GoWithMi Whitepaper

Thursday, February 13, 2020
Download document
Save for later
Add to list
Whitepaper.io is a project of OpenBook

Ledger of Everything Global decentralized location service infrastructure that benefits all White Paper V3.0.0 GoWithMi Team 2019.08

Abstract GoWithMi aims to build a distributed map ecosystem through blockchain, AI and GIS (Geographic Information System) technology, which enables anyone (any node) to initiate and autonomously perform use, update and maintenance of maps without depending on any centralization organization in the entire process. The maturity of 5G, IoT, Edge AI and other technologies as well as the breakout of blockchain reveal the advent of the new era of Internet of everything, which will become the third generation of the Internet after the first generation of PC Internet and the second generation of mobile Internet. The spatial location information of people and things in the real world need map to realize digital, so that they can be perceived by the digital universe. In this sense, the map is the “Ledger of Everything” that records the spatial location. It provides a connection service from the real world to the digital world by recording the “spatial location” which the real world mapped to the digital world. Just like the original intention of the Bitcoin was to break the “payment monopoly”, GoWithMi was originally designed to break the “map monopoly”. In the first and second generation of the Internet, the map is in the hands of a handful of Internet tycoons represented by Google. This triggers a series of obvious drawbacks: abuse of personal information, leak of personal privacy [1], arbitrary price hike of map service that businesses need[2], or even shut down of map service [3], and so on. If this situation stay unchanged, the Internet of everything will still be threatened by “map monopoly”. To change this situation and break the map monopoly, GoWithMi explores the “decentralization” practice mainly from technology and community governance: 1. GoWithMi’s main chain – the technical realization of peer-to-peer map network Gaia Gaia is a peer-to-peer map network completely. Map data update, map service provision and maintenance can be initiated from any node in the network and cooperate as needed throughout the whole network without any centralization mechanism. The TPS or supporting governance mechanisms of the existing blockchain projects such as BTC, ETH, and EOS, cannot meet the application needs in non-financial scenarios. We propose the spatial geographic computing architecture (GeoMesh Computing Architecture) to allow the data structure and core algorithm in the map industry to combine with the peer-to- 2 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

peer (Peer-to-Peer) network and consensus mechanism of the blockchain in their “gene level”, so as to provide a peer-to-peer map network that combines the map service with the decentralized self-organization of blockchain. The detailed method is: ⚫ Designing a new basic map data structure GeoMesh, which enables basic map data to be segmented into a large number of discrete and small-sized GeoMesh data; ⚫ Designing a consensus mechanism like PoW (PoCW, Proof of Compilation Work) that allows multiple nodes to create and update map data by mesh (Mesh) like block-by-block ledger-keeping; ⚫ Realizing a peer-to-peer network mechanism that allows GeoMesh to be synchronized between nodes on demand; ⚫ Reconstructing the core algorithm to allow the node to provide a complete map service in a specified area based on the currently cached GeoMesh rather than synchronizing all map data; ⚫ Designing a proof mechanism (PoG, Proof of GeoMesh) to enable the node to prove its map service capability simply through the effectively cached GeoMesh; ⚫ Designing a ledger-keeping consensus design for blockchains in non- financial scenarios (DPoG) that allows all nodes to maintain the same voting rights and map service obligations to the ledger-keeping nodes. Based on the designing philosophy of GCA (GeoMesh Computing Architecture), Gaia will coordinate the computing resources of all nodes in the whole network effectively . Compared with the PoW consensus, the PoCW makes those node computing power which are participated in this consensus spend in the "map data compilation", a service benefits the map service, and let the computing power game between the nodes can be used to ensure the consistency of the core map service resource "GeoMesh". Based on the results of the PoCW work, the entire network nodes can synchronize the GeoMesh on demand to ensure the consistency of the entire network map service (PoG) and the fairness of the accounting node election (DPoG). Based on the planning of such computing power, under the premise of ensuring data consistency, service consistency and fair governance, the computing resources of most nodes of the whole network can be released to the map service itself, thereby increasing the TPS of the network by a thousand times or even thousands of times compared with that of the existing public chain. 3 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

2. The organization and governance of the map ecosystem and other business that goes with the main chain Gaia, including: ⚫ An AI and human-machine collaborative data collection network combining Token and Edge AI that could shorten the single data data collection time from ten minutes to ten milliseconds. ⚫ A decentralized governance mechanism for map data rights which takes into account both the investment efficiency and investment efficiency of map data. ⚫ A decentralized governance mechanism for core ecological business resources that ensure optimal utilization of core resources. If BTC is a Peer to Peer version of “financial ledger” , then GoWithMi is the Peer to Peer version of “Ledger of Everything”. The blockchain revolution represented by BTC, ETH, and EOS reveals the possibility of participants playing positive-sum game based on a distributed network to co-construct ecology. GoWithMi hopes to extend this kind of “self-governance” in positive-sum game to non-financial scenarios, to decentralize the “ledger-keeping rights” of the map - “Ledger of Everything”, thus build the next generation of global decentralized location service infrastructure that benefits all. 4 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

