🇲🇰 WhatTheStack Conference 2025

Thinking Outside the Box: Alternative Architectures for Modern Web Applications

In a world dominated by traditional architectures, developers often overlook other patterns that could better serve their apps. This talk explores lessons we can learn from different parts of the globe and various industries, via two approaches that challenge our traditional thinking about web apps.

First, we'll explore the Mini App architecture, a paradigm that emerged from China's unique ecosystem. We'll examine how these—operating within super app environments like WeChat—represent a radical shift from traditional web apps. We'll discuss their dual-thread architecture and lifecycle management, and how this approach might achieve performance gains.

Then, we'll question what would happen if the DOM and other native view systems weren't based on a tree abstraction. For that, we'll discuss the Entity Component System (ECS) architecture, traditionally associated with game development, and see how its data-oriented design can solve complex state management challenges and improve performance in ways traditional component hierarchies cannot.

Transcript

1. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS • SEPTEMBER 20, 2025. здраво, Skopje! 👋 🇲🇰
2. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS MATHEUS ALBUQUERQUE • @ythecombinator
3. Matheus Albuquerque ↝ 👨💻 STAFF SWE @ MEDALLIA ↝ ⚡ GOOGLE DEVELOPER EXPERT ↝ 𝕏 YTHECOMBINATOR
4. Preamble THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
5. In my humble opinion, we have a lot to learn from… other tools. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
6. 2021 2014 LEARNING FROM OTHER TOOLS…
7. LEARNING FROM OTHER TOOLS… 2021 2005
8. In my humble opinion, we have a lot to learn from… the past. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
9. LEARNING FROM THE PAST…
10. 2023 1993 LEARNING FROM THE PAST…
11. In my humble opinion, we have a lot to learn from… other industries. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
12. LEARNING FROM OTHER INDUSTRIES… 2018 2014
13. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
14. In my humble opinion, we have a lot to learn from… other countries. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
15. LEARNING FROM OTHER COUNTRIES…
16. LEARNING FROM OTHER COUNTRIES…
17. LEARNING FROM OTHER COUNTRIES…
18. ↝ MY HISTORY WITH THAT ↝ OVERVIEW ↝ ARCHITECTURE ↝ ECOSYSTEM ↝ WHY DOES IT MATTER? THIS SESSION PRESENTS… MINI & SUPER APPS ECS ↝ MY HISTORY WITH THAT ↝ OVERVIEW ↝ ARCHITECTURE ↝ ECOSYSTEM ↝ WHY DOES IT MATTER?
19. Mini Apps & Super Apps THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
20. MY HISTORY WITH THAT… THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
21. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
22. ME BACK IN 2022…
23. OVERVIEW THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
24. SUPER APPS? ↝ HOSTS TO OTHER APPS THAT RUN WITHIN THEM. ↝ THEY'VE TAKEN THE CHINESE MARKET BY STORM. ↝ WECHAT (微信) BY TENCENT, ALIPAY ( 支 付宝) BY ANT GROUP, BAIDU (百度), BYTEDANCE'S DOUYIN (抖 音 ) AND TIKTOK (蒂克托克).
25. SUPER APPS! 900 MILLION+ 661 MILLION+ Monthly active users engage with Mini Programs on WeChat. Monthly active users engage with Mini Programs on Alipay. STATISTA, NUMBER OF MONTHLY ACTIVE USERS OF LEADING MINI- PROGRAM PLATFORMS IN CHINA IN DECEMBER 2024 STATISTA, NUMBER OF MONTHLY ACTIVE USERS OF ALIPAY MINI PROGRAM IN CHINA IN SELECTED MONTHS FROM SEPTEMBER 2020 TO MAY 2024
26. MINI APPS?
27. ARCHITECTURE THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
28. ARCHITECTURE: TWO THREADS ↝ VIEW LAYER: ONLY DEALS WITH RENDERING AND LISTENS TO EVENTS. ↝ LOGIC LAYER: IT PROCESSES DATA, SENDS IT TO THE VIEW LAYER, AND RECEIVES FEEDBACK FROM IT. ↝ AS OPPOSED TO A REGULAR WEB APP, A SCRIPT CAN'T CAUSE THE PAGE TO BECOME UNRESPONSIVE.
