Mappix

mappix

Imagine you are walking in a very thick forest without a guide map. Every step you take feels like a lucky guess, and there are high chances that you might lose your way. Now, think of a product the user will use, but without the user's experience, it is the same as the forest with no guide map. User mapping has the role of the guide map in the forest, and thus, the designers can easily walk the users through their journey using user mapping. What is User Mapping? 🤔 It is a methodology of visualizing how a person uses a product or service. It helps the designers understand who are the users, what they are doing, and how they feel at each step. Touchpoints are instances of interaction with your product by the users. App registration or button selection is one of the examples. Pain Points are the things that users complain about. For example, it may take too long to load a page, or if a button is hard to find. Why is User Mapping Important? 🌟 Think of it like you are going hiking with a guide map that will guide you about the wrong paths and the faster route. Without it, you are wasting your time and feeling frustrated. The user map does the same; it helps designers to find the problems earlier before they become major issues. For instance, a food delivery app design team during the user mapping process noticed that the user had trouble finding the checkout button. This helped them to quickly identify the pain point and make the necessary changes. Types of User Mapping 🧩 There are different types of User Maps, each with their own goals. It works just like having different maps for different places. Journey Mapping: It is a road map that shows the entire process that one will go through in order to complete a certain task.
Example: A pizza ordering example found out the following: searching the menu, choosing a pizza, adding it to a cart, and getting the purchase completed. Empathy Mapping: It is more about what users think, how they feel, and what they would want at various times of the journey.
For example, people who are using the fitness app may experience irritation or confusion during the first phase of using it. Experience Mapping: It covers everything that includes how users relate to the brand.
Example: A journey of a consumer from the moment of seeing the ad to buying the product online. It shows the bigger picture including the link between the users and the brand. In simple words: The shopper's path from seeing an advertisement to purchasing a product online. Service Blueprinting: It focuses on the internal mechanisms that create the user experience.
Example: The way a delivery company organizes its kitchen, customer care, and drivers to ensure the process keeps running smoothly. Principles of Effective User Mapping 🌟 Empathy First: Think about how users are perceiving and feeling what the situation is about, at every moment. Visualize Simply: Your map should have the same readability and usability as a straight road map. Work as a Team: It is very important to hear from people all over the business, developers, customer support, and designers are all looking at the same thing but from different angles. Focus on Action: Your map should provide every insight that leads you to a clear step to improve the experience. Iterate: User needs are subject to change with time, therefore, constantly update your maps according to feedback. How to Create a User Map 📝 Let's say you are developing an application for people to order their tasks by the day easily. The next step is what user mapping could give your work: Start with Research: Communicate with users, and watch what they are doing. Define Personas: Build up the features of the different users. As an example, we can say that Tom, a busy professional, has to master meetings and deadlines. Choose Your Map: Depending on what you want to know, decide whether you need a journey map, empathy map, or experience map. Map the Journey: Break down each step of the way. In the case of Tom, it’s about opening the app, adding tasks, and checking them off. Refine: Continue developing your map based on the new data and the user feedback you get, as user needs change over time. Therefore, it will be important to monitor and incorporate the feedback you get to make sure you still have the right and updated map. A Day-to-Day Example: Navigating a Coffee Shop App ☕ Consider a typical coffee shop app in which you can pay for your drinks in advance to avoid the hassle. One of the important items that might be discovered after Journey Mapping is the situation when a user leaves the cart because the advertisements are all in the targeted ads till the end. In the case of Empathy Mapping, the developers might denote that users are angry when the app slows down. After the team finds out about these problems, they can solve them by introducing the total cost sooner and the app's speed faster. And that is more effective than making users happy in the long run. How Atomic Design Helps the Mapping Process 🚀 Rapid user mapping and simplified templates can be accomplished purely by atomic design. Let’s say that you already have a library of buttons, navigation bars, and icons that you can bring in from it. The strategy of reusability allows for the use of the same components which you can tune if you need to fulfil the desired outcome and it will save your time accordingly without a big shift. For instance, you can tell a user has trouble with checking out simply because they can't find the representative button. You can in this situation modify one single button for the whole product which will save you time and the layout will be consistent. Final Thoughts: A Smooth Journey Ahead 🌟 User mapping is not just a task but its a complete set of strategy. It’s about the user's path, finding their pain points, and making the user experience as smooth as possible. As you need a map to guide you through the forest, you need a user map to visualize and design your project. The user map is a skill that will help you connect with your users on a deeper level, solving their problems and making the interface of the design more interactive and engaging. Therefore, take your map and start designing the way to achieve success. 🌍 Imagine you are walking in a very thick forest without a guide map. Every step you take feels like a lucky guess, and there are high chances that you might lose your way. Now, think of a product the user will use, but without the user's experience, it is the same as the forest with no guide map. User mapping has the role of the guide map in the forest, and thus, the designers can easily walk the users through their journey using user mapping. What is User Mapping? 🤔 It is a methodology of visualizing how a person uses a product or service. It helps the designers understand who are the users, what they are doing, and how they feel at each step. Touchpoints are instances of interaction with your product by the users. App registration or button selection is one of the examples. Touchpoints are instances of interaction with your product by the users. App registration or button selection is one of the examples. Pain Points are the things that users complain about. For example, it may take too long to load a page, or if a button is hard to find. Pain Points are the things that users complain about. For example, it may take too long to load a page, or if a button is hard to find. Why is User Mapping Important? 🌟 Why is User Mapping Important? 🌟 Think of it like you are going hiking with a guide map that will guide you about the wrong paths and the faster route. Without it, you are wasting your time and feeling frustrated. The user map does the same; it helps designers to find the problems earlier before they become major issues. For instance, a food delivery app design team during the user mapping process noticed that the user had trouble finding the checkout button. This helped them to quickly identify the pain point and make the necessary changes. Types of User Mapping 🧩 There are different types of User Maps, each with their own goals. It works just like having different maps for different places. Journey Mapping: It is a road map that shows the entire process that one will go through in order to complete a certain task.
Example: A pizza ordering example found out the following: searching the menu, choosing a pizza, adding it to a cart, and getting the purchase completed. Journey Mapping: It is a road map that shows the entire process that one will go through in order to complete a certain task. Example: A pizza ordering example found out the following: searching the menu, choosing a pizza, adding it to a cart, and getting the purchase completed. Journey Mapping: It is a road map that shows the entire process that one will go through in order to complete a certain task. Journey Mapping: Example: A pizza ordering example found out the following: searching the menu, choosing a pizza, adding it to a cart, and getting the purchase completed. Example: A pizza ordering example found out the following: searching the menu, choosing a pizza, adding it to a cart, and getting the purchase completed. Empathy Mapping: It is more about what users think, how they feel, and what they would want at various times of the journey.