目录 Abstract .......................................................................................................................................... 2 Chapter I Mission and Vision .................................................................................................... 7 Mission .................................................................................................................................. 7 Vision .................................................................................................................................... 7 Chapter II The Challenges and Solutions for Blockchains in Non-financial Scenarios 8 Data and Computing Power .............................................................................................. 8 Consensus in the Real World ............................................................................................ 8 Governance of Map Data .................................................................................................. 9 Chapter III The Main Chain Gaia ........................................................................................... 10 Distributed Storage and Distributed Computing ........................................................ 10 [email protected] ............................................................................................................. 10 IPFS ............................................................................................................................ 10 Map Technology and Spatial Geographic Grid Computing Architecture ............. 10 Background Knowledge of Map Technology .................................................... 10 Spatial Geographic Grid Computing Architecture (GCA) .............................. 11 Root Node and Automatic Compilation ....................................................................... 14 Root Node ................................................................................................................. 14 Data Compilation Workload Proof (PoCW) ...................................................... 14 Automatic Compilation and GeoMesh ................................................................ 14 Edge node and Distributed Map Service...................................................................... 16 Edge Node ................................................................................................................. 16 Proof of Service Capability (PoG) ....................................................................... 16 Peer-to-Peer Network and GeoMesh ................................................................... 16 Map dAPI .................................................................................................................. 17 Ledger and DPoG ............................................................................................................. 18 The Governance of Ledger-keeping Right and GeoMesh............................... 18 Service Voting Proof DPoG ........................................................................... 18 Chapter IV Human-machine Collaboration in the Real World ........................................ 19 Data Collection and Map................................................................................................. 19 The Governance of Distributed Data Collection ........................................................ 19 Token-driven Edge AI and Distributed Human-machine Collaboration ............... 20 AI space data mining machine ............................................................................ 21 The Near Space Vehicle.......................................................................................... 21 Chapter V The Governance of Map Data ............................................................................. 23 From the Estate Monopoly in the Real World to the Data Monopoly in the Internet .............................................................................................................................................. 23 Decentralized Land Governance Mechanism: Harberger taxes .............................. 23 Spatial Data Node and Map Data Management ......................................................... 24 Chapter VI Map Distributed Business and Ecological Value Transfer ........................... 25 5 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Maps are the top-level user traffic ................................................................................. 25 Traffic Management and Distributed Business .......................................................... 25 GMAT and Ecological Value Transfer.......................................................................... 27 The Incentives of GMAT ....................................................................................... 27 The Mortgage of GMAT ........................................................................................ 27 The Consumption of GMAT.................................................................................. 27 The Distribution and Usage of GMAT ......................................................................... 28 The Distribution Plan of Token............................................................................. 28 The Token Usage Plan ............................................................................................ 30 Chapter VII Team Introduction ............................................................................................... 31 Core Team .......................................................................................................................... 31 Investors and Advisors ..................................................................................................... 33 Investment Institutions..................................................................................................... 35 Partner& Customer ........................................................................................................... 36 Chapter VIII Development and Prospects ............................................................................ 37 Development Plan ............................................................................................................. 37 The Status of Current Products ...................................................................................... 38 Prospects ............................................................................................................................. 39 Ledger of Everything and Space Oracle ............................................................. 39 Worldwide Decentralized Location Service Facilities and “Prisoner's Dilemma” .................................................................................................................. 40 Privacy Map and Interest Map .............................................................................. 40 The Future of Distributed Autonomous Driving ............................................... 41 References ................................................................................................................................... 42 6 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter I Mission and Vision Mission GoWithMi is committed to becoming the global decentralized location service infrastructure by building a global human-machine collaborative map network and reshaping the map ecosystem through blockchain, AI and GIS technology. Meanwhile, as a decentralized map component for all public chains, GoWithMi is dedicated to facilitate blockchain industry to empower transportation, travel, accommodation, e-commerce, advertising, gaming, location intelligence, autonomous driving and other large-scale application scenarios that rely heavily on maps. Vision Internet infrastructure like map services should be designed and managed as public product. GoWithMi’s main chain, Gaia, will fill the global capital and technology gaps, allowing people in emerging markets to build a map network through idle computer or even idle phone rather than purchasing many expensive servers; allowing people to update and improve map services by themselves according to their own needs rather than relying on large international companies to collect bulk data; empowering blockchain and Internet developers around the world with a map dAPI that will not leak privacy, will not be forced to raise prices and shut down; empowering all businesses that need maps, including transportation, travel, accommodation and e-commerce, with a low cost and high quality underlying map service that never shut down. 7 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter II The Challenges and Solutions for Blockchains in Non-financial Scenarios Data and Computing Power The requirements for scalability and data storage for blockchains in non- financial projects far exceed the existing blockchain projects. Baidu Map needs to provide 120 billion location services per day [4]; that’s about 1,388,889 TPS every minute, which is 86,806 times of Ethereum, 347 times of EOS, and 694 times of TRON. The Gigabyte-level data storage required for map data is also significantly larger than the existing ledger data storage size. Obviously, it is difficult to meet such computing and storage needs under the existing blockchain technology framework and consensus. In this whitepaper, we introduce the GeoMesh technology architecture which enables the peer-to-peer map network to solve both the computing and storage needs of the map service, and propose a complete decentralized solution for all key services of the map service. For details, please refer to Chapter III. Consensus in the Real World From the perspective of game theory, the existing blockchain consensus can be abstracted into a positive-sum game based on distributed systems. Whether under the PoW or the PoS consenus, the game behaviors of participants has been abstracted into extremely simple behaviors, such as calculation of SHA256 hash and the mortgage of stake. However, existing Internet services require a lot of collaborative work in the real world. The complexity of such collaboration cannot be abstracted into simple behaviors. Although we can set up mortgages, competition and other measures to motivate and punish participants, participants’ learning costs and cognitive biases in work in the real world will still lead to unexpected outcomes in the consensus-driven human collaboration. 8 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Fortunately, the outbreak of the AI technology revolution brings us a new possibility. We can standardize and streamline the behavior of all participants with the aid of AI algorithm. By cooperating with Edge AI hardware, AI could automatically assist to complete the work that would otherwise take participants hours or even days to learn. Participants can complete the data collection work that meets the requirements simply by taking the equipment to the designated area. Such human- machine collaboration makes the consensus-driven collaboration in reality as elegant as PoW in BTC. Governance of Map Data Most Internet services can be considered as an extension of their data applications, and data is the core production material of the Internet. Most of the existing blockchain governance can be understood as the governance of “ledger- keeping rights”. Most projects do not design and answer the question of how to govern the ownership and the maintenance obligation of data. The core production material of map is map data. Based on its spatial characteristics, GoWithMi maps all the rights and maintenance obligations of the map data to different pieces of virtual lands that are corresponding to reality. Each piece of land becomes a space node (GoZone). Based on such data governance design, we have transformed the global map data governance into spatial node governance based on independent regions. We are fortunate to b design a spatial node governance mechanism based on “Harberger taxes”. In the digital space world, “Harberger taxes” does not have real-world constraints, so we could govern the rights and obligations of data elegantly through the governance of spatial nodes. 9 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter III The Main Chain Gaia Distributed Storage and Distributed Computing [email protected] [email protected][6] is the world’s best-known and largest distributed computing project. It began to use distributed computing to analyze electromagnetic waves from the universe and find extraterrestrial civilization in the 1990s. The success of [email protected] proves the feasibility of providing services through distributed computing. Inspired by [email protected], Gaia regarded map service as a public course that the international community could participate in the same way as finding extraterrestrial civilizations, and provide global decentralized location service infrastructure through the integration of idle computing power. IPFS IPFS is a master of successful open source distributed systems in recent decades, providing an excellent distributed storage protocol solution for subsequent projects. Gaia focuses on the development of distributed map services on the network transport layer. It can be considered as a distributed map network that realizes some IPFS protocols in some sense, which is exemplary for the distribution transformation of the subsequent Internet projects.. Map Technology and Spatial Geographic Grid Computing Architecture Background Knowledge of Map Technology In general, map technology can be divided into two parts: ⚫ Map Data Compilation A series of data collected from the real world, such as road shape, traffic rules, the name and location of POI (Point Of Interest), cannot be directly used by commercial-level map services. To reorganize the data structure, different indexes need to be added based on different map service needs. For example, POI name 10 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

