such as sender address or subject (both unencrypted) starting with a specific bit string. In Web 3.0, this part allows peers to communicate in real time, update, and self-organize, posting priorities that do not require inherently trusted or later referenced information. In the traditional Web, this part is the bulk of the information transferred over HTTP in an AJAX-style implementation. The third part of Web 3.0 is the consensus engine. Bitcoin introduced many of us to the idea of consensus-based applications. However, this is only the first step. A consensus engine is a protocol for codifying certain rules of interaction, knowing that future interactions (or lack thereof) will automatically and irrevocably lead to strict adherence to the rules. It is effectively an all-encompassing social contract and draws strength from the network effects of consensus. The fact that the consequences of betraying an agreement can be felt in all other agreements is crucial to building a strong social contract and thereby reducing betrayal or willful ignorant change. For example, the more isolated a reputation system is from a personal social interaction system, the less effective that reputation system is. Because users place their intrinsic worth on what their friends, partners, or colleagues think of them, a reputation system combined with something like Facebook or Twitter will perform better than one without it. A particularly poignant example is the dilemma of whether and when to re-friend employers or dating partners on Facebook. A consensus engine will be used for all trusted releases and information changes. This will be achieved through a fully generalized global transaction processing system, the first viable example of which is the Ethereum project. The traditional Web cannot fundamentally solve the consensus, but can only rely on the centralized trust of authoritative institutions such as ICANN, Verisign, and Facebook, and simplify private and government websites and the software to build them. The fourth and final component of the Web 3.0 experience is the technology that brings it all together -- the browser and the user interface. Interestingly, this looks very similar to the browser interface we already know and love. It will have the URI bar, the back button, and of course, the main part will be assigned to the display of the DApp (webpage/website). Using this consensus-based name resolution system (unlike NameCoin in an application), the URI can be reduced to a unique address (ie hash) for that application front end. This can be extended to a collection of files needed for the front end (for example, an archive containing .html, .js, .css, and .jpg files) through the information publishing system. This is the static part of the DApp(-let). It does not contain dynamic content; rather, it is served through other communication channels. In order to collect and submit dynamic but publicly available content that must be absolutely sure of its provenance and must be persistent (as in fixed), such as reputation, balance, etc., there is a Javascript based API that works with its The consensus engine interacts. To collect and deliver dynamic, possibly private content that is necessarily volatile and subject to destruction or lack of availability, a p2p messaging engine is used. This brings up some superficial differences; well see addresses that look like traditional client-server URL patterns, such as https://address/path, start turning into new forms of addresses, such as goldcoin and uk.gov. Name resolution will take place through a consensus engine based contract and can be easily redirected or augmented by users. A period will allow multiple levels of name resolution, e.g. uk.gov might pass the gov subname into the name resolver given by uk. Due to the constantly flowing nature of information, browsers can automatically and inadvertently get information through the updates of the consensus backend and the maintenance of the peer-to-peer network, so we will see background DApps or small applications play a role in our Web3.0 experience play an important role. From time to time, well be updating what we care about, through dynamic icon infographics like the always-visible Mac OS Dock, or dashboard-style dynamic applets. After the initial sync process, the page load time will be reduced to zero because the static data is pre-downloaded and the dynamic data (served via the consensus engine or p2p messaging engine) is also kept up-to-date. While synchronizing, the user experience will be very solid, although the actual information displayed may be out of date (although this cannot happen easily and can be annotated). As a Web 3.0 user, all interactions will be performed safely, securely, anonymously, and many services are trustless. And for those cases where a third party is required, the tool would allow users and app developers to spread trust across multiple different, potentially competing entities, greatly reducing the amount of trust held by a single entity. Say hello to Web 3.0, the secure social operating system. original:https://gavwood.com/dappsweb3.html Translation: PolkaWorld Community