For example, people who are using the fitness app may experience irritation or confusion during the first phase of using it. Empathy Mapping: It is more about what users think, how they feel, and what they would want at various times of the journey. For example, people who are using the fitness app may experience irritation or confusion during the first phase of using it. Empathy Mapping: It is more about what users think, how they feel, and what they would want at various times of the journey. Empathy Mapping: For example, people who are using the fitness app may experience irritation or confusion during the first phase of using it. For example, people who are using the fitness app may experience irritation or confusion during the first phase of using it. Experience Mapping: It covers everything that includes how users relate to the brand.
Example: A journey of a consumer from the moment of seeing the ad to buying the product online. It shows the bigger picture including the link between the users and the brand. In simple words: The shopper's path from seeing an advertisement to purchasing a product online. Experience Mapping: It covers everything that includes how users relate to the brand. Example: A journey of a consumer from the moment of seeing the ad to buying the product online. It shows the bigger picture including the link between the users and the brand. In simple words: The shopper's path from seeing an advertisement to purchasing a product online. Experience Mapping : It covers everything that includes how users relate to the brand. Experience Mapping Example: A journey of a consumer from the moment of seeing the ad to buying the product online. It shows the bigger picture including the link between the users and the brand. In simple words: The shopper's path from seeing an advertisement to purchasing a product online. Example: A journey of a consumer from the moment of seeing the ad to buying the product online. It shows the bigger picture including the link between the users and the brand. In simple words: The shopper's path from seeing an advertisement to purchasing a product online. Service Blueprinting: It focuses on the internal mechanisms that create the user experience.
Example: The way a delivery company organizes its kitchen, customer care, and drivers to ensure the process keeps running smoothly. Service Blueprinting: It focuses on the internal mechanisms that create the user experience. Example: The way a delivery company organizes its kitchen, customer care, and drivers to ensure the process keeps running smoothly. Service Blueprinting: It focuses on the internal mechanisms that create the user experience. Service Blueprinting: Example: The way a delivery company organizes its kitchen, customer care, and drivers to ensure the process keeps running smoothly. Example: The way a delivery company organizes its kitchen, customer care, and drivers to ensure the process keeps running smoothly. Principles of Effective User Mapping 🌟 Empathy First: Think about how users are perceiving and feeling what the situation is about, at every moment. Empathy First: Think about how users are perceiving and feeling what the situation is about, at every moment. Empathy First: Visualize Simply: Your map should have the same readability and usability as a straight road map. Visualize Simply: Your map should have the same readability and usability as a straight road map. Visualize Simply: Work as a Team: It is very important to hear from people all over the business, developers, customer support, and designers are all looking at the same thing but from different angles. Work as a Team: It is very important to hear from people all over the business, developers, customer support, and designers are all looking at the same thing but from different angles. Work as a Team: Focus on Action: Your map should provide every insight that leads you to a clear step to improve the experience. Focus on Action: Your map should provide every insight that leads you to a clear step to improve the experience. Focus on Action: Iterate: User needs are subject to change with time, therefore, constantly update your maps according to feedback. Iterate: User needs are subject to change with time, therefore, constantly update your maps according to feedback. Iterate: How to Create a User Map 📝 Let's say you are developing an application for people to order their tasks by the day easily. The next step is what user mapping could give your work: Start with Research: Communicate with users, and watch what they are doing. Start with Research: Communicate with users, and watch what they are doing. Start with Research: Define Personas: Build up the features of the different users. As an example, we can say that Tom, a busy professional, has to master meetings and deadlines. Define Personas: Build up the features of the different users. As an example, we can say that Tom, a busy professional, has to master meetings and deadlines. Define Personas: Choose Your Map: Depending on what you want to know, decide whether you need a journey map, empathy map, or experience map. Choose Your Map: Depending on what you want to know, decide whether you need a journey map, empathy map, or experience map. Choose Your Map: Map the Journey: Break down each step of the way. In the case of Tom, it’s about opening the app, adding tasks, and checking them off. Map the Journey: Break down each step of the way. In the case of Tom, it’s about opening the app, adding tasks, and checking them off. Map the Journey: Refine: Continue developing your map based on the new data and the user feedback you get, as user needs change over time. Therefore, it will be important to monitor and incorporate the feedback you get to make sure you still have the right and updated map. Refine: Continue developing your map based on the new data and the user feedback you get, as user needs change over time. Therefore, it will be important to monitor and incorporate the feedback you get to make sure you still have the right and updated map. Refine: A Day-to-Day Example: Navigating a Coffee Shop App ☕ Consider a typical coffee shop app in which you can pay for your drinks in advance to avoid the hassle. One of the important items that might be discovered after Journey Mapping is the situation when a user leaves the cart because the advertisements are all in the targeted ads till the end. One of the important items that might be discovered after Journey Mapping is the situation when a user leaves the cart because the advertisements are all in the targeted ads till the end. In the case of Empathy Mapping, the developers might denote that users are angry when the app slows down. After the team finds out about these problems, they can solve them by introducing the total cost sooner and the app's speed faster. And that is more effective than making users happy in the long run. In the case of Empathy Mapping, the developers might denote that users are angry when the app slows down. After the team finds out about these problems, they can solve them by introducing the total cost sooner and the app's speed faster. And that is more effective than making users happy in the long run. How Atomic Design Helps the Mapping Process 🚀 Atomic Design Rapid user mapping and simplified templates can be accomplished purely by atomic design. Let’s say that you already have a library of buttons, navigation bars, and icons that you can bring in from it. The strategy of reusability allows for the use of the same components which you can tune if you need to fulfil the desired outcome and it will save your time accordingly without a big shift. For instance, you can tell a user has trouble with checking out simply because they can't find the representative button. You can in this situation modify one single button for the whole product which will save you time and the layout will be consistent. Final Thoughts: A Smooth Journey Ahead 🌟 User mapping is not just a task but its a complete set of strategy. It’s about the user's path, finding their pain points, and making the user experience as smooth as possible. As you need a map to guide you through the forest, you need a user map to visualize and design your project. The user map is a skill that will help you connect with your users on a deeper level, solving their problems and making the interface of the design more interactive and engaging. Therefore, take your map and start designing the way to achieve success. 🌍