queries needs to add inverted indexes; POI peripheral queries or map displays need to add spatial indexes; and path calculation needs to convert road data into multiple layers of strongly connected directed graphs. These tasks are generally weekly- level because they are full-scale structure conversion of existing data. ⚫ Map Service The map service provides a series of location-based services such as place-name query, map display, path calculation, and route guidance on the basis of generating map data though compiling the collected data. All map-related business services can be regarded as the combination of these basic map services. Obviously, we need a complete and exclusive transformation of the map technology to provide a peer-to-peer map solution. Spatial Geographic Grid Computing Architecture (GCA) The spatial geographic computing architecture (GCA, GeoMesh Computing Architecture) proposed in this white paper represents a technical system and architecture design for distributed map services, of which the design philosophy is inherited from Grid Computing [7]. “Mesh” could better illustrate the spatial properties of the map services in the geographic information industry. GCA aims to design a new map service based on P2P network. Through the redesign of the basic map data structure, the map data can achieve distributed production based on PoCW (Proof of Compilation Work) consensus. While the map data consistency among the network is guaranteed, each node synchronizes data on demand, and achieves the network-wide consistency of the map service through the network-wide consistency of the map data, so that most nodes of the entire network no longer waste valuable computing resources on competing the ledger-keeping right under PoW. Instead, the resources are devoted to the map service itself, thereby increasing the TPS of the network by a thousand times or even thousands of times compared with that of the existing public chain. The core idea of GeoMesh is to make the map data which organized through GeoMesh the core service resource in Gaia network. All nodes can prove their own 11 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