29. ARCHITECTURE: TWO THREADS VIEW LOGIC WEBVIEW • NATIVE RENDERING JAVASCRIPT ENGINE NATIVE SERVER EVENT ↓ DATA ↑ DATA ↓ EVENT ↑ ⁷ HTTPS • WEBSOCKET
30. ARCHITECTURE: TWO THREADS VIEW LOGIC iOS: WKWebView ANDROID: DIFFERENT PROPRIETARY ENGINES (E.G., TENCENT’S X5 OR BAIDU'S T7) DEVTOOLS: CHROMIUM iOS: JavaScriptCore ANDROID: V8 DEVTOOLS: NW.js OR ELECTRON
31. ARCHITECTURE: EXECUTION
32. VIEW THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
33. TEMPLATING ↝ THERE IS USUALLY A MARKUP LANGUAGE WITH A TEMPLATE MECHANISM SIMILAR TO HTML. ↝ THE MINIAPP RUNTIME TRANSLATES THE MARKUP LANGUAGE INTO HTML OR NATIVE CODE. ↝ EACH PAGE IS RENDERED IN A DIFFERENT WEBVIEW BUT ALL WEBVIEWS SHARE THE IMAGE CACHE.
34. TEMPLATING
35. TEMPLATING: PRIOR ART WXML VUE.JS < ! - - wxml - - > <view wx:if="{{view == 'one'}}">One</view> <view wx:elif="{{view == 'two'}}">Two</view> <view wx:else="{{view == 'three'}}">Three</view> / / page.js Page({ data: { view: "three", }, });
36. TEMPLATING: PRIOR ART 2014 2009
37. TEMPLATING: PRIOR ART 2009 1998
38. TEMPLATING: PRIOR ART
39. COMPONENTS: BUILT-IN
40. COMPONENTS: CUSTOM
41. LOGIC THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
42. SCRIPTING
43. SCRIPTING ↝ A SUBSET OF JS NOT TO BE CONFUSED WITH IT. ↝ NO SUPPORT FOR: eval(), new Function(), document, window, STANDARD ES MODULES, AND OTHERS. ↝ SOME HOSTS PROVIDE BUILT-IN POLYFILLS TO MITIGATE CERTAIN INTEROPERABILITY ISSUES.
44. SCRIPTING: APIs
45. SCRIPTING: APIs?
46. STYLING THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
47. STYLING ↝ DIALECTS OF CSS THAT EXTEND IT WITH RESPONSIVE PIXELS. ↝ STYLES CAN BE: ↝ IMPORTED WITH AN @IMPORT RULE. ↝ DECLARED INLINE, WITH INTERPOLATION.
48. PERFORMANCE THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
49. SNAPSHOTS ↝ A PRE-RENDERING PROCESS ON SOME PAGES IN THE CLOUD TO GENERATE AN INTERMEDIATE, COMPRESSED FORMAT. ↝ THE MINIAPP ENGINE LOADS, PARSES AND RENDERS THE SNAPSHOT. ↝ E.G.: 50K JS SOURCE ⇢ 3K SNAPSHOT.
50. VIRTUAL DOM ↝ THE PAGE RENDERING OF MINIAPPS OFTEN USES VDOM. ↝ SOME MINIAPP ENGINES USE WEBASSEMBLY TO IMPLEMENT THE CORE LOGIC OF VDOM AND MODIFY THE JS ENGINE TO ENHANCE THE PERFORMANCE OF CALLS BETWEEN JS AND WEBASSEMBLY.
51. SKYLINE + GLASS-EASEL: OVERVIEW
52. SKYLINE + GLASS-EASEL: OVERVIEW ↝ WECHAT’S NEWER RENDERING ENGINE. ↝ DESIGNED TO BE FASTER AND MORE EFFICIENT THAN THE TRADITIONAL WEBVIEW-BASED ENGINE. ↝ REPLACES THE LEGACY ARCHITECTURE WITH A MUCH MORE STREAMLINED RENDERING PIPELINE TO ACHIEVE NATIVE-LIKE PERFORMANCE.
