我們傳統的互聯網極易受到各種問題影響。但在區塊鏈和 Web3 時代，我們擁有合理的替代方案：去中心化網絡。
2025 年 11 月某個寒冷的天氣中午，互聯網大範圍突然消失。從基輔到加州的用戶，打開社交媒體、電郵或工作應用程式時，看到的不是熟悉的內容，而是一堆錯誤訊息。原因？Cloudflare 突發大故障 ── 這家公司負責全球約五分之一的網絡流量。當 Cloudflare 於 11 月 18 日當機，大批主流平台（如 X，前稱 Twitter，以及 OpenAI 的 ChatGPT）都無法使用，影響數千人。工程師忙於搶修 Cloudflare 網絡上的「大規模 500 錯誤」時，人們不得不反思更深層的問題：現今互聯網仍然存有關鍵單點失效的風險。
這並非首次遇到類似事故。就在數星期前，一次 Amazon Web Services 故障亦導致 Snapchat、Reddit 等熱門網站癱瘓。
這些事故凸顯出網絡其實非常依賴有限幾間集中的基礎設施供應商。「一項服務只能堅固如其最脆弱的一環……而這一環未壞前通常察覺不到危機，」The Register 在 Cloudflare 事件時如此點評。
這一次，最弱的一環斷裂，把互聯網的一大部分拉下水。對不少人而言，這是互聯網脆弱性的又一次警鐘——也是推動更有彈性、去中心化網絡的契機。如果只需一家企業出事就可讓半個互聯網「感冒」，也許真要重新思考網絡該怎樣架構。
去中心化網絡的概念並不新穎——科技圈早有討論——但每逢大型網絡事故或資料醜聞，都令其迫切性再升級。支持者認為，真正的去中心化（或「分散式」）網絡，就算某一伺服器、企業、網絡失效，網站或服務依然可以正常運作。這種架構下，不會再有任何巨企壟斷我們大量網絡生活，是個迷人願景：當中央樞紐出現問題，網絡依然暢通，也沒任何單一機構可輕易箝制或審查。Cloudflare 這場大型事故之後，這種去中心化的想法得到更多支持者。有網絡檔案館資深評論：「現時網絡其實不算可靠，太集中化──要做到可靠、同時保障私隱兼有趣味，我們就得建立一個『分散網絡』。」
甚麼是去中心化網絡？（它如何運作？）
所謂去中心化網絡（常稱 Web3），是一種全新網絡架構，着重把主控權和數據分佈到眾多節點，而不是集中於少數伺服器或公司。簡言之，就是要顛覆現時互聯網的權力結構。
現時多數網絡活動，都由大型企業數據中心或官方系統主導。無論你在社交平台貼文、儲存檔案、或網上理財，基本上都依靠某間公司的伺服器。科技未來學者 Bernard Marr 指出，這傳統方法歷來就是最簡易的網絡建設路線：一間公司架設伺服器，提供服務，用戶按遊戲規則使用。相反，去中心化網絡構想類似點對點網絡，讓用戶直接互聯，而非經中央伺服器，也利用先進密碼學來保障安全。中間不再只有一家公司（及他們訂下的規矩）控制一切，而是權力分散至整個社群。
去中心化網絡的核心是區塊鏈技術，配合其他分散式協議。區塊鏈──比特幣開創、以太坊等平台推廣──讓數據和交易可以在全球電腦網絡透明保存、運作，毋須任何單一機構主宰。其原理是利用加密及分布式運算：每位用戶都有自己私密密鑰，能解鎖自家資料或資產，數據橫跨全球許多伺服器存有副本，而非集中在單一地方。
如有人試圖更改某伺服器的記錄，多個副本即能比對發現不符。
單一伺服器故障不致使數據下線，亦不會有中央管理員偷偷修改紀錄。
理論上，你主導自己於去中心化網絡上的資料——而非交由 Google、Amazon 或 Facebook 各自為政。
這種架構令系統實現「無需信任」（trustless）及「無需許可」（permissionless）──這是 Web3 的潮語。「無需信任」指完成交易時不用靠中介平台保證，代碼和網絡共識就能保證完整性。例如你想直接轉加密貨幣給朋友，區塊鏈演算法即可取代銀行作為驗證和操作單位。
「無需許可」則代表你毋須經「守門人」同意加入生態。在現時網絡，支付平台或社交網站可以單方面封鎖你；但在分散式網絡，只要符合理協，中央權威沒法封你帳或拒你服務。支持者認為，這將令自由和創新更易實現。
Coinbase 行政總裁 Brian Armstrong 指出，「我們需要實現分散化協議，打造更全球化、公平和自由的金融體系」，他舉例指比特幣和以太坊等加密網絡運作便無需中心化主控。
要強調，「去中心化」並非毫無規則——而是規則改由程式和用戶共識執行，而非公司或政府高壓統領。不少去中心化項目以開源及社群自治形式運行。有些以自動化智能合約（自執行代碼）實現規則透明化。