map service capability (Proof of GeoMesh) by themselves by storing and maintaining a part of GeoMesh data. And from the perspective of Gaia network, each node’s holding of service resources like GeoMesh is the proof of the contribution to the map service. Each node uses the effective holding of GeoMesh as a stock to vote for the ledger-keeping nodes in the network. This is called DPoG (Delegated Proof of GeoMesh). In general, the Spatial Geographic Computing Architecture (GCA) can be divided into the following parts: ⚫ A new basic map data structure GeoMesh which enables basic map data to be segmented into a large number of discrete and small-sized GeoMesh data; ⚫ PoCW (Proof of Compilation Work) consensus mechanism that allows multiple nodes to create and update map data by mesh (Mesh) like block-by-block ledger- keeping; ⚫ A peer-to-peer network mechanism that allows GeoMesh to be synchronized between nodes on demand; ⚫ The core algorithm goes with GeoMesh that allows the node to provide a complete map service in a specified area based on the currently cached GeoMesh rather than synchronizing all map data; ⚫ A proof of the node’s map service capability (Proof of GeoMesh). The node could prove its map service capability simply through the effectively cached GeoMesh; ⚫ A ledger-keeping consensus for blockchains in non-financial scenarios (DPoG) that allows all nodes to maintain the same voting rights and map service obligations to the ledger-keeping nodes. The map service features access aggregation. Hardly do the taxi driver in Beijing need to know the location of the Los Angeles supermarket. Access aggregation makes GeoMesh enjoy a very high cache hit rate, thus making it possible to provide city- level map services with a single node with limited GeoMesh. The core of GeoMesh’s data structure is decoupling and parallelizing map data. Decoupling means that an independent GeoMesh data can provide the data needed for map services in a few square kilometers without any other data; Parallelizing means that several adjacent GeoMesh data can provide all the data of the map service in the corresponding range in parallel without any additional processing means. 12 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Figure 1. GeoMesh segements the map data. The data structure features of GeoMesh allow map nodes to synchronize and cache GeoMesh as simple as the way IPFS nodes cache images. In terms of network structure, I designed a two-level node structure, which is Root Node and Edge Node. Any hardware device, including server, PC or mobile phone, can be a node to participate in the Gaia map network. They can choose to become one kind of node or both kinds of nodes at the same time. Figure 2. A demonstration of Gaia nodes 13 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Root Node and Automatic Compilation Root Node The “root” in the “root node” comes from the root name server [8]. It uses the role of the root name server in the Domain Name Service (DNS) [9] to analogize the role of the root node in the Gaia network. In the actual operation of the DNS system, most of the domain name resolutions do not need to access the root name server, and the mapping relationship of the top- level domain (TLD) [10] has been cached in the nodes of the DNS system. But we can rebuild the entire DNS system based on the root name server at any time. In the Gaia network, GeoMesh data is generated between the root nodes through PoCW. GeoMesh will synchronize to the Gaia nodes as needed. Most of the map service requests do not require the root node, either. The root node’s role in Gaia is more like a provider of rules than a provider of services. Data Compilation Workload Proof (PoCW) Bitcoin’s PoW mechanism provides a very good example of how to maintain the data consistency among the whole network. Based on the “one CPU, one vote” principle, the decision of the “honest majority” is expressed as the longest chain which outruns other competing chains, thus ensuring the ledger data of the whole network is consistent. Proof of Compilation Work (PoCW) inherits the design principle of PoW. It makes the data compilation of map production a legder-keeping process and the GeoMesh-based data compilation and production a block production process; it also makes the decision of the “honest majority” expressed as the longest chain, thus ensuring the map data consistency among the whole network. Automatic Compilation and GeoMesh The automatic compilation of map data without human intervention is the premise of the de-centralized production of map data. So far, there is no map data automatic compilation project available for commercial use worldwide. We need to make different modifications based on the data compilation of different map service modules. Fortunately, we are “standing on the shoulders of giants”. 14 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Spatial index technology is relatively mature in the map industry. Quadtree [11] has almost become the industry standard; Web Mercator projection [12] and GeoHash [13] provide very good references in application. The inverted index technology is also relatively mature, and a large number of existing projects such as Solr [14] and ElasticSearch [15] provide good references for creating real-time index. The difficulty of GeoMesh technology lies in the automatic compilation of path planning, of which the purpose is to transform the road network composed of all roads into a strongly connected directed graph [16], so as to use the Dijkstra algorithm [17] or the A* algorithm [18] to obtain the optimal path from location A to lacation B. Although “boundary node” technology allows road network data to be segmented based on spatial index without additional indexes, the inspection of road networks and even the production of strongly connected directed graphs require human intervention for a long time, which is almost the most labor-intensive part of the indoor operation in traditional map business. Fortunately, the advancement of the algorithm makes full automation possible. Tarjan’s algorithm [19] allows the system to automatically detect “isolated subgraphs” on the road network, preventing drop-dead halt in path computation. Contraction hierarchies algorithm[20] provides a way to rarefy road network without losing any optimal path between higher nodes, making the high-level road network possible, which the automatic compilation and production of mainland-level path calculation requires. Figure 3. Multi-layer bidirectional A* pathfinding algorithm that is almost the industry standard. 15 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Edge node and Distributed Map Service Edge Node The edge nodes in the Gaia network have no cap on the number. The role of the edge nodes is to provide nearby map services based on their own hardware, network conditions and geographic location. The maps could range from a city, a country or to the whole world. Edge nodes can be deployed on servers, PCs, and even mobile phones. We call it “everything could mine”. Proof of Service Capability (PoG) PoG is a mechanism that allows Edge Nodes to prove their map capabilities through the effectively cached GeoMesh. After Gaia guarantees the map service consistency through GeoMesh data consistency based on proof of compilation workload (PoCW) , the spatial index of GeoMesh in parallel directly describes the range of map services that can be provided. In this way, Edge Node can prove its service capability by caching such service resources of GeoMesh, which referred to as PoG (Proof of GeoMesh). Gaia matches the Edge Node which proves its service capabilities through PoG with the nearby corresponding map service needs. Peer-to-Peer Network and GeoMesh One of the original design purposes of GeoMesh data structure is to make the size of the map data conform to the transmission requirements of the P2P network. In the PoC (Proof of Concept) test, one single GeoMesh data can be controlled at around 800K, and can provide map display and path planning services within the range of 3.6*2.4 square kilometers.. 16 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Figure 4. The available map area of 800k GeoMesh in the PoC test. With GeoMesh as the carrier, the global map data is transmitted and stored in the P2P network in a distributed way. Each node only needs to synchronize and store the required parts. Gaia’s map service will be provided by a large number of edge nodes scattered around the world. Since real-time synchronization is not required, the node’s own computing power is used to provide map services. In this way, Gaia provides low-cost map services with no TPS limit theoretically. Map dAPI Based on the P2P map network, Gaia can provide a secure and transparent map service that never ends, which we call dAPI. Any blockchain project, Internet company or individual developer can deploy their own edge nodes based on open source code in the future. The deployment environment can be a secure environment chosen by the developer. After the edge node is connected to the network, the developer only needs to pay Gas to get the map service. 17 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Figure 5. dAPI based on P2P network Gaia provides decentralized map dAPI in a completely transparent way, eliminating the possibility of evil conduct at the center. Ledger and DPoG The Governance of Ledger-keeping Right and GeoMesh We tried to design a consensus to govern the ledger-keeping right on Gaia, so that each node’s right to vote for the ledger-keeping node is related to its contribution to the entire network map service. GeoMesh provides a simple solution to directly link the map service capability of each node to the nodes’s effectively cached map data. So the contribution of each node to the whole map network ecology can be abstractly quantized into the node’s effectively cached map data in GeoMesh format. Service Voting Proof漑DPoG漒 DPoG can be considered as a variant of DPoS (Delegated Proof of Stake) in essence, except that we convert the interest brought by the holding into the interest brought by service resources held by the nodes (GeoMesh). Each node obtains voting rights to the ledger-keeping node based on the valid GeoMesh cache, thereby unifying their rights and obligations in Gaia. 18 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter IV Human-machine Collaboration in the Real World Data Collection and Map Information like the location of the store, the name of the road, etc., need to be collected and processed by people to form the original data in the electronic map in a standard format. The map industry refers to such work as “map data collection”. In traditional map enterprises, the biggest cost comes from data collection. Both Chinese American companies finish this work through intensive labor. Even Google can’t invest enough manpower and material resources in each country in the emerging markets to ensure that map data quality in emerging markets is the same as in developed countries. This is why Google Maps can’t deliver map services of the same quality throughout the world. Distributed data collection does not have this problem. Through the management of consensus and incentives, we can break through the organization and management limitations in traditional industry and let people all over the world participate in data collection. The Governance of Distributed Data Collection Most governance of distributed cooperation can be considered as the establishment of a self-assessment system. All self-assessment systems can be easily divided into two parts: self-declaration and verification by others. So distributed data collection is also divided into two phases: ⚫ Collection phase (self-declaration): After collecting the data, the data does not immediately enter the compilation process. The collector needs to mortgage his or her own token to inspire others to verify. ⚫ Verification by others: After the collector mortgages the token, it will attract other participants to verify the data. If the data is verified to be true, both the collector and the verifier will acquire incentives. If it’s verified to be false, the verifier will split the collector’s mortgaged token. 19 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Token-driven Edge AI and Distributed Human-machine Collaboration Edge AI can be considered as a distributed revolution in AI computing, freeing AI computing power from the constraints of the center and providing it to edge devices. We believe that Edge AI is naturally driven by Token, and this Token-driven Edge AI is naturally combined with human-involved data collection activities. We will use GIS and Edge AI technology to create an ecological spatial data mining machine. The carrier can be AI hardware for car, near space vehicle or even satellite. We’ll make map data collection process standardized, automated and efficient to let the consensus algorithm perfectly facilitate human to participate in collecting map data in the real world, thus realizing perfect human-machine collaboration, interpersonal collaboration and machine collaboration. Figure 6. AI collaboration of token-driven Air-ground integration The AI data collection protocol will be open to third parties, making it possible for all Edge AI devices to participate in data collection. 20 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Currently we have made the following AI collection attempts: AI space data mining machine Figure 7. AI hardware in development This is GoWithMi’s exclusive data collection hardware. After the miner fixes the equipment on the vehicle, the camera on the equipment will receive continuous driving video. The customized AI chip will automatically identify the traffic signs, lane lines, intersection turns, POI along the way and other spatial data that maps need. The AI hardware shortens the time of collecting one single unit of map data from ten minutes (collected completely by human) to ten milliseconds, which greatly accelerates the data collection process. The Near Space Vehicle The near space vehicle will continue to hover at a fixed altitude of 20,000 meters for several months, with a “gaze” range covering large cities. Figure 8. Aerial view of a near space vehicle 21 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