Understanding the Importance of User Mapping in UI/UX Design 🗺️

The TechBeat: Mapping CX/UX Research Competencies to Stay Human-Centered in Product Development (10/22/2024)

The TechBeat: Mapping CX/UX Research Competencies to Stay Human-Centered in Product Development (10/20/2024)

The TechBeat: Mapping CX/UX Research Competencies to Stay Human-Centered in Product Development (10/16/2024)

Mapping CX/UX Research Competencies to Stay Human-Centered in Product Development

Intro I absolutely love Paul Graham's style. I can't boast about knowing all of his essays by heart, but there are a few of my favorites, and I'm overwhelmed by the amount of wisdom he manages to put into each paragraph. Recently, I stumbled upon one of his long reads – “How to do great work.” Reminding me of “You and your research” by Richard Hamming, the essay covers a broad set of topics which made my head spin when I tried to grasp it. I read it 10 times and still couldn't feel that the knowledge had settled in my head. Luckily, I discovered the newsletter by Jason Chen where 60,000 characters have been distilled into a simple and powerful mind map. But something was still off – a few key pieces from the original article were missing. The mind map was a step in the right direction, but I needed it to be more bespoke to work for me. So, feeling inspired, I decided to build my own detailed and structured map, which you can find here. In this two-part article, I'm sharing my reflections and favorite pieces from each part of the essay. I initially planned to put everything in a single text, but split it in half instead. Here's the first part: Finding a field and progressing in it and learning and cultivating the Habits of great work. If you liked this article, I highly recommend spending time reading the original slowly and thoughtfully and using the mindmap later as a reference. Topic 1: Finding a Field and Making Progress in It The essay opens with Paul's reflection on what great work is and what the pre-conditions for it are. He shares that it should be something you're naturally good at, ambitious about, and the area itself provides enough room for greatness. So, the next obvious question is how to find this magical combination. 4-Step Process to Achieve Greatness Paul provides a very straightforward process to apply to anything, provided that you're feeling curious enough to bother. It's reiterated a few times, in the first few paragraphs and later, that the best way to start applying this process is to pick an area of curiosity and start making progress. One of the most sobering thoughts is that no one will tell you what to work on – you are on your own, and the main person in charge. The Pros or Regularity and Cons of Procrastination What follows next are reflections on how to make consistent progress in interesting areas because even something you're good at and generally enjoy doing will require hard work and include highs and lows. I was particularly encouraged by the thoughts on the compound nature of any work and the importance of investing regularly, even if it's a small investment of time or money. The idea of making little incremental steps that stack up is trivial but hard to follow in reality. The reason it's so hard to keep up is that any exponential growth, in the beginning, looks like a straight line – you do things, and for a long time, it looks like nothing is moving forward, but the more you do, the steeper the slope and the easier it is to make progress. Paul also warns that since consistent effort is really important, procrastination and gaps in consistency are a threat. He distinguishes two types of procrastination – not progressing on a given day and not progressing in your area of interest in general. The latter is more dangerous, as it's harder to detect and quite often hides behind the mask of real work. I'm trying to build a habit of writing, yet it took me a few months to publish this article because I was busy in other areas of my life, and it always seemed like a good excuse. The Benefits of Active and Passive Thinking The paragraph about taking a break is a good reminder for hard workers who are easily carried away with work and end up exhausting their bodies and brains. Paul points out that your mind is a complicated device, and you can't just kick it 24/7 to produce greatness. Yes, you will make some progress, but you will also burn out and the quality will suffer, so you'll have to take a few steps back to fix the mistakes and recover. Hence, it's quite important to reserve time for recharging and also allow for creative wandering – walk and do nothing to allow your brain to operate in a different mode and make connections. Hacking Yourself Into the Flow State Finally, Paul points out that even if you've picked a field you love and started investing in it consistently, it doesn't mean your level of motivation will remain permanent. So, when you are feeling low, it's okay to “trick” yourself by saying you'll just spend a few minutes working on something. Next thing you know, if the field you're working on is really engaging for you, you will find yourself an hour or two later completely losing track of time and making good progress. Topic 2: Habits of Great Work The next big chunk of the article describes how to probe if the work you are doing is really great, and how to keep the “greatness” bar always high. Some lucky people have these high standards built in, but I think Paul makes it clear what to focus on for building the habits for anyone. Aim to Be the Best, Not ”The Rest” It's easy to be inspired by great visionaries and innovators and tell yourself you're going to only deliver the best, but then life happens – you find yourself tired and slipping into a “good enough” attitude. According to Paul, what matters is exactly the attitude: you shouldn't be discouraged if you miss the standard that you set for yourself, but if you don't aim high again, you won't even do okay in the end. That reminded me of the point Marc Andreessen made on the Silicon Valley phenomenon (from 31:25 in this podcast) – peer pressure pushes people forward, and they can't help but match the high standards with their products. The useful thing to help you in the absence of peer pressure is strictness – your inner voice that often tells you if you did great or not. It's important to stop, listen to this voice, and try to be just a tiny bit better today. The more you listen to it, the easier it gets. Do Not Copy The paragraph on copying is quite interesting because looking up to someone and trying to mimic their behavior seems like a pretty good tactic named “fake it till you make it.” Paul writes a lot more on the danger of blindly copying in the following paragraphs, but here, he lands a very important thought – trying to mimic a style requires effort, yet it's still going to feel fake in the end. So, save your effort and channel it instead into perfecting the work itself. After making the work stand out, the original style and all the cool things will follow naturally. Be Earnest Being earnest is the opposite of being fake, so it's almost the other half of the previous habit. The first key thing for being earnest is to start noticing the moments when you are not quite sincere with yourself, and maybe even assume that there are more moments like this even if you don't notice them. That's what is defined as intellectual honesty – being ready to question yourself and put yourself under scrutiny more often. The second component is to embrace your inner nerdiness – partially because suppressing it and trying to conform will require energy you might otherwise put into making your work great. Re-do and Cut Until Every Part Feels Great The idea of simplicity resonated most with me in this paragraph. I often find myself annoyed by things made more complex than they should be (normally it happens when lots of systems have to integrate together, like in programmable messaging or fintech), and asking if it could be simplified and reduced. Paul's evidence from math and art made me feel better because he also states that being powerful and flexible shouldn't be complicated. Another great thought from this paragraph is consistency – to create a piece of great work, you need to make quite a few incremental steps, and each one of them should be great. So, even if it's tempting to rush through something and keep it in a “whatever” state, the better approach would be to have a few more iterations until it really matches the standards you set for yourself in this whole project. Wrapping Up In this first part of my attempt to map the “How to do great work” article by Paul Graham, I covered two big pieces – How to choose a field and progress in it, and the Habits of great work. In the second part, I will share my findings on Getting ideas, Working on ideas, and the last few paragraphs that I