亦有項目試驗 DAOs（去中心化自治組織）等新模式，持有代幣者可以表決決策。整體目標是將權力重回用戶手上。不必把資料、內容、交易交給中央平台後祈求對方守規矩，去中心化網絡讓你能以自家條件管理自身數碼生活，密碼學保護你的私隱和安全。正如早期網絡先驅說：「網絡寫法會塑造我們的網絡生活」──去中心化運動正是想以自由和韌性為本，重新設定網絡。
去中心化網絡的利與弊
去中心化網絡的前景無疑極具吸引力，優點就是現時網絡各種積弊的解藥。首先是強大的韌性：避免單點失效，分散式網絡即使部分遭遇攻擊或離線，也能持續運作。
正如 Cloudflare 事故這種大當機，在分散系統下就不會牽連太廣。
檔案或網站可經數以百計節點分布全球，一旦其中一份仍在線，用戶就能瀏覽。這同時帶來更強的抗審查性。今天，要移除網上資訊，政府或公司通常只需向平台施壓或切斷主機服務。在分散式網絡下，沒有單一「關閉開關」或「咽喉位」可一刀封殺。訊息難以徹底封鎖，有助言論自由和知識傳播（數碼圖書館員、網絡運動者長久以來都支持這種願景）。
另一常見優點是用戶自主和私隱保障。
因去中心化服務上的數據通常有加密，並鎖定由你持有的密鑰，就由你決定誰能查閱。
你的個人數據、身份、內容不會像以往般被巨企伺服器存放、挖掘或外洩。這帶來巨大影響：社交網絡無法再追蹤你行為賣廣告，也不會再有上百萬用戶資料外洩醜聞。理想情況下，你主導自己數據，可以「隨身攜帶」——舉例你的社交媒體檔案存在個人錢包或私密資料庫，需要用時插入任何服務，不再被單一平台鎖死。
所以，去中心化網絡有望大幅提升私隱和個人自主度，這亦是 Tim Berners-Lee（網絡之父）及其他理念者一直追求的方向。
經濟與創新同樣受惠。分散化可削弱巨企壟斷，令競爭更公平。倘若沒有單一公司掌控平台，開發者和創業者可在開放協議上自由搭建新服務，無需申請許可。這情景令人憶起互聯網早期的開放與創新，或會推動新一輪技術浪潮。社群可根據自身需要，自建網絡和應用程式，設置原生代幣機制（如加密貨幣或各類代幣）獎勵參與及維護系統的人。以金融為例，分散式金融（DeFi）應用讓人可點對點借貸、交易資產，毋須銀行，費用往往更低，且用戶覆蓋全球。
支持者認為，Web3 和加密技術可以「更新」現有金融系統甚至其他產業，消弭守門人，讓用戶真正持有平台股份。這種新願景即是集體普惠：用戶「共同擁有」社交平台或叫車服務，不再只是產品。
然而，理想背後也有不少現實挑戰和缺點。其中一大難題就是複雜性。現時的去中心化應用（dApps）往往需要 dealing with crypto wallets, secret keys, and unfamiliar interfaces – a far cry from the user-friendly experiences people are used to. As Deloitte notes, “the on-ramp to Web3 is not a one-click solution”, and until using a decentralized service is as seamless as using Google or Amazon, mainstream users will struggle.
處理加密錢包、私鑰同唔熟悉嘅介面——完全唔係一般人習慣用嘅「易用」體驗。正如德勤（Deloitte）所講：「進入Web3唔係一鍵搞掂。」直到用去中心化服務變得同用Google或Amazon一樣順暢，大眾用戶都會感到困難。
Managing one’s own keys (which act like the password that, if lost, means losing access forever) is daunting. Mistakes can be costly and irreversible on blockchain systems. User experience problems have absolutely slowed Web3 adoption , and solving them is crucial if the decentralized web is to go beyond tech enthusiasts.