The near space vehicle could provide decimeter-level picture within the gaze range. All the spatial attributes of a city such as green spaces, water systems, roads, and buildings can be obtained directly through the picture. Figure 9. Image sent by the near space vehicle With near space vehicle, you can get the basic display data of a city in a few days or even hours, and can detect all events in the city in real time, from traffic jams to social events (demonstrations, riots) and even natural disasters (tsunami, volcanoes). 22 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter V The Governance of Map Data From the Estate Monopoly in the Real World to the Data Monopoly in the Internet The status quo of the Internet is similar to that of Europe more than two hundred years ago when the landlord holds the core production material -land. All the dividends acquired in the industrial revolution since 1750 were in the hand of the landlord, and the wages of the working class hadn’t changed for one hundred years. The same story is happening in the Internet ecosystem where large Internet companies control data - the core production material in the Internet economy, and gain a lot of benefits, yet the Internet users can’t get any return. After making the distributed management of data technically possible through Gaia, we need to provide a mutually beneficial data governance system. Decentralized Land Governance Mechanism: Harberger taxes Harberger taxes is essentially a “self-assessment system” for the land’s value. Through the owner’s self-assessment of land prices and forced circulation trading, Harberger taxes takes into account both investment efficiency and investment efficiency in a decentralized way. To be specific, any landowner can declare the price of his land in accordance with his own ideas and pay taxes according to the price. Meanwhile, anyone can purchase the land at the price set by the owner. If the owner declared a low price to lower the tax, his land will soon be bought by others; if the owner declared a high price to prevent others from buying the land, he or she has to pay a high tax continuously. Based on such a game, the owner will declare the price of his land reasonably and eventually transfer the land to the person who can truly maximize its value. Harberger taxes manages the rights and obligations of land in an elegant way. Neither there will have “public land effect” resulted from unmaintained and over- developed shared land, nor “land monopoly” where owners refuse to fulfill their obligations and just reap the benefits. 23 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Spatial Data Node and Map Data Management We believe that Harbiner taxs can also be used in map data management. The detailed approach is to divide the areas on the map into virtual land of map data according to the corresponding reality. The map data will be assigned to a certain piece of virtual land with certainty according to its spatial characteristics, and each virtual land is a spatial data node (GoZone); The global map data is divided into independent nodes based on the space node for governance; “Harberger taxes” governance ensures optimal ecological governance efficiency between space data nodes in a decentralized way, allowing virtual land to transfer to the people who can truly exert its value and mobilizing the community to participate in the map ecological construction with the highest efficiency. 24 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter VI Map Distributed Business and Ecological Value Transfer Maps are the top-level user traffic 20 years development from PC Internet to the mobile Internet proves that the world’s top-level traffic only has four types: e-commerce, social, information and map. Only applications in these four areas can serve hundreds of millions or even billions of users for years. Especially in the era of the third generation Internet of everything represented by IoT+5G+Edge AI, the map will become the only top-level traffic that will connect more than 50 billion IOT devices. The trend of Internet development in the past two years is the exploration of monetizing these top-level user traffic. Pinduoduo (Nasdaq: PDD) which monetizes the traffic in social and Qutoutiao (Nasdaq: QTT) which monetizes the traffic in information provide us with good references. Traffic Management and Distributed Business User traffic is as precious to the Internet economy as oil is to the industrial society. We need to manage this core resource more fairly in a more distributed way. Distributed business is such a governance method. The detailed idea is to abstract traffic conversion services such as discount coupon and advertisements into resources without threshold that everyone can bid for, and the highest bidder get to use the resources. In this way, the resources will be transferred to the merchants with higher conversion efficiency. 25 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Figure 10. Discount coupons and digital hygiene ads along the dApp This decentralized method will ensure the effective use of map traffic. The GMAT consumed by the auction will be re-allocated to the owner of the GoZone space node where the advertisement resource is located, so as to incent its maintenance of the map data and local business promotion of the distributed business. Figure 11. Token circulation and its role in distributed business 26 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

GMAT and Ecological Value Transfer GoWithMi’s token is called GMAT. GoWithMi uses GMAT to realize the value transfer from data production to map service or even to distributed business. The Incentives of GMAT ⚫ Root Node participates in data production under PoCW consensus to obtain GMAT incentives. ⚫ Edge Node provides map service to get GMAT incentives ⚫ Providing ledger-keeping services based on the ledger-keeping node selected by DPoG to obtain GMAT incentives ⚫ Map data collectors complete effective data collection to get incentives ⚫ Map data verifiers obtain incentives by verifying data ⚫ GoZone space nodes can obtain incentives for distributed business The Mortgage of GMAT ⚫ Map data collectors in the real world need to mortgage GMAT as a credit endorsement ⚫ In the real world, merchants need to mortgage GMAT as an endorsement of their service quality before placing advertisements The Consumption of GMAT ⚫ Calling the dAPI service consumes GMAT ⚫ Holding GoZone as a space node consumes GMAT ⚫ Participating in the competition of advertising resources and advertising consumes GMAT 27 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