call Life advice. Found this helpful? Drop a comment, and pass it along to anyone you think could use it! Intro I absolutely love Paul Graham's style. I can't boast about knowing all of his essays by heart, but there are a few of my favorites, and I'm overwhelmed by the amount of wisdom he manages to put into each paragraph. Recently, I stumbled upon one of his long reads – “ How to do great work .” Reminding me of “ You and your research ” by Richard Hamming, the essay covers a broad set of topics which made my head spin when I tried to grasp it. I read it 10 times and still couldn't feel that the knowledge had settled in my head. How to do great work You and your research Luckily, I discovered the newsletter by Jason Chen where 60,000 characters have been distilled into a simple and powerful mind map. But something was still off – a few key pieces from the original article were missing. The mind map was a step in the right direction, but I needed it to be more bespoke to work for me. the newsletter So, feeling inspired, I decided to build my own detailed and structured map, which you can find here . In this two-part article, I'm sharing my reflections and favorite pieces from each part of the essay. I initially planned to put everything in a single text, but split it in half instead. Here's the first part: Finding a field and progressing in it and learning and cultivating the Habits of great work . here Finding a field and progressing in it Habits of great work If you liked this article, I highly recommend spending time reading the original slowly and thoughtfully and using the mindmap later as a reference. the mindmap Topic 1: Finding a Field and Making Progress in It The essay opens with Paul's reflection on what great work is and what the pre-conditions for it are. He shares that it should be something you're naturally good at, ambitious about, and the area itself provides enough room for greatness. So, the next obvious question is how to find this magical combination. 4-Step Process to Achieve Greatness Paul provides a very straightforward process to apply to anything, provided that you're feeling curious enough to bother. It's reiterated a few times, in the first few paragraphs and later, that the best way to start applying this process is to pick an area of curiosity and start making progress. One of the most sobering thoughts is that no one will tell you what to work on – you are on your own, and the main person in charge. The Pros or Regularity and Cons of Procrastination What follows next are reflections on how to make consistent progress in interesting areas because even something you're good at and generally enjoy doing will require hard work and include highs and lows. I was particularly encouraged by the thoughts on the compound nature of any work and the importance of investing regularly, even if it's a small investment of time or money. The idea of making little incremental steps that stack up is trivial but hard to follow in reality. The reason it's so hard to keep up is that any exponential growth, in the beginning, looks like a straight line – you do things, and for a long time, it looks like nothing is moving forward, but the more you do, the steeper the slope and the easier it is to make progress. Paul also warns that since consistent effort is really important, procrastination and gaps in consistency are a threat. He distinguishes two types of procrastination – not progressing on a given day and not progressing in your area of interest in general. The latter is more dangerous, as it's harder to detect and quite often hides behind the mask of real work. I'm trying to build a habit of writing, yet it took me a few months to publish this article because I was busy in other areas of my life, and it always seemed like a good excuse. The Benefits of Active and Passive Thinking The paragraph about taking a break is a good reminder for hard workers who are easily carried away with work and end up exhausting their bodies and brains. Paul points out that your mind is a complicated device, and you can't just kick it 24/7 to produce greatness. Yes, you will make some progress, but you will also burn out and the quality will suffer, so you'll have to take a few steps back to fix the mistakes and recover. Hence, it's quite important to reserve time for recharging and also allow for creative wandering – walk and do nothing to allow your brain to operate in a different mode and make connections. Hacking Yourself Into the Flow State Finally, Paul points out that even if you've picked a field you love and started investing in it consistently, it doesn't mean your level of motivation will remain permanent. So, when you are feeling low, it's okay to “trick” yourself by saying you'll just spend a few minutes working on something. Next thing you know, if the field you're working on is really engaging for you, you will find yourself an hour or two later completely losing track of time and making good progress. Topic 2: Habits of Great Work The next big chunk of the article describes how to probe if the work you are doing is really great, and how to keep the “greatness” bar always high. Some lucky people have these high standards built in, but I think Paul makes it clear what to focus on for building the habits for anyone. Aim to Be the Best, Not ”The Rest” It's easy to be inspired by great visionaries and innovators and tell yourself you're going to only deliver the best, but then life happens – you find yourself tired and slipping into a “good enough” attitude. According to Paul, what matters is exactly the attitude: you shouldn't be discouraged if you miss the standard that you set for yourself, but if you don't aim high again, you won't even do okay in the end. That reminded me of the point Marc Andreessen made on the Silicon Valley phenomenon (from 31:25 in this podcast ) – peer pressure pushes people forward, and they can't help but match the high standards with their products. in this podcast The useful thing to help you in the absence of peer pressure is strictness – your inner voice that often tells you if you did great or not. It's important to stop, listen to this voice, and try to be just a tiny bit better today. The more you listen to it, the easier it gets. Do Not Copy The paragraph on copying is quite interesting because looking up to someone and trying to mimic their behavior seems like a pretty good tactic named “fake it till you make it.” Paul writes a lot more on the danger of blindly copying in the following paragraphs, but here, he lands a very important thought – trying to mimic a style requires effort, yet it's still going to feel fake in the end. So, save your effort and channel it instead into perfecting the work itself. After making the work stand out, the original style and all the cool things will follow naturally. Be Earnest Being earnest is the opposite of being fake, so it's almost the other half of the previous habit. The first key thing for being earnest is to start noticing the moments when you are not quite sincere with yourself, and maybe even assume that there are more moments like this even if you don't notice them. That's what is defined as intellectual honesty – being ready to question yourself and put yourself under scrutiny more often. The second component is to embrace your inner nerdiness – partially because suppressing it and trying to conform will require energy you might otherwise put into making your work great. Re-do and Cut Until Every Part Feels Great The idea of simplicity resonated most with me in this paragraph. I often find myself annoyed by things made more complex than they should be (normally it happens when lots of systems have to integrate together, like in programmable messaging or fintech), and asking if it could be simplified and reduced. Paul's evidence from math and art made me feel better because he also states that being powerful and flexible shouldn't be complicated. Another great thought from this paragraph is consistency – to create a piece of great work, you need to make quite a few incremental steps, and each one of them should be great. So, even if it's tempting to rush through something and keep it in a “whatever” state, the better approach would be to have a few more iterations until it really matches the standards you set for yourself in this whole project. Wrapping Up In this first part of my attempt to map the “How to do great work” article by Paul Graham, I covered two big pieces – How to choose a field and progress in it , and the Habits of great work . How to choose a field and progress in it Habits of great work In the second part, I will share my findings on Getting ideas , Working on ideas , and the last few paragraphs that I call Life advice . Found this helpful? Drop a comment, and pass it along to anyone you think could use it! Getting ideas Working on ideas Life advice