管理自己私鑰（就好似密碼一樣，唔見咗就永遠搵唔返）令人卻步。喺區塊鏈上犯錯可以好貴同無得補救。用戶體驗真係拖慢咗Web3嘅普及，如果想去中心化網絡唔只限於科技愛好者，呢個問題一定要解決。
Another issue is performance and scalability. Decentralized networks, especially blockchain-based ones, have historically been slower and more resource-intensive than their centralized counterparts. For example, Bitcoin can process only a handful of transactions per second and early Ethereum struggled with high fees and congestion when usage spiked. Though newer networks and upgrades have improved speeds, there’s often a trade-off between decentralization and efficiency. Truly distributed systems have to coordinate data among many nodes, which can introduce lag or limits on throughput.
另一個問題就係效能同擴展性。去中心化網絡，尤其係區塊鏈底層嗰啲，一向都比中心化系統慢同食資源。例如，比特幣每秒淨係得幾單交易，早期以太坊用戶多時就塞車兼收費高。雖然新一代網絡同升級加快咗速度，但好多時都要喺去中心化同效率之間取捨。一個真分散嘅系統要喺咁多個節點之間協調數據，自然會有延遲或者容量限制。
By contrast, a centralized service can be heavily optimized in one data center. This leads to debates: some newer “Layer 1” blockchains sacrifice some decentralization to achieve higher speeds – which arguably defeats the purpose if taken too far.
相反，中心化服務可以集中喺一個數據中心極速優化。呢點都引伸到一啲爭論：有啲新一代「Layer 1」區塊鏈為求快啲，犧牲咗啲去中心化嘅程度——如果去得太盡，其實都違反初衷。
The bottom line is that to compete with Web2 platforms at scale, decentralized technologies must overcome technical challenges around speed, capacity, and energy usage (early blockchains infamously used huge amounts of electricity, though newer consensus mechanisms are greener).
最終，為咗大規模同Web2平台競爭，去中心化技術一定要喺速度、容量、能源消耗等技術難題上解決（早期區塊鏈好出名用電多，雖然新共識機制環保好多）。
Governance and accountability pose further cons. If something goes wrong in a decentralized network – say, a bug that loses users money or a harmful piece of content spreading – who is responsible? With no central owner, it can be unclear how to resolve disputes or enforce laws. Total decentralization can be a double-edged sword: it removes the corporate overlord, but also means there’s no help desk to reset your password, and no authority to reverse fraudulent transactions or moderate illegal content. This raises safety and legal concerns. For instance, regulators worry that anonymous, decentralized platforms could facilitate money laundering or other crimes without oversight.
治理同問責都係另一個弊處。如果去中心化網絡出事——例如漏洞令用戶損失金錢、或者有有害內容瘋傳——咁邊個負責？冇一個中心化老細，有咩糾紛或者要執法時就成疑問。完全去中心化係一把雙刃劍：你甩咗企業老爺，但都即係冇人幫你重設密碼、冇機構可以倒轉詐騙交易、或者管理非法內容。呢啲都好多安全同法律風險。例如，監管機構擔心匿名去中心化平台會淪為洗黑錢或者其他犯罪活動嘅溫床，唔受監管。
Likewise, a fully decentralized social media might become a haven for disinformation or abuse if there’s no mechanism to control malicious behavior. Advocates are experimenting with community moderation and on-chain governance to tackle this, but it’s an evolving challenge.
同樣，如果冇機制制衡，完全去中心化嘅社交媒體有機會變成假消息或濫用問題嘅溫床。支持者現時都研究緊社群自治同區塊鏈上鏈治理等對策，但始終都係一個不斷演化嘅挑戰。
Finally, there’s the risk that the “decentralized” ideal doesn’t live up to the hype in practice.
最後，「去中心化」呢個理想喺現實中有機會未必兌現。
Skeptics like Twitter co-founder Jack Dorsey point out that many Web3 projects are backed by powerful venture capital firms – meaning power may just be shifting from one set of gatekeepers to another. “You don’t own ‘Web3.’ The VCs and their LPs do… It’s ultimately a centralized entity with a different label,” Dorsey quipped in late 2021.
懷疑論者，例如Twitter聯合創辦人Jack Dorsey，就指好多Web3項目背後都有大財團或者風險資本支持——其實只係權力由一班守門人，轉咗去第二班。「你唔係擁有『Web3』，VC同佢哋啲LPs先擁有⋯⋯最終都係換咗個名嘅中心化企業啫。」Dorsey喺2021年底咁講。
In other words, if a few wealthy investors control the major blockchain networks or token supplies, the web might not be as egalitarian as advertised. This criticism serves as a reminder that technology alone doesn’t guarantee decentralization; governance and ownership matter too. The decentralized web movement will have to ensure it doesn’t simply create new oligarchs under the banner of decentralization.