The Distribution and Usage of GMAT The Distribution Plan of Token Distribution Proportion Quantity Explanation Plan It is used for long-term system development, community construction, business ecology and research of cutting-edge science and technology, targeting supporters who can grow as the project processes. Among them, the 6% seed lock-up period is 19 months, the initial unlock is 5%, release 5% one-month Private sale 12% 1,788,000,000 after landing on the exchange, and release 15% every three months thereafter; 5% private sale, the lock-up period is 6 months, the initial unlock is 10%, release 15% one week after landing on the exchange, release 25% every two months thereafter; 1% pre- sale, no lock-up period. It is used for sustainable community ecological construction and maintenance. It’s open to lawful citizens of countries other than Chinese mainland and the United States who are not Gate.io Startup 9% 1,341,000,000 subject to regulation influence and are willing to participate in the co-construction of GoWithMi distributed smart maps. There is no lock period. The founding team made outstanding contributions to the birth and construction of GoWithMi. This part is used for related teams and personnel who contribute to the early Founding team 10% 1,490,000,000 construction and continuous operation of GoWithMi. The lock-up period is 42 months, and then 10% will be unlocked after the project goes online for half a year and 15% every six months thereafter. 28 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

It is used to invest in the early stage of R&D, project operation, marketing and etc. It is for early supporters and consultants who have Early supporters 8% 1,192,000,000 made outstanding contributions to the development of GoWithMi. The lock-up period is the same as founding team’s. In order to promote the long-term prosperity and ecological development of GoWithMi, continuous promotion and operation are needed. This part is used for long-term marketing, business cooperation, cutting-edge science and technology research and etc. Due Market and 16% 2,384,000,000 to the large investment required in the early commerce stage of Southeast Asia operation, the usage plan is established, 30% can be extracted within 6 months after the token issued, and 10% can be extracted every quarter after landing on the exchange. The development plan of each stage shall be carried out according to the Council. The source of GoWithMi's sustainable competitiveness lies in the continuous expansion and updating of map data, which is used to reward map ecology contributors and block producers. Gradually release each year after the spatial digital real estate and the main Mining 30% 4,470,000,000 network launch online, the release quota incentives reduced half every 2 years, complete mining in 10 years. Specifically, the first year and the second year respectively released 1,154,000,000 GMATs, and the third and fourth years respectively released 577,000,000 GMATs, and so on. It is used for follow-up development, establishing cooperative partner and improving ecology and etc. The use of funds is executed Foundation 15% 2,533,000,000 according to the foundation’s resolutions, and reserve related information will be published regularly. Start releasing 12 months after landing on the exchange and executed when two-thirds of the 29 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

member of the Council vote yes. The Token Usage Plan Project Proportion Explanation Research and It is mainly used for rewarding the founding team, recruiting development of 30% experts and developing personnel, technical patents and technology intellectual property protection. It is used for rewarding cold start, business development and Marketing 25% training, technology exchange and sharing, the publication operation of journal, alliance establishment or participation and etc. Business It is used for developing important business partners, 10% cooperation important cooperation channels and etc. Safety and It is used for code safety audit, legal compliance, third party 5% compliance audit and etc. It is used for incubation, investment and absorption of new Reinvestment 10% blockchain technologies, new teams, new projects that are conducive to the ecological construction of GoWithMi. It is used for sustainable development of projects and risk Reserves 20% response. 30 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter VII Team Introduction Core Team GoWithMi team is a cross-disciplinary professional team composed of continuous successful entrepreneurs in the global GIS industry, global map navigation engine technology experts, star product managers, blockchain technology experts, and local operation experts. The main members come from Navteq(Here), NavInfo (002405.SZ), JD.com(NASDAQ :JD), Sohu.com(NASDAQ :SOHU),development community of Ethereum and EOS. There are 29 people in the team, including 15 people in blockchain and map technology, 10 people in Indonesia operation, and 4 people in global market and business. The core members are as follows: Dong (Oliver) Li Founder & CEO Oliver is the world's leading expert in the spatial information industry. With over 20 years of global operating experiences and continuous successful entrepreneurship ranging from big data, mobile internet, and the map industry; One of his successful exits was as co-founder of Cennavi, the largest UGC map in China, being acquired in 2008 by NavInfo (002405.SZ), the largest listed map company in China. Oliver's next company is LBAdvisor Technology, a big data-focused company that was acquired in 2016 by Blue Focus (300058.SZ), a global top 10 PR company. Yi (Simon) Ren Co-founder & CTO Simon is one of the earliest core developers of map navigation engine in China. With over 15 years of map product development experiences. He is the top technology expert in the global location- based services (LBS) field, with complete front and back end and maps full engine self-development capabilities and International technical team management capabilities. He led the China map product team of SK Group, served as the head of the core product research and development for NavInfo (002405.SZ), 31 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Dong Yangfan Founding Partner & SVP Mr. Dong is the expert in commercialize the modern technologies for the LBS and O2O industries. He used to lead his company to provide the services for tens of thousands business partners cross China mainland, Hong Kong and Taiwan, and acquired millions of users.Mr. Dong is the expert in commercialize the modern technologies for the LBS and O2O industries. He used to lead his company to provide the services for tens of thousands business partners cross China mainland, Hong Kong and Taiwan, and acquired millions of users.With his experience in LBS/O2O industries and combined with his view of block chain technology, Mr. Dong is now leading GoWithMi’s business and market team to establish the distributed business environment for those business owners who were still suffering the lacking of I.T infrastructures in south east Asia. Dr. Jeff Flowers Blockchain technology VP Dr. Jeff is an early participant and communicator of the Silicon Valley blockchain community. He serves as an advisor to a number of Ethereum based projects such as POA as well as an active member of multiple decentralization communities in and around the Bay Area. He was also a Director and a Professor of Curriculum at Blockchain University and continues to teach at DLT Education. Iwan Suryaputra Founding Partner & COO in Indonesia Iwan worked for 20 plus years in mobile Internet and map industry in Indonesia. He served as CEO for PT Surya Teknologi Perkasa, a mobile monitoring and mapping service provider in Indonesia as well as CEO for Gowes, a bicycle-sharing company in Indonesia. 32 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Yuanyuan (Yolanda) Pan Founding Partner & CPO With over 10 years of product and project management experiences in the map industry software products, Yolanda is an outstanding product leader and had successfully participated in the creation of products with ten million daily users in China. She has a strong corporate strategy decomposition, execution, design and landing ability, served as a star product manager in the mobile phone map industry. Yolanda was a senior product manager for AMAP, a most famous and popular map software in China, and in charge of multiple subprojects including traveling, payment, searching, channel, and life services. With her leading of the AMAP’s product process system construction, operation and R&D management, users have a breakthrough from hundreds of thousands to hundreds of millions. Investors and Advisors Chuang Tao The world’s map expert, a successful entrepreneur, and well-known investor. Mr. Tao is a tenured professor at the University of Calgary in Canada and chief geospatial information advisor to the United Nations. He is the founder of Microsoft Bing Map, founder of PPTV and founder of Z Ventures Group, the world’s largest mapping investment organization. Congwu Cheng Founder and former CEO of AMAP. He led AMAP to become a top player in China's map industry. AMAP was listed in NASDAQ in 2010 and later delisted after being 100% acquired by Alibaba Group. 33 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Dr. Jerry Fuqua Blockchain Advisor, he is a continuous successful entrepreneur, investor, manager, and technology advisor. He is currently active in a number of areas involving crypto currencies, blockchain, and gamification. Jerry was on the Executive Committee of the MIT- Stanford Venture Lab (VLAB) and actively involved in the Silicon Valley crypto ecosystem, including efforts in promoting Silicon Valley’s first Bitcoin meetup community “SV Bitcoin meetup”. Jerry was an early pioneer of commercial web services, initially as the co-founder and CTO of Internet Information Systems Inc. and World Point Interactive. He was also involved in AI and robotics as a researcher and program manager for the US Department of Defense. As the Principal of Fuqua Associates, Jerry consulted for number of notable organizations including Sony, IBM, United Airlines, and the State of Hawaii. Suryandy Jahja The best IT investment expert in Indonesia. Mr. Jahja is the co- founder and CEO of Kresna investment, the largest IT investment group in Indonesia. Kresna owns several listed companies. In 2017, the overall performance of Kresna ranked the first in Indonesia. 34 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Investment Institutions 35 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Partner& Customer GoJek: The largest O2O service China Beidou civil Center China Lodging Group: Nasdaq listed company in Indonesia. company, China's largest hotel group. Bing Map: Microsoft map platform. Mapbox: The largest To B map WAYZ: Unicorn level unmanned map platform globally. data Enterprises. Mars Finance: Famous Blockchain Media. Shopee: The Largest B2C corporation in Zoomy:The largest mobile advertising Southeast Asia. company in Indonesia. Akulaku: The largest consumer lending TOYOTA Indonesia Mazda Indonesia company in Southeast Asia. HONDA Indonesia 36 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