53. SKYLINE + GLASS-EASEL: OVERVIEW ↝ A NATIVE RENDERING ENGINE WITH ITS OWN COMPONENT SYSTEM (GLASS-EASEL) INSTEAD OF EMBEDDING A SYSTEM BROWSER. ↝ LAYOUT, STYLE, AND COMPOSITING ARE HANDLED INSIDE THE ENGINE, NOT IN A BROWSER PROCESS. ↝ PAGES SHARE ONE RENDERING ENGINE INSTANCE INSTEAD OF SPINNING UP A NEW WEBVIEW PER PAGE.
54. SKYLINE + GLASS-EASEL: WORKLETS const { runOnUI } = wx.worklet function someWorklet(greeting) { 'worklet' console.log(greeting) } / / JS Thread someWorklet('hello') / / UI Thread runOnUI(someWorklet)('hello') = CODE THAT CAN BE RUN ON EITHER THREAD.
55. SKYLINE + GLASS-EASEL: WORKLETS const { runOnJS, runOnUI } = wx.worklet function someFunc(greeting) { console.log('hello', greeting); } function someWorklet() { 'worklet' runOnJS(someFunc)('skyline') } runOnUI(someWorklet)()
56. SKYLINE NEXT.JS SKYLINE + GLASS-EASEL: DIRECTIVES
57. SKYLINE SOLID START SKYLINE + GLASS-EASEL: DIRECTIVES
58. SKYLINE LYNX SKYLINE + GLASS-EASEL: DIRECTIVES
59. ECOSYSTEM THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
60. OPEN SOURCE
61. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
62. STANDARDIZATION
63. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
64. OTHER HOSTS
65. OTHER HOSTS
66. WHY DOES IT MATTER? THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
67. WHY DOES IT MATTER?
68. WHY DOES IT MATTER?
69. WHY DOES IT MATTER?
70. WHY DOES IT MATTER? — HIGH RESPONSE RATE AND LOW COST OF WECHAT-BASED DATA COLLECTION OF INFANT AND YOUNG CHILD FEEDING DATA IN RURAL QINGHAI PROVINCE, CHINA
72. LEARNING FROM OTHER INDUSTRIES…
73. LEARNING FROM OTHER INDUSTRIES…
74. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
76. “This attitude does not only work in the context of the game industry, though. Any other industry with strong architectural requirements can be a source of inspiration and knowledge for us and help us write better code on the web” An Actor, a model and an architect walk onto the web... • Surma
77. Entity Component Systems THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
78. MY HISTORY WITH THAT… THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
79. ME BACK IN 2016…
80. ME BACK IN 2017…
81. ME BACK IN 2017…
82. THE CREATOR OF REACT IN 2024…
83. THE CREATOR OF REACT IN 2024…
84. OVERVIEW THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
85. HARDWARE THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
86. CPUs ARE FAST ↝ BASIC MATH CAN BE DONE IN A FEW CYCLES. ↝ COMPLEX MATH REQUIRES A FEW MORE CYCLES—BUT IT’S STILL OK. ↝ SIMD MAKES THE THROUGHPUT EVEN BETTER BY PROCESSING MULTIPLE VALUES IN THE TIME IT TAKES TO DO A SINGLE INSTRUCTION.
87. MEMORY IS NOT AS FAST ↝ CPUS CAN ONLY OPERATE ON DATA THAT IS PHYSICALLY CO-LOCATED. ↝ EITHER IN A REGISTER OR IN THE L1 CACHE. ↝ EVERYTHING ELSE NEEDS TO BE LOADED FIRST.
88. MEMORY IS NOT AS FAST CPU MEMORY 100 NS ⁵ L1 CACHE 0.5 NS ⁵ L2 CACHE 4 NS ⁵
89. MEMORY IS NOT AS FAST CPU MEMORY 200X L1 CACHE 1X L2 CACHE 8X
90. MEMORY IS NOT AS FAST
91. MEMORY IS NOT AS FAST ↝ MEMORY ACCESS IS SLOW. ↝ CACHING CAN ALLEVIATE MOST OF THE LATENCY IF THE CPU CAN PREDICT THE ACCESS PATTERN—USUALLY, THE SIMPLE ONES. ↝ LINEAR PATTERNS ARE USUALLY GOOD. ↝ RANDOM PATTERNS ARE USUALLY BAD.