即係話，如果主要區塊鏈或者Token供應都由少數大財主操控，呢個「人人平等」嘅網絡就可能只係一個包裝。呢個批評提醒大家，得技術都保障唔到去中心化，治理同擁有權一樣咁重要。去中心化網絡運動一定要確保唔會只係換咗名嘅新寡頭。
Decentralized vs. Today’s Internet: Key Differences
Figure: Classic network models – a centralized network (left) relies on one core node, a decentralized network (center) has multiple hubs, and a distributed network (right) has no central authority.
圖：經典網絡模式——左邊係中心化網絡（靠一個核心節點），中間係去中心化（有多個樞紐），而右邊分布式網絡就完全冇中央管理。
The more distributed, the more the system can route around failures or control.
分布愈分散，系統就愈能夠自動繞過失效或者外部操控。
To grasp how the decentralized web diverges from the status quo, consider how information flows today. The current Web 2.0 model is largely centralized: data is stored on specific servers, and you typically access it by reaching out to those servers (often owned by whoever provides the service). It’s a client-server architecture. For example, when you visit a website or use a cloud app, your browser is fetching content from that company’s server farm. If that server (or the network path to it) is down, the content becomes unavailable. Control is also centralized – whoever runs the server can decide what content it serves, who gets access, and can potentially log or modify what you’re doing.
要了解去中心化網絡同現狀有幾唔同，可以諗吓而家資訊點流動。現時嘅Web2.0模式基本都係中心化：數據儲落特定服務器，你想攞就要搵番果啲服務器（通常都係提供服務嘅公司擁有）。呢種叫做客戶端-服務器架構。例如你上個網站或者用雲端App，其實你部瀏覽器係喺對方公司啲伺服器度攞內容。如果果台（或者去果台條路）死咗，你就冇得睇。權力都被集中咗——營運伺服器嗰個決定派咩內容，邊個有權登入，甚至可以記錄或改你做咗咩。
In contrast, the decentralized web uses a peer-to-peer model where information is distributed across many nodes.
相反，去中心化網絡用點對點架構，資訊分布喺好多個節點。
There is no single “origin server” for a piece of data. Instead, any node in the network that holds the data can serve it to others. This is sometimes called content-addressed networking. A current web address (URL) points to a location on a specific server. A decentralized web address might point to a content hash – a unique fingerprint of the data – and the network can retrieve it from any node that has that content. In practical terms, it’s like the difference between calling a particular library branch to request a book versus asking a network of libraries if anyone has the book and can share it.
每份數據都並無「原始伺服器」，只要網絡上有節點儲咗該數據，就可以派俾其他人。呢種方式叫內容定址（Content-addressed Networking）。現時Web網址(URL)係指向某台伺服器嘅某個位；但去中心化網址會指向內容哈希值——即係資料獨有嘅指紋——任何一個有呢份data嘅節點都可以出貨。實際上，好似你以前要搵書打俾分館，依家就直情問成個圖書館網絡有冇人有本書可以分享。
One pioneering system enabling this is IPFS (InterPlanetary File System), which lets files be retrieved from dozens of computers globally rather than one host, similar to how BitTorrent shares files among users.
實踐呢種模式嘅先驅之一係IPFS（星際檔案系統），文件可以由世界各地好多部電腦攞返，而唔只靠一台主機，好似BitTorrent咁多人分開Share。
This structural shift brings several key differences.
呢個架構上轉變帶嚟幾個重點分別。
Reliability is one: the internet’s underlying design has always been distributed (able to route around broken nodes), but the web layer built on top wasn’t. A decentralized web extends the original internet ethos to content itself. If one node holding a piece of data goes offline, the data isn’t lost – other peers can fill in. Websites could be served like swarms, not from single data silos. This is why the decentralized web is often called the “distributed web”: it would be far more fault-tolerant, much like the internet’s packet routing is resilient by design. Outages would require knocking out many nodes, not just hitting one target.
可靠性係其一：互聯網底層設計本身已經分布式（可以繞過壞咗嘅節點），但Web層就唔係。去中心化網絡係將呢個原則延伸去到內容本身。如果一個節點跌咗，資料冇咗？——其實唔使怕，其他peer可以補位。網站可以群體式派送，而唔係淨係喺單一數據孤島出貨。呢個咪就係點解去中心化網絡又叫「分布式Web」：佢容錯力高好多，好似互聯網本身包咗路由冗餘一樣。要斷線唔只要攻擊一個目標，而係搞掂晒好多節點。
Control and governance form another major difference. On today’s web, control is highly centralized in platform providers. Facebook alone decides what’s allowed on Facebook and can unilaterally ban users or content. On a decentralized social network, control would be more federated or user-driven – for instance, each user or community might moderate their own slice, and there’s no single company that can dictate terms to everyone.