Chapter VIII Development and Prospects Development Plan 2017 • The market research Q2~Q4 • The first phase of map data mapping platform based on image data has been completed, which can automatically generate high-precision road network, terrain and landform, building outline and so on based on satellite image map and UAV aerial photo, solving the cold start problem of map data,laying a foundation for low-cost and rapid replication in the future worldwide. • The research and development of "map automatic compile engine" has been completed, which has realized the full automation of map data production for the first time in the world. The update speed of map data has been successfully promoted to the day level, and it will reach the minute level in the future, laying a foundation for the decentralization map production on the main chain Gaia. • Launch of MVP 2018 • The mobile terminal map product is officially launched, which can provide a complete display map of Indonesia Q1~Q2 and a real-time 3D map of traffic in Jakarta. • The official partner of the Asian games in Indonesia, providing real-time traffic services for the Asian games with distributed maps. 2018 • Developed mobile AR data collection, POI coverage in Jakarta has reached 97.95 POI / square kilometer, better Q3~Q4 than Google maps. • Continue to carry out distributed business expansion based on map; obtain orders from Honda, Mazda, Toyota and other car manufacturers that continuously generate revenue. • DApp integrate decentralize wallet. • Cooperate with Indonesian and Southeast Asian unicorn enterprises, go-jek, and Akulaku, on map services. • The core technology of full data on chain completes POC Proof of Concept . 2019 • The initial display map covers Southeast Asia Q1~Q2 • AI hardware spatial data mining machine project approval. • The first phase of GoZone Spatial data node was launched. 2019 • Open governance of spatial data node. Q3~Q4 • The first version of the AI hardware spatial data mining machine was released. • Continue to expand map-based distributed business partnerships and scale up GoWithMi's services for distributed businesses. • Map display covers South America and Africa. • Indonesia offers a full range of map services including map display, POI search, route planning and route guide. • The map dAPI start services. 2020 • Main chain online. • Second generation of HD map AI hardware spatial data mining machine released. • GoZone covers Southeast Asia. • Full map service complete global coverage. 37 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

The Status of Current Products The project is progressing smoothly according to the roadmap and is currently in the stage of “2019 Q3~Q4”. The technical details are as follows: ⚫ Main Chain: The core design of Gaia’s main chain has been audited, and no technical risks has been detected. The main chain will be launched next year on the main network. GeoMesh technology prototype has passed POC test. One single GeoMesh data can be controlled at around 800K and able to provide map display and path planning services within the range of 3.6*2.4 square kilometers; High-level road network self-adaptive compilation enters the performance test phase. At present, it only takes 68 seconds to generate a strongly connected high-level road network in Southeast Asia’s 1,888,583 km road network. On chain standard map service has completed 80%. Point of Interest (POI) search module 1.0 has completed, which can support the POI search of 527,575 place names in Indonesia (the data size is 10~30 times that of OpenStreetMap). Path planning and guiding module 1.0 has completed and is being tested within a total area of 47,266 square kilometers in Jakarta, Indonesia and adjacent regions. The path planning and guiding module 1.1 have completed 80%; together with the “high-level road network automatic compilation function”, it can provide mainland-level cross-country path calculation. The first phase of indoor map technology has completed with a data coverage of 922,303 square meters. ⚫ Main dApp: dAPP has released 16 Android versions and 6 IOS versions iteratively with 1,200,000 downloads and 22,000 daily active users. Both Android version and IOS version have built-in decentralized wallet ⚫ Spatial data mining machine: The initial technical development of AI spatial data mining machine has been completed and the details will be announced recently. ⚫ Application achievements: 38 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