Mind Mapping “How to Do Great Work”: My Take on Paul Graham's Essay (Part One)

Карта розуму “Як рабіць вялікую працу”: мой погляд на эсэ Пола Грэма (частка першая)

Мисловна карта „Как да вършим страхотна работа“: Моят поглед към есето на Пол Греъм (първа част)

Mapas mentales “Cómo hacer un gran trabajo”: mi opinión sobre el ensayo de Paul Graham (primera parte)

Mind Mapping "Nola egin lan bikaina": Paul Graham-en saiakera nire iritzia (lehen zatia)

Pemetaan Pikiran “Cara Melakukan Pekerjaan Hebat”: Pandangan Saya terhadap Esai Paul Graham (Bagian Satu)

マインドマップ「素晴らしい仕事をする方法」: ポール・グラハムのエッセイに対する私の見解 (パート 1)

Domu kartēšana “Kā paveikt lielisku darbu”: mans skatījums uz Pola Grehema eseju (pirmā daļa)

Yuyay Mapeo “Imaynatataq hatun llamkayta ruwachwan”: Paul Grahampa ensayomanta hapisqay (Ñawpaq kaq raki)

แผนที่ความคิด “วิธีทำงานที่ยอดเยี่ยม”: ความคิดเห็นของฉันเกี่ยวกับเรียงความของ Paul Graham (ตอนที่หนึ่ง)

Mind Mapping "Paano Gumawa ng Mahusay na Trabaho": Ang Aking Pananaw sa Sanaysay ni Paul Graham (Unang Bahagi)

Zihin Haritalama “Harika İşler Nasıl Yapılır”: Paul Graham'ın Denemesi Hakkındaki Görüşüm (Birinci Bölüm)

مائنڈ میپنگ "عظیم کام کیسے کریں": پال گراہم کے مضمون پر میرا خیال (حصہ اول)

Authors:
(1) Aviad Rom, The Data Science Institute, Reichman University, Herzliya, Israel;
(2) Kfir Bar, The Data Science Institute, Reichman University, Herzliya, Israel. Table of Links Abstract and Introduction Related Work Methodology Experimental Settings Results Conclusion and Limitations Bibliographical References 5. Results The results are summarized in Table 1. The table rows are organized into three distinct groups. The first group features combinations of various baseline models. The second group showcases combinations incorporating our proprietary HeArBERT model, while the third group highlights combinations involving the contrasting HeArBERTNT model. We train multiple machine-translation combinations, in both directions Arabic-to-Hebrew and Hebrewto-Arabic, based on the same encoder-decoder architecture, initialized with different language model combinations. We assign various combinations of language models to the encoder and decoder components, ensuring that the selected models align with the respective source and target languages. In other words, we make sure that the language model which we use to initialize the encoder, is compatible with the system’s source language. Since mBERT and CAMeLBERTET , as well as our two models HeArBERT and HeArBERTNT , can potentially handle both languages, we experiment with combinations where each of them is assigned to both, the encoder and decoder components at the same time. The results demonstrate that the second group, which utilizes our HeArBERT to initialize a Hebrew decoder, surpasses the performance of all other combinations. Notably, the pairing of GigaBERT with HeArBERT is the standout performer across both directions. It surpasses the performance of using HeArBERT for initializing the Arabic encoder by only a few minor points. The performance of HeArBERTNT , as reported in the third group is consistently lower than HeArBERT in both directions. The difference seems to be more significant in the Arabic-to-Hebrew direction. Overall, the results show that transliterating the Arabic training texts into the Hebrew script as a pre-processing step is beneficial for an Arabic-to-Hebrew machine translation system. The impact of the transliteration proves to be less pronounced in the reverse translation direction. Using the extended (ET) version of CAMeLBERT has a reasonable performance. However, it performs much worse than the best model in both directions, indicating that extending the vocabulary with transliterated Arabic tokens does contribute to better capturing the meaning of Hebrew tokens in context. This implies that joint pre-training on both languages is essential for achieving a more robust language representation. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Aviad Rom, The Data Science Institute, Reichman University, Herzliya, Israel; (2) Kfir Bar, The Data Science Institute, Reichman University, Herzliya, Israel. Authors: Authors: (1) Aviad Rom, The Data Science Institute, Reichman University, Herzliya, Israel; (2) Kfir Bar, The Data Science Institute, Reichman University, Herzliya, Israel. Table of Links Abstract and Introduction Abstract and Introduction Related Work Related Work Methodology Methodology Experimental Settings Experimental Settings Results Results Conclusion and Limitations Conclusion and Limitations Bibliographical References Bibliographical References 5. Results The results are summarized in Table 1. The table rows are organized into three distinct groups. The first group features combinations of various baseline models. The second group showcases combinations incorporating our proprietary HeArBERT model, while the third group highlights combinations involving the contrasting HeArBERTNT model. We train multiple machine-translation combinations, in both directions Arabic-to-Hebrew and Hebrewto-Arabic, based on the same encoder-decoder architecture, initialized with different language model combinations. We assign various combinations of language models to the encoder and decoder components, ensuring that the selected models align with the respective source and target languages. In other words, we make sure that the language model which we use to initialize the encoder, is compatible with the system’s source language. Since mBERT and CAMeLBERTET , as well as our two models HeArBERT and HeArBERTNT , can potentially handle both languages, we experiment with combinations where each of them is assigned to both, the encoder and decoder components at the same time. The results demonstrate that the second group, which utilizes our HeArBERT to initialize a Hebrew decoder, surpasses the performance of all other combinations. Notably, the pairing of GigaBERT with HeArBERT is the standout performer across both directions. It surpasses the performance of using HeArBERT for initializing the Arabic encoder by only a few minor points. The performance of HeArBERTNT , as reported in the third group is consistently lower than HeArBERT in both directions. The difference seems to be more significant in the Arabic-to-Hebrew direction. Overall, the results show that transliterating the Arabic training texts into the Hebrew script as a pre-processing step is beneficial for an Arabic-to-Hebrew machine translation system. The impact of the transliteration proves to be less pronounced in the reverse translation direction. Using the extended (ET) version of CAMeLBERT has a reasonable performance. However, it performs much worse than the best model in both directions, indicating that extending the vocabulary with transliterated Arabic tokens does contribute to better capturing the meaning of Hebrew tokens in context. This implies that joint pre-training on both languages is essential for achieving a more robust language representation. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Training a Bilingual Language Model by Mapping Tokens onto a Shared Character Space: Results