控制同治理都係一大差別。依家呢個Web，大權都集中喺平台老細手上。得Facebook話事Facebook有咩可以上／唔可以上，可以單方面Ban人Ban內容。去中心化社交網絡就唔同，可以聯邦式，或者由用戶群自己話事，例如每個用戶或群組睇住自己片段，冇一間公司可以一錘定音。
Even domain naming could change: rather than using the centralized DNS run by ICANN (which can censor or seize domains via registrars), people are experimenting with blockchain-based domain name systems (like *Ethereum Name Service’s *.eth domains) that no one company can simply confiscate.
就連域名系統都想變天：唔係用ICANN管理、易被登記商封殺/充公嘅集中式DNS，而係實驗緊以區塊鏈為本，例如Ethereum Name Service嘅*.eth域名，冇一間公司可以話充就充公。
In short, today’s internet is built on implicit trust in central entities, whereas a decentralized web shifts trust to transparent code and consensus.
簡單講，現時網絡靠隱性信任中心化實體；去中心化網絡就將信任轉移俾透明程式同共識機制。
Identity and data ownership also differ. Currently, users juggle accounts with every service – each one storing your profile and data on their servers. The decentralized web envisions a world where you have a single, sovereign identity (or a set of identities) that you control. You’d log in with a crypto wallet or digital identity that you manage, not a password stored on a company database. Your personal data might live in an encrypted storage that only you can unlock, and you grant services permission to use it when needed.
身份同數據擁有權都差好遠。依家你要同N個服務開N個帳戶，每間公司存你啲Profile同數據喺自己伺服器。去中心化網絡就想做到一個（或者一組）你自己話事嘅主權身份。你以自己管理嘅加密錢包或數字身份登入，而唔係公司數據庫嘅密碼。你啲個人數據會儲喺你自己專用、加密嘅儲存空間內，只係你可以開鎖，有需要先批服務用。
This flips the script from the status quo, where we routinely hand over personal information to use “free” services. As Sir Tim Berners-Lee describes in his Solid project, it’s like each person having their own data vault (or “pod”) and services come to your pod to fetch what they need, with your consent, rather than you uploading your data to them permanently. The effect would be to greatly reduce the leverage of tech giants who currently stockpile user data as a resource. Instead, users would be the ultimate source of truth for their data – an idea often summarized as “self-sovereign data.”
呢種做法完全顛覆咗現況：而家我哋用「免費」服務都要交齊私隱俾人。正如Tim Berners-Lee喺Solid計劃講，等於每個人有個自己嘅數據保險箱（Pod），服務要data就嚟問你攞，你肯先得，而唔係永久將數據上載畀佢哋。咁樣會大幅減低科技巨頭靠囤積用戶數據嚟壓榨市場。最終你先係自己數據絕對話事人——所謂「自我主權數據」。
Additionally, the business models and incentives on a decentralized web are likely to differ from today’s ad-driven, centralized models. In Web2, companies monetize by locking in users (network effects) and extracting value from data or transactions. In Web3, many services have built-in cryptocurrencies or tokens that reward users for contributing to the network’s operation.
另外，去中心化網絡嘅商業模式同獎勵結構都應該同今日靠廣告、集中壟斷嘅唔同。Web2裏面，企業主靠綁死用戶（網絡效應）、抽數據或者交易價值賺錢，但Web3好多服務內建加密貨幣或Token，獎勵用戶貢獻網絡運作。
For example, if you provide storage space to a file-sharing network, you might earn tokens; if you curate quality content, a social platform might reward you rather than just profiting off you.
例如你提供硬碟空間俾分享檔案嘅網絡，就有Token收；你幫社交平台整理好內容，平台可以回饋你，而唔係淨係靠你賺錢。
These tokenized incentives could create more participatory economies.
呢啲Token化獎勵制度可以造就更高參與度嘅新經濟。
However, they also introduce new dynamics – speculation, governance votes based on token holdings, and so on – which are quite unlike the way traditional web companies operate. It’s a grand experiment in aligning the interests of a platform’s users with the platform’s success, in theory avoiding the exploitation or privacy invasion we see in some of today’s internet giants’ practices.
但同時都會帶嚟新挑戰——投機炒賣、用Token持有量來投票治理等等，係傳統網絡公司冇見過嘅玩法。理論上，係一個令用戶同平台利益一致嘅實驗，盡量防止依家啲網絡巨頭呃盡你私隱或者剝削嘅現象。
Technologies Enabling the Decentralized Web
What will it take to build this new web?