We have gained revenue from car manufacturers and cooperated with many well-known fast moving consumer goods manufacturers in Indonesia to pilot a larger scale distributed business operation in hundreds of restaurants and thousands of supermarkets. Figure 12. A display of the current products Prospects Ledger of Everything and Space Oracle Map service is required when the smart contract is involved in the real world scenario, such as decentralized cab-hailing, decentralized take-out and decentralized e-commerce. If dAPI need to solve issues of using smart contract to mobilize the people and things in the real world like calculating driving route, dispatching the delivery man to fetch the take-out, real-time display of the cargo’s location, etc., we need space oracles to solve the problem of letting smart contracts perceive whether people and things effectively complete services. The main reason why the space oracle cannot be replaced by the centralized map is related to the blockchain smart contract’s “self-declaration, verification by others” design. After a node performs the smart contract, it still needs verification from other nodes. This means the space oracle needs to give the verification node verifiable and tamper-resistant historical spatial data. Obviously, that’s impossible for centralized map which emphasizes timeliness and provides context-free services. 39 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

As a result, decentralized “financial ledger” like blockchain needs to use decentralized “Ledger of Everything” like GoCenterMi to drive everything. As Gaia’s technology system and network architecture are the same as the blockchain, the production of GeoMesh can be considered as a block generation behavior in essence, which supports the complete traceability of its historical chain. All public chains need technology like Gaia to ensure the purity of performing smart contracts related to the real world in a decentralized way. Worldwide Decentralized Location Service Facilities and “Prisoner's Dilemma” Prisoner’s dilemma exists in the map industry globally—any large Internet company doesn’t want its own map, a HUB that connects its own real-world information and digital information, on competitors’ servers. As a distributed and completely neutral location service infrastructure, Gaia will free large Internet companies from the meaningless competition of “reinventing the wheel” other than providing a safe, low-cost solution for small and medium-sized developers,. Privacy Map and Interest Map With GoWithMi providing a decentralized universal underlying map solution, segmented map application scenario becomes possible, which otherwise could not be realized due to the high sunk cost. ⚫ Privacy Map Just like the browsing tracks on the Internet, people’s tracks in space sometimes involve privacy issue. Gaia’s distributed map makes privacy map possible. Map service that based on open source, verifiable and distributed nodes completely eliminates the possibility of user information being hijacked by the bottom layer. And for the upper layer, MPC (multi-party computing) or zero-knowledge proof technology allow users to use the full map service without revealing their spatial location. It is worth emphasizing that the closed source code of centralized map and the “black box operation” of the central server deployment make it impossible for centralized map to prove its protection of user privacy. For centralized map, claiming to realize privacy map is an ostrich policy. ⚫ Community of Interest Map 40 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

The current map is more concerned with general-purpose functions because of data cost, but this does not mean that there is no market demand for maps that provide data and services for niche interests. Based on GoWithMi’s data collection system, anyone can build maps according to their own needs. Users can form sub-map communities like “Sea-fishing Map”, “Cycling Map”, “Scenic Map” and “Agricultural Map” to build a more in-depth community by themselves. The Future of Distributed Autonomous Driving Tesla’s crowdsourcing collection of autopilot data [21] reveals the trend of autonomous driving. Autonomous driving’s requirements for data richness and real- time performance have gradually surpassed the limit of traditional centralized data production. While the improvement of 5G communication bandwidth blurs the difference between the edge and the center, AI blurs the difference between professional and non-professional data collection. Token-driven crowdsourcing data collection will be an important data source for autonomous driving in the future. GoWithMi’s design of the entire map ecology is in line with this trend. 41 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved

References [1] According to Report on the Traffic of the Major Cities in China 2016 released by Amap, Cadillac owners like going to bathing centers the best. URL: https://wenku.baidu.com/view/36f751530a4e767f5acfa1c7aa00b52acec79c51.html [2] <Google Maps API Price Hike Is Threatening the Future of Some Companies> https://gadgets.ndtv.com/apps/features/google-maps- apis-new-pricing-impact-1907242 [3] <YouTube, Google Maps and more: The major Android apps Huawei users could lose>, URL: https://www.trustedreviews.com/news/youtube-google-maps-apps-huawei-users-miss-3757529 [4] Baidu Maps Open Platform http://lbsyun.baidu.com/ [6] [email protected] https://en.wikipedia.org/wiki/[email protected] [7] Grid Computing https://en.wikipedia.org/wiki/Grid_computing [8] Root name server https://en.wikipedia.org/wiki/Root_name_server [9] Domain Name System https://en.wikipedia.org/wiki/Domain_Name_System [10] top-level domain https://en.wikipedia.org/wiki/Top-level_domain [11] Quadtree: https://en.wikipedia.org/wiki/Quadtree [12] GeoHash: https://en.wikipedia.org/wiki/Geohash [13] Web Mercator projection:https://en.wikipedia.org/wiki/Web_Mercator_projection [14] Solr https://en.wikipedia.org/wiki/Apache_Solr [15] ElasticSearch https://en.wikipedia.org/wiki/Elasticsearch [16] Strongly connected graph: https://en.wikipedia.org/wiki/Strongly_connected_component [17] Dijkstra's algorithm: https://en.wikipedia.org/wiki/Dijkstra%27s_algorithm [18] A* algorithm: https://en.wikipedia.org/wiki/A*_search_algorithm [19] Tarjan's algorithm:https://en.wikipedia.org/wiki/Tarjan%27s_strongly_connected_components_algorithm [20] Contraction hierarchies:https://en.wikipedia.org/wiki/Contraction_hierarchies [21] Your Tesla Is Watching–and Recording–You All the Time:https://www.lamag.com/citythinkblog/tesla-recording-data-privacy/ og/tesla-recording-data-privacy/ 42 / 42 GoWithMi White Paper V3.0.0 All Rights Reserved