92. DATA-ORIENTED DESIGN THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
93. DATA-ORIENTED DESIGN
94. DATA-ORIENTED DESIGN ↝ A COLLECTION OF TECHNIQUES TO ANALYZE THE ACCESS PATTERNS OF THE DIFFERENT KINDS OF DATA AND SELECT DATA STRUCTURES THAT OPTIMALLY LEVERAGE THE UNDERLYING HARDWARE. ↝ THE OPTIMIZATIONS ARE OFTEN RELATED TO OPTIMIZING USAGE OF THE CPU CACHE, LIMITING LOADING/STORING TO RAM (CACHE MISSES), AND USAGE OF SIMD INSTRUCTIONS.
95. AoS X SoA ARRAY OF STRUCTURES STRUCTURE OF ARRAYS type Entity = { position: { x: number; y: number }; velocity: { x: number; y: number }; hitPoints: number; }; const entities: Entity[] = [ { position: { x: 0, y: 0 }, velocity: { x: 1, y: 1 }, hitPoints: 100 }, { position: { x: 10, y: 5 }, velocity: { x: -1, y: 0 }, hitPoints: 80 }, / / . . . ]; const positions = new Float32Array([0, 0, 10, 5 / * const velocities = new Float32Array([1, 1, -1, 0 / * const hitPoints = new Int16Array([100, 80 / * / const i = 0; console.log(positions[i * 2], positions[i * 2 + 1]); / console.log(velocities[i * 2], velocities[i * 2 + 1]); console.log(hitPoints[i]);
96. AoS X SoA ARRAY OF STRUCTURES STRUCTURE OF ARRAYS type Entity = { position: { x: number; y: number }; velocity: { x: number; y: number }; hitPoints: number; }; const entities: Entity[] = [ { position: { x: 0, y: 0 }, velocity: { x: 1, y: 1 }, hitPoints: 100 }, { position: { x: 10, y: 5 }, velocity: { x: -1, y: 0 }, hitPoints: 80 }, . ]; const positions = new Float32Array([0, 0, 10, 5 / * . . . * /]); const velocities = new Float32Array([1, 1, -1, 0 / * . . . * /]); const hitPoints = new Int16Array([100, 80 / * . . . * /]); / / Entity i = index in all arrays const i = 0; console.log(positions[i * 2], positions[i * 2 + 1]); / / x, y console.log(velocities[i * 2], velocities[i * 2 + 1]); console.log(hitPoints[i]);
97. AoS X SoA ARRAY OF STRUCTURES STRUCTURE OF ARRAYS ✅ EASY TO READ, WRITE, AND MODEL FOR US. ✅ ENTITIES ARE INTUITIVE BUNDLES. ❌ POOR DATA LOCALITY. ❌ ITERATING OVER MILLIONS SCALES BADLY. ❌ HARD TO PARALLELIZE BECAUSE DATA IS INTERLEAVED. ✅ EXCELLENT DATA LOCALITY. ✅ CACHE-FRIENDLY + SIMD ACCELERATION POSSIBLE. ✅ PARALLELIZABLE: SYSTEMS OPERATE ON “COLUMNS” INDEPENDENTLY. ❌ LESS INTUITIVE FOR DEVS USED TO OBJECTS. ❌ HARDER TO DEBUG AN ENTITY.
98. AoS X SoA ARRAY OF STRUCTURES STRUCTURE OF ARRAYS ✅ EASY TO READ, WRITE, AND MODEL FOR US. ✅ ENTITIES ARE INTUITIVE BUNDLES. ❌ POOR DATA LOCALITY. ❌ ITERATING OVER MILLIONS SCALES BADLY. ❌ HARD TO PARALLELIZE BECAUSE DATA IS INTERLEAVED. ✅ EXCELLENT DATA LOCALITY. ✅ CACHE-FRIENDLY + SIMD ACCELERATION POSSIBLE. ✅ PARALLELIZABLE: SYSTEMS OPERATE ON “COLUMNS” INDEPENDENTLY. ❌ LESS INTUITIVE FOR DEVS USED TO OBJECTS. ❌ HARDER TO DEBUG AN ENTITY.