咁要砌出呢套新網絡，需要啲咩技術？
In practice, the decentralized web isn’t a single thing, but a stack of technologies and protocols coming together. At the foundation is blockchain – the distributed ledger technology that proved decentralization could really work at scale (starting with Bitcoin). Blockchains provide a way to achieve consensus across a global network of nodes, so everyone agrees on the state of data without a central referee.
實際上，去中心化網絡唔係一件產品，而係一堆基礎技術同協定組合。最底層就係區塊鏈——呢種分布式帳簿證明咗去中心化係可以大規模運作嘅（由比特幣開始）。區塊鏈可以令全球節點達成數據共識，不需要中央裁判都可以全部人信晒。money (cryptocurrency) and smart contracts. Ethereum, for instance, is a blockchain that can execute Turing-complete programs (smart contracts) on a decentralized network of thousands of computers. It’s the backbone for many decentralized applications, from financial protocols to games and marketplaces.
金錢（加密貨幣）同智能合約。例如，Ethereum 係一條可以喺去中心化網絡上運行圖靈完備程式（即智能合約）嘅區塊鏈，有成千上萬部電腦參與。佢係好多去中心化應用（由金融協議到遊戲同市場平台）嘅支柱。
Other blockchains (Solana, Polkadot, Avalanche, and more) are also competing, each with different approaches to speed, security, and decentralization. Together, these form the transactional and computational layer of Web3 – effectively the new “servers” in the cloud, except spread across many independent operators.
其他區塊鏈（Solana、Polkadot、Avalanche等等）都喺度競爭，各有唔同嘅速度、安全同去中心化嘅做法。呢啲一齊組成咗 Web3 嘅交易同運算層——即係新一代「雲端伺服器」，但係分佈喺好多獨立營運者手上。
But decentralizing compute and transactions is just one piece. Equally important is decentralized storage and data delivery.
去中心化運算同交易只係其中一部分。同樣重要嘅，係去中心化嘅儲存同數據傳送。
This is where technologies like IPFS (InterPlanetary File System) and its cousin Filecoin come in. IPFS is a protocol that allows files to be stored and fetched from a peer-to-peer swarm of computers, rather than one central server. It addresses content by a unique hash of the file, and peers on the network can serve that content if they have it. In practice, this means a website or video on IPFS isn’t sitting in one data center – it’s potentially spread across dozens of nodes.
呢度就需要用到 IPFS（星際文件系統）同佢嘅「表兄弟」Filecoin 呢啲技術。IPFS 係一個協議，可以令文件喺點對點嘅電腦群之間儲存同下載，而唔使靠中央伺服器。佢用文件嘅獨特雜湊（hash）嚟定位內容，網絡入面嘅節點儲有內容就可以提供出嚟。實際上，即係話一個網站或者影片唔會只係擺喺某一個數據中心——可能係分佈喺幾十個節點。
Filecoin adds an incentive layer on top of IPFS, rewarding nodes (with cryptocurrency) for storing files over time, thereby creating a robust, self-healing storage network.
Filecoin 喺 IPFS 基礎上加入咗經濟誘因機制，即係用加密貨幣獎勵節點長期儲存文件，令呢個儲存網絡更堅固同有自我修復能力。
There are other decentralized storage projects too, like Arweave (which focuses on permanent, archived storage), Storj and Sia (which distribute encrypted chunks of users’ files across many hosts). These systems aim to ensure that the web’s content remains available and verifiable, without needing traditional web hosting. In fact, even during the Cloudflare outage, some tech-savvy users noted that certain content on IPFS was still reachable through alternative gateways – an early hint of resilience.
仲有其他類似嘅去中心化儲存項目，例如 Arweave（主打永久備份儲存）、Storj 同 Sia（將用戶加密咗嘅文件分拆到多個主機）。呢啲系統個目標，都係確保網上內容可以持續存取同驗證到，而唔使用傳統網站服務供應商。事實上，就算 Cloudflare 當日斷咗線，有啲精通技術嘅用戶都發現，IPFS 上面嘅某啲內容仲可以經 alternative gateways 搵返到——證明咗佢嘅彈性。
Another key technology area is decentralized naming and identity. The traditional DNS (Domain Name System) is hierarchical and centralized at the top levels. In a decentralized web, you’ll want human-readable addresses that aren’t tied to centralized authorities. Blockchain-based naming services are tackling this. Ethereum Name Service (ENS), for example, lets users register “.eth” domain names (like alice.eth) which can map to cryptocurrency wallets, smart contracts, or even websites hosted on IPFS. These records are stored on Ethereum’s blockchain itself, making them censorship-resistant domain names. There are others, such as Handshake and Unstoppable Domains, attempting similar with different approaches. For user identity, there’s work on DIDs (Decentralized Identifiers) and identity hubs where you control your credentials and profile, and only you can cryptographically prove your identity to apps (instead of logging in via Google or Facebook). These tools help replace the centralized gatekeepers of identity and naming that we rely on today.