Authors:
(1) Aviad Rom, The Data Science Institute, Reichman University, Herzliya, Israel;
(2) Kfir Bar, The Data Science Institute, Reichman University, Herzliya, Israel. Table of Links Abstract and Introduction Related Work Methodology Experimental Settings Results Conclusion and Limitations Bibliographical References Abstract We train a bilingual Arabic-Hebrew language model using a transliterated version of Arabic texts in Hebrew, to ensure both languages are represented in the same script. Given the morphological, structural similarities, and the extensive number of cognates shared among Arabic and Hebrew, we assess the performance of a language model that employs a unified script for both languages, on machine translation which requires cross-lingual knowledge. The results are promising: our model outperforms a contrasting model which keeps the Arabic texts in the Arabic script, demonstrating the efficacy of the transliteration step. Despite being trained on a dataset approximately 60% smaller than that of other existing language models, our model appears to deliver comparable performance in machine translation across both translation directions. Keywords: bilingual language model, transliteration, Arabic, Hebrew 1. Introduction Pre-trained language models have become essential for state-of-the-art performance in mono and multilingual natural language processing (NLP) tasks. They are pre-trained once and can then be fine-tuned for various downstream NLP tasks. It has been shown that language models generalize better on multilingual tasks when the target languages share structural similarity, possibly due to script similarity (K et al., 2020). Typically, language models are trained on sequences of tokens that often correspond to words and subword components. Arabic and Hebrew are two Semitic languages that share similar morphological structures in the composition of their words, but use distinct scripts for their written forms. The Hebrew script primarily serves Hebrew, but is also employed in various other languages used by the Jewish population. These languages include Yiddish (or “JudeoGerman”), Ladino (or “Judeo-Spanish”), and JudeoArabic, which comprises a cluster of Arabic dialects spoken and written by Jewish communities residing in Arab nations. To some extent, Judeo-Arabic can be perceived as an Arabic variant written in the Hebrew script. Most of the vocabulary in JudeoArabic consists of Arabic words that have been transliterated into the Hebrew script. Words in two languages that share similar meanings, spellings, and pronunciations are known as cognates. Arabic and Hebrew cognates share similar meanings and spellings despite different scripts. The pronunciation of such cognates are not necessarily the same. Numerous lexicons have been created to record these cognates. One of those lexicons[1] lists a total of 915 Hebrew-Arabic spelling 1https://seveleu.com/pages/ equivalents, of which 435 have been identified as authentic cognates, signifying that they possess identical meanings. Analyzing a parallel HebrewArabic corpus, named Kol Zchut[2] using this lexicon, we found instances of those cognates in about 50% of the sentences. The purpose of this work is to take advantage of the potentially high frequency of cognates in Arabic and Hebrew in building a bilingual language model using only one script. Subsequently, the model will be fine-tuned on NLP tasks, such as machine translation, which can benefit from the innate bilingual proficiency to achieve better results. To ensure that cognates are mapped onto a consistent character space, the model uses Arabic texts that are transliterated into the Hebrew script, which mimics the writing system used in Judeo-Arabic. We call this new language model HeArBERT. [3] We test our new model on machine translation, which is considered a downstream task requiring knowledge in two languages, and report on some promising results. In summary, the primary contributions of our work are: (1) we release a new bilingual Arabic-Hebrew language model; and, (2) we show that pre-training a bilingual language model on transliterated texts, as a way for aligning tokens onto the same character space, is beneficial for machine translation. This paper is available on arxiv under CC BY 4.0 DEED license. [1] https://seveleu.com/pages/semiic-syntax-morpho/comparative-sem [2] https://releases.iahlt.org/ [3] https://huggingface.co/aviadrom/HeArBERT Authors: (1) Aviad Rom, The Data Science Institute, Reichman University, Herzliya, Israel; (2) Kfir Bar, The Data Science Institute, Reichman University, Herzliya, Israel. Authors: Authors: (1) Aviad Rom, The Data Science Institute, Reichman University, Herzliya, Israel; (2) Kfir Bar, The Data Science Institute, Reichman University, Herzliya, Israel. Table of Links Abstract and Introduction Abstract and Introduction Related Work Related Work Methodology Methodology Experimental Settings Experimental Settings Results Results Conclusion and Limitations Conclusion and Limitations Bibliographical References Bibliographical References Abstract We train a bilingual Arabic-Hebrew language model using a transliterated version of Arabic texts in Hebrew, to ensure both languages are represented in the same script. Given the morphological, structural similarities, and the extensive number of cognates shared among Arabic and Hebrew, we assess the performance of a language model that employs a unified script for both languages, on machine translation which requires cross-lingual knowledge. The results are promising: our model outperforms a contrasting model which keeps the Arabic texts in the Arabic script, demonstrating the efficacy of the transliteration step. Despite being trained on a dataset approximately 60% smaller than that of other existing language models, our model appears to deliver comparable performance in machine translation across both translation directions. Keywords : bilingual language model, transliteration, Arabic, Hebrew Keywords 1. Introduction Pre-trained language models have become essential for state-of-the-art performance in mono and multilingual natural language processing (NLP) tasks. They are pre-trained once and can then be fine-tuned for various downstream NLP tasks. It has been shown that language models generalize better on multilingual tasks when the target languages share structural similarity, possibly due to script similarity (K et al., 2020). Typically, language models are trained on sequences of tokens that often correspond to words and subword components. Arabic and Hebrew are two Semitic languages that share similar morphological structures in the composition of their words, but use distinct scripts for their written forms. The Hebrew script primarily serves Hebrew, but is also employed in various other languages used by the Jewish population. These languages include Yiddish (or “JudeoGerman”), Ladino (or “Judeo-Spanish”), and JudeoArabic, which comprises a cluster of Arabic dialects spoken and written by Jewish communities residing in Arab nations. To some extent, Judeo-Arabic can be perceived as an Arabic variant written in the Hebrew script. Most of the vocabulary in JudeoArabic consists of Arabic words that have been transliterated into the Hebrew script. Words in two languages that share similar meanings, spellings, and pronunciations are known as cognates. Arabic and Hebrew cognates share similar meanings and spellings despite different scripts. The pronunciation of such cognates are not necessarily the same. Numerous lexicons have been created to record these cognates. One of those lexicons[1] lists a total of 915 Hebrew-Arabic spelling 1https://seveleu.com/pages/ equivalents, of which 435 have been identified as authentic cognates, signifying that they possess identical meanings. Analyzing a parallel HebrewArabic corpus, named Kol Zchut[2] using this lexicon, we found instances of those cognates in about 50% of the sentences. The purpose of this work is to take advantage of the potentially high frequency of cognates in Arabic and Hebrew in building a bilingual language model using only one script. Subsequently, the model will be fine-tuned on NLP tasks, such as machine translation, which can benefit from the innate bilingual proficiency to achieve better results. To ensure that cognates are mapped onto a consistent character space, the model uses Arabic texts that are transliterated into the Hebrew script, which mimics the writing system used in Judeo-Arabic. We call this new language model HeArBERT. [3] We test our new model on machine translation, which is considered a downstream task requiring knowledge in two languages, and report on some promising results. In summary, the primary contributions of our work are: (1) we release a new bilingual Arabic-Hebrew language model; and, (2) we show that pre-training a bilingual language model on transliterated texts, as a way for aligning tokens onto the same character space, is beneficial for machine translation. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv [1] https://seveleu.com/pages/semiic-syntax-morpho/comparative-sem [2] https://releases.iahlt.org/ [3] https://huggingface.co/aviadrom/HeArBERT