99. ARRAY OF STRUCTURES X Y HP dX dY POSITION VELOCITY
100. dX dY ARRAY OF STRUCTURES UNITS 0 ⇢ 2 UNITS 3 ⇢ 5 UNITS … ⇢ N X Y X Y X Y X Y X Y X Y X Y X Y X Y dX dY dX dY dX dY dX dY dX dY dX dY dX dY dX dY HP HP HP HP HP HP HP HP HP
101. ARRAY OF STRUCTURES: DAMAGE HP HP HP HP HP HP HP HP HP UNITS 0 ⇢ 2 UNITS 3 ⇢ 5 UNITS … ⇢ N
102. HP STRUCTURE OF ARRAYS POSITIONS VELOCITIES HIT POINTS X Y X Y X Y X Y X Y X Y X Y dX dY dX dY dX dY dX dY dX dY dX dY dX dY HP HP HP HP HP HP HP HP HP HP HP HP HP
103. STRUCTURE OF ARRAYS POSITIONS VELOCITIES HIT POINTS X Y X Y X Y X Y X Y X Y dX dY dX dY dX dY dX dY dX dY dX dY HP HP HP HP HP HP HP HP HP HP HP HP THREAD A THREAD B THREAD C
104. DATA-ORIENTED DESIGN ↝ SMALL LOOPS THAT OPERATE ON ARRAYS OF FIELDS. ↝ MINIMIZE THE AMOUNT OF DATA IN CACHE THAT ISN’T READ OR WRITTEN DURING THE UPDATE. ↝ SEVERAL TRANSFORMS CAN RUN IN PARALLEL ON THE SAME LOGICAL OBJECT IF THEY DON’T SHARE FIELDS.
105. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
106. DATA-ORIENTED DESIGN: BENCHMARKS
107. DATA-ORIENTED DESIGN: BENCHMARKS
108. DATA-ORIENTED DESIGN: BENCHMARKS
109. DATA-ORIENTED DESIGN: BENCHMARKS
110. THRIVING WHEN AT SCALE…
111. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
112. ARCHITECTURE THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
113. ARCHITECTURE ↝ ENTITIES: UNIQUE IDENTIFIERS WITH ZERO OR MORE COMPONENTS THEY CAN CHANGE DYNAMICALLY. ↝ COMPONENTS: PLAIN DATATYPES WITHOUT BEHAVIOR. ↝ SYSTEMS: FUNCTIONS MATCHED WITH ENTITIES THAT HAVE A CERTAIN SET OF COMPONENTS.
114. ARCHITECTURE: COMPONENTS… X Y dX dY HP updatePosition() damage()
115. ARCHITECTURE: ENTITIES… X Y dX dY HP updatePosition() damage()
116. ARCHITECTURE: SYSTEMS… X Y dX dY updatePosition() damage() HP
117. ECOSYSTEM THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
118. THE GAMES WE PLAY…
119. THE GAMES WE PLAY…
120. THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
121. THE BROWSER APPS WE BUILD…
122. WHY DOES IT MATTER? THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
123. CODE DATA COUPLING… CODE COUPLING DATA COUPLING ↝ MIXING UP CONCERNS. ↝ BAD FOR MAINTAINABILITY. ↝ OFTEN SEEN AS AN ISSUE. ↝ MIXING UP DATA SLICES NOT ACCESSED BY THE SAME CODE. ↝ BAD FOR DATA SHARING. ↝ NOT OFTEN SEEN AS AN ISSUE.
124. CODE DATA COUPLING… CODE COUPLING DATA COUPLING ↝ MIXING UP CONCERNS. ↝ BAD FOR MAINTAINABILITY. ↝ OFTEN SEEN AS AN ISSUE. ↝ MIXING UP DATA SLICES NOT ACCESSED BY THE SAME CODE. ↝ BAD FOR DATA SHARING. ↝ NOT OFTEN SEEN AS AN ISSUE.
125. HARDWARE CONSTRAINTS…
126. HARDWARE CONSTRAINTS…
127. HARDWARE CONSTRAINTS…
128. Closing Thoughts THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
129. Software development continually goes through cycles… THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
130. Software development continually goes through cycles… What’s di ff erent now? What has turned into lost art? THINKING OUTSIDE THE BOX: ALTERNATIVE ARCHITECTURES FOR MODERN WEB APPLICATIONS
132. MATHEUS ALBUQUERQUE • @ythecombinator ⬆ ALL THE LINKS! THAT’S ALL, FOLKS! ФАЛА! 👋 🇲🇰 QUESTIONS?