另一個重要技術範疇係去中心化命名同身分認證。傳統 DNS（網域名稱系統）本身係分層同集中管理。去中心化網絡下，你都想要方便人理解嘅網址，但又唔係綁死喺中央機構度。區塊鏈命名服務就幫你搞掂。好似 Ethereum Name Service（ENS），用戶可以登記 ".eth" 域名（例如 alice.eth），呢啲域名可以指向加密貨幣錢包、智能合約、甚至喺 IPFS 託管嘅網站。紀錄全部擺喺 Ethereum 區塊鏈，做到防審查嘅功能。其他例如 Handshake、Unstoppable Domains 都用唔同嘅方法做緊類似嘅事。身分認證方面，有 DIDs（去中心化身分識別）同 identity hub，俾你自己完全控制賬號資料，同時用加密方法證明自己身份，而唔使經 Google 或 Facebook 登入。呢啲工具可以取代現時集中的名字同身份「守門人」。
Smart contracts and protocols form the application logic layer of the decentralized web.
智能合約同協議就組成咗去中心化網絡嘅應用邏輯層。
On Ethereum and similar platforms, developers have created protocols for everything from decentralized exchanges to social media. These are essentially programs that run automatically on the blockchain.
喺 Ethereum 同其他類似平台，開發者已經創建咗多種協議，由去中心化交易所到社交媒體都有。呢啲其實係一啲自動運作、跑喺區塊鏈上嘅程式。
For example, a decentralized exchange (DEX) like Uniswap is just a smart contract on Ethereum that lets users swap tokens directly from their wallets – no central exchange operator needed. The code defines how liquidity pools work, how prices are determined, and anyone can interact with it or even build on top of it. There are smart contract protocols for lending (Compound, Aave), for media (Mirror, a decentralized publishing platform where writers own their content via NFTs), for music streaming (Audius), and many more. Interconnecting these, projects like The Graph provide decentralized indexing, letting dApps query blockchain data in a trustless way (a bit like Google for blockchain data, but community-run).
例如，一個去中心化交易所（DEX）如 Uniswap，其實就只係 Ethereum 上面一個智能合約，俾用戶直接用自己嘅錢包換代幣——唔使有中央營運嘅交易所。啲程式碼就定咗流動性池點運作、點計價錢，任何人都可以用或者上面再擴展。借貸（如 Compound、Aave）、媒體（例如 Mirror——寫手可以用 NFT 擁有內容嘅去中心化出版平台）、音樂串流（Audius）等功能，都有相應智能合約協議。為咗將佢哋互相連接，類似 The Graph 嘅項目做到去中心化索引，等 dApp 可以去信任查詢區塊鏈數據（有啲似 Google 幫區塊鏈數據做搜尋，不過係由社群運作）。
Another piece of the puzzle is peer-to-peer networking and communication protocols: for truly decentralized messaging or social feeds, protocols like Libp2p (used by IPFS) or GossipSub can propagate data among nodes without a server hub. And for real-time communications, there’s Matrix (an open decentralized chat protocol) or newer attempts like P2P versions of WebRTC.
另一個重要組件係點對點網絡同通訊協議：如果真係要做到去中心化嘅即時通訊或者社交訊息推送，可以用 Libp2p（IPFS 用緊）或 GossipSub 之類，唔使中央伺服器就可以傳播數據。即時通訊方面，有 Matrix（開放式去中心化聊天協議），或者更新嘅 P2P WebRTC 解決方案。
Crucially, many of these technologies are already available in some form.
但好重要嘅係，以上呢啲技術，其實好多已經有初步可用嘅形態。
Blockchain networks are live (with Ethereum even transitioning to a more energy-efficient model now), IPFS is operational and used under the hood by Brave browser and others, and ENS has registered millions of names that people use for crypto wallets. However, they are not yet seamless or ubiquitous. They often require technical know-how to use directly. So an important technology piece is actually bridging tools and middleware to connect the decentralized web with the traditional web.