Training a Bilingual Language Model by Mapping Tokens onto a Shared Character Space

Mapping IT System Capabilities: The Key to Unlocking Valuable Business Data

Spatial computing represents a transformative approach to how we interact with digital information, fundamentally distinct from traditional computing through its emphasis on spatial awareness. Under the spatial computing paradigm information and contents are contextualized in space and into specific locations. The devices that deliver such content are aware and capable of computing both their global location (coordinates) and the physical 3D structure of the space they are in, coherently blending content with it. There are several components that enable Spatial Computing, but at its core, the system needs to connect 3D experiences to physical locations, be able to precisely understand where the streaming device is in space, and compute in real-time position and occlusions. But why use VPS and not other systems like GPS, Beacons, or QRCodes you may ask? There are three main principles that guided us in such a choice: Censorship resistance
Availability of the re-localization system for AR devices
Precision Let’s go through the different re-localization technologies to see how they compare to VPS: GPS: is a solution to localize in space with low precision, is based on a publicly accessible infrastructure of satellites, is censorship resistant (excluding extreme scenarios), and can be easily accessed with almost any smartphone with on-edge computation. On the downside, it is very imprecise indoors and in locations with tall buildings. Is a good system to get a rough idea of where you are in space, yet it is not precise enough for most of the spatial computing use cases. VPS: is a solution to localize in space with high precision, it is censorship resistant - as long as you can use an electronic device with a camera - since it leverages the visual appearance of locations, it can be performed both on edge and on the cloud. The downside compared to GPS is that it requires preliminary 3D mapping of the target location, an activity that through the OVER mapping tool requires 3-5 minutes for an area of 300 sqm. An important feature that is not shared by any of the other systems is that the data structures that empowers VPS also entails the understanding of the 3D structure of physical locations and thus enables the creation of remotely explorable digital wins of the mapped location. It is the only solution that supports all of the spatial computing use cases.


Magnetic fields: is a solution to localize in space with medium precision, it is censorship resistant since it leverages the magnetic field of the target locations (similarly to what VPS does with the visual field). It leverages on-edge computation yet it requires the device to be equipped with a magnetic compass (not all devices are equipped with it). It requires contextualization/calibration. It is not precise enough for all of the spatial computing use cases.


Wifi: is a solution to localize in space with medium precision, it is censorship resistant since is based on a “public good”, the existing wifi networks of the target location. Is based on on-edge computation leveraging device Wifi capabilities. On the downside, contrary to GPS, Wifi is not globally distributed, so the system will work only if there are existing Wifi hotspots, it requires calibration/contextualization. It is not precise enough for all of the spatial computing use cases and it is very sensitive to updates to the wifi network infrastructures.


Beacons: is a solution to localize in space with medium precision, it is not censorship resistant since it requires permission for the installation of physical devices. Is based on on-edge computation, and it leverages Bluetooth/Wifi hardware of devices. On the downside compared to both GPS and VPS it is costly (cost of the devices + installation) and requires maintenance (power supply) and contextualization/calibration. It is not precise enough for all of the spatial computing use cases.


QR Codes/Visual markers: this is one of the oldest techniques in AR and is a solution to localize in space with high precision. It is not censorship-resistant since it requires permission to install visual markers in the target physical space. Is based on on-edge computation leveraging the camera and SLAM of the device. As for Wifi, Beacons and Magnetic Fields, it requires additional contextualization in space and calibration. It is precise enough for spatial computing use cases yet it lacks spatial awareness. Decentralization is a means to an end that is censorship resistance, it would be illogical to build a decentralized localization system where you need to ask permission to install QR Codes or Beacons. Leveraging Wifi and/or Magnetic Fields could have been a solution but the lack of ubiquity of the first one and the requirement for a magnetic compass of the second one undermines the application scope of such tech. The only logical solution to the problem of building a censorship-resistant global localization system is combining GPS for rough localization and VPS for precise localization and context/3D space awareness. Spatial computing represents a transformative approach to how we interact with digital information, fundamentally distinct from traditional computing through its emphasis on spatial awareness. Under the spatial computing paradigm information and contents are contextualized in space and into specific locations. The devices that deliver such content are aware and capable of computing both their global location (coordinates) and the physical 3D structure of the space they are in, coherently blending content with it. There are several components that enable Spatial Computing, but at its core, the system needs to connect 3D experiences to physical locations, be able to precisely understand where the streaming device is in space, and compute in real-time position and occlusions. But why use VPS and not other systems like GPS, Beacons, or QRCodes you may ask? There are three main principles that guided us in such a choice: Censorship resistance Availability of the re-localization system for AR devices Precision Censorship resistance Availability of the re-localization system for AR devices Precision Let’s go through the different re-localization technologies to see how they compare to VPS: GPS: is a solution to localize in space with low precision, is based on a publicly accessible infrastructure of satellites, is censorship resistant (excluding extreme scenarios), and can be easily accessed with almost any smartphone with on-edge computation. On the downside, it is very imprecise indoors and in locations with tall buildings. Is a good system to get a rough idea of where you are in space, yet it is not precise enough for most of the spatial computing use cases. GPS: is a solution to localize in space with low precision, is based on a publicly accessible infrastructure of satellites, is censorship resistant (excluding extreme scenarios), and can be easily accessed with almost any smartphone with on-edge computation. On the downside, it is very imprecise indoors and in locations with tall buildings. Is a good system to get a rough idea of where you are in space, yet it is not precise enough for most of the spatial computing use cases. VPS: is a solution to localize in space with high precision, it is censorship resistant - as long as you can use an electronic device with a camera - since it leverages the visual appearance of locations, it can be performed both on edge and on the cloud. The downside compared to GPS is that it requires preliminary 3D mapping of the target location, an activity that through the OVER mapping tool requires 3-5 minutes for an area of 300 sqm. An important feature that is not shared by any of the other systems is that the data structures that empowers VPS also entails the understanding of the 3D structure of physical locations and thus enables the creation of remotely explorable digital wins of the mapped location. It is the only solution that supports all of the spatial computing use cases. Magnetic fields: is a solution to localize in space with medium precision, it is censorship resistant since it leverages the magnetic field of the target locations (similarly to what VPS does with the visual field). It leverages on-edge computation yet it requires the device to be equipped with a magnetic compass (not all devices are equipped with it). It requires contextualization/calibration. It is not precise enough for all of the spatial computing use cases. Wifi: is a solution to localize in space with medium precision, it is censorship resistant since is based on a “public good”, the existing wifi networks of the target location. Is based on on-edge computation leveraging device Wifi capabilities. On the downside, contrary to GPS, Wifi is not globally distributed, so the system will work only if there are existing Wifi hotspots, it requires calibration/contextualization. It is not precise enough for all of the spatial computing use cases and it is very sensitive to updates to the wifi network infrastructures. Beacons: is a solution to localize in space with medium precision, it is not censorship resistant since it requires permission for the installation of physical devices. Is based on on-edge computation, and it leverages Bluetooth/Wifi hardware of devices. On the downside compared to both GPS and VPS it is costly (cost of the devices + installation) and requires maintenance (power supply) and contextualization/calibration. It is not precise enough for all of the spatial computing use cases. QR Codes/Visual markers: this is one of the oldest techniques in AR and is a solution to localize in space with high precision. It is not censorship-resistant since it requires permission to install visual markers in the target physical space. Is based on on-edge computation leveraging the camera and SLAM of the device. As for Wifi, Beacons and Magnetic Fields, it requires additional contextualization in space and calibration. It is precise enough for spatial computing use cases yet it lacks spatial awareness. VPS: is a solution to localize in space with high precision, it is censorship resistant - as long as you can use an electronic device with a camera - since it leverages the visual appearance of locations, it can be performed both on edge and on the cloud. The downside compared to GPS is that it requires preliminary 3D mapping of the target location, an activity that through the OVER mapping tool requires 3-5 minutes for an area of 300 sqm. An important feature that is not shared by any of the other systems is that the data structures that empowers VPS also entails the understanding of the 3D structure of physical locations and thus enables the creation of remotely explorable digital wins of the mapped location. It is the only solution that supports all of the spatial computing use cases. VPS: is a solution to localize in space with high precision, it is censorship resistant - as long as you can use an electronic device with a camera - since it leverages the visual appearance of locations, it can be performed both on edge and on the cloud. The downside compared to GPS is that it requires preliminary 3D mapping of the target location, an activity that through the OVER mapping tool requires 3-5 minutes for an area of 300 sqm. An important feature that is not shared by any of the other systems is that the data structures that empowers VPS also entails the understanding of the 3D structure of physical locations and thus enables the creation of remotely explorable digital wins of the mapped location. It is the only solution that supports all of the spatial computing use cases. Magnetic fields: is a solution to localize in space with medium precision, it is censorship resistant since it leverages the magnetic field of the target locations (similarly to what VPS does with the visual field). It leverages on-edge computation yet it requires the device to be equipped with a magnetic compass (not all devices are equipped with it). It requires contextualization/calibration. It is not precise enough for all of the spatial computing use cases. Magnetic fields: is a solution to localize in space with medium precision, it is censorship resistant since it leverages the magnetic field of the target locations (similarly to what VPS does with the visual field). It leverages on-edge computation yet it requires the device to be equipped with a magnetic compass (not all devices are equipped with it). It requires contextualization/calibration. It is not precise enough for all of the spatial computing use cases. Wifi: is a solution to localize in space with medium precision, it is censorship resistant since is based on a “public good”, the existing wifi networks of the target location. Is based on on-edge computation leveraging device Wifi capabilities. On the downside, contrary to GPS, Wifi is not globally distributed, so the system will work only if there are existing Wifi hotspots, it requires calibration/contextualization. It is not precise enough for all of the spatial computing use cases and it is very sensitive to updates to the wifi network infrastructures. Wifi: is a solution to localize in space with medium precision, it is censorship resistant since is based on a “public good”, the existing wifi networks of the target location. Is based on on-edge computation leveraging device Wifi capabilities. On the downside, contrary to GPS, Wifi is not globally distributed, so the system will work only if there are existing Wifi hotspots, it requires calibration/contextualization. It is not precise enough for all of the spatial computing use cases and it is very sensitive to updates to the wifi network infrastructures. Beacons: is a solution to localize in space with medium precision, it is not censorship resistant since it requires permission for the installation of physical devices. Is based on on-edge computation, and it leverages Bluetooth/Wifi hardware of devices. On the downside compared to both GPS and VPS it is costly (cost of the devices + installation) and requires maintenance (power supply) and contextualization/calibration. It is not precise enough for all of the spatial computing use cases. Beacons: is a solution to localize in space with medium precision, it is not censorship resistant since it requires permission for the installation of physical devices. Is based on on-edge computation, and it leverages Bluetooth/Wifi hardware of devices. On the downside compared to both GPS and VPS it is costly (cost of the devices + installation) and requires maintenance (power supply) and contextualization/calibration. It is not precise enough for all of the spatial computing use cases. QR Codes/Visual markers: this is one of the oldest techniques in AR and is a solution to localize in space with high precision. It is not censorship-resistant since it requires permission to install visual markers in the target physical space. Is based on on-edge computation leveraging the camera and SLAM of the device. As for Wifi, Beacons and Magnetic Fields, it requires additional contextualization in space and calibration. It is precise enough for spatial computing use cases yet it lacks spatial awareness. QR Codes/Visual markers: this is one of the oldest techniques in AR and is a solution to localize in space with high precision. It is not censorship-resistant since it requires permission to install visual markers in the target physical space. Is based on on-edge computation leveraging the camera and SLAM of the device. As for Wifi, Beacons and Magnetic Fields, it requires additional contextualization in space and calibration. It is precise enough for spatial computing use cases yet it lacks spatial awareness. Decentralization is a means to an end that is censorship resistance, it would be illogical to build a decentralized localization system where you need to ask permission to install QR Codes or Beacons. Leveraging Wifi and/or Magnetic Fields could have been a solution but the lack of ubiquity of the first one and the requirement for a magnetic compass of the second one undermines the application scope of such tech. The only logical solution to the problem of building a censorship-resistant global localization system is combining GPS for rough localization and VPS for precise localization and context/3D space awareness.

The Essential Role of VPS and 3D Mapping in Spatial Computing for Accurate Re-localization

Stories, News & Mentions about MAPPIX

Mappix’s Stories on HackerNoon