區塊鏈網絡已經係現實中運行緊（Ethereum 而家都開始轉向節能運作模式），IPFS 已經可以用，Brave 瀏覽器等等都暗地裏用咗呢啲技術。ENS 已經註冊咗幾百萬個域名，好多人用嚟接收加密貨幣。不過，依家佢哋都仲未算係無縫、普及，大多數情況直用都要一定技術知識。所以一個重要技術環節，其實係要用橋接工具同 middleware，將去中心化網絡同傳統網絡連接返埋一齊。
For instance, web browsers are beginning to integrate these technologies – the Brave browser has native IPFS support, meaning it can resolve ipfs://... addresses directly and retrieve content from the distributed network. Opera and others have experimented with similar integration.
例如，部分網頁瀏覽器已經開始整合呢啲技術——Brave 已經支援原生 IPFS，可以直接識別 ipfs://... 地址並喺分散網絡搵返內容。Opera 等其他瀏覽器亦試過類似功能。
There are also “gateways” that let anyone access IPFS content via a normal HTTPS link (though gateways themselves can be centralized points, they help with on-boarding).
亦有啲所謂「gateway」設施，俾人可以用普通 HTTPS 連結睇到 IPFS 內容（雖然 gateway 本身可以係集中點，但有助新手上手）。
Likewise, browser plugins or built-in crypto wallets (like those in Brave or upcoming in Chrome via standards) allow users to interact with blockchain-based sites (often called dApps) just as easily as they do with Web2 sites. All these connective technologies are aimed at making the decentralized web invisible in usage – you shouldn’t need to know what IPFS or Ethereum is to benefit from them.
同樣，瀏覽器插件或者內置加密貨幣錢包（例如 Brave 已有，Chrome 之後都有標準支援）都俾用戶咁易操作區塊鏈網站（平時叫 dApps），好似用 web2 網站咁直接。呢啲連接技術嘅目標，係令去中心化網絡用起嚟「無形」，即係你唔使明咩叫 IPFS 或 Ethereum 都可以受惠。
The final ingredient isn’t a technology per se but a challenge: standards and interoperability.
最後一個要素唔完全係技術，而係挑戰：標準化同互通性。
For the decentralized web to truly function as a web (a unified network of networks), different projects and chains will need to talk to each other. Initiatives like cross-chain bridges and emerging standards (for example, W3C’s work on decentralized identity, or multi-chain token standards) are trying to ensure we don’t end up with a bunch of isolated mini-webs. It’s akin to ensuring that email providers can all exchange email despite different software – a common protocol is key. Technologists are working on this, but it’s a space to watch. In summary, the decentralized web is being built with blockchains, distributed storage, crypto-based identity, open protocols, and new web browsers – a potent mix that together could re-architect internet services as we know them.
如果要實現真正嘅「網絡」——統一咁將網絡連在一起——唔同項目同區塊鏈之間都要能夠互通。啲跨鏈橋、同新興標準（例如 W3C 研究緊嘅去中心化身份、或多鏈代幣標準）都想確保唔會最後只係變成一堆孤立嘅「小網域」。有啲似唔同電郵供應商之間，雖然用唔同軟件都能互發電郵——一個共用協議係關鍵。技術人員正努力做緊呢件事，值得大家關注。總結來講，去中心化網絡係用區塊鏈、分散存取、用加密技術管理身份、開放協議同新型瀏覽器等技術拼合出來——有機會大大改變而家互聯網嘅底層架構。
Real-World Examples of the Decentralized Web in Action
而家去中心化網絡仲係起步階段，但其實已經有一啲現實例子證明咗呢個新生態嘅潛力同挑戰。
One prominent example is in finance: decentralized exchanges (DEXs). Consider Uniswap, a DEX running on the Ethereum blockchain.
例如，金融領域一個好突出例子就係去中心化交易所（DEXs）。以 Uniswap 為例，佢係運行喺 Ethereum 區塊鏈嘅去中心化交易所。
Without any centralized operator, Uniswap enables users to swap cryptocurrencies directly from their own wallets, using an automated liquidity pool mechanism. It launched only a few years ago, but by 2023 Uniswap was handling trading volumes on par with, or even exceeding, those of big centralized crypto exchanges.
完全冇中央營運商，Uniswap 俾用戶直接用自己錢包兌換加密貨幣，同時靠自動化流動性池作機制。Uniswap 只係幾年前推出，但去到 2023 年，佢每日交易量已經同主流集中交易所睇齊，甚至超過某啲大平台。
Uniswap’s rise shows how a Web3 application can challenge traditional gatekeepers (in this case, exchanges like Coinbase or Binance) by offering an open, user-driven alternative. It’s not perfect – users still face issues like high transaction fees during peak times – but it proves the decentralized model can be competitive at scale.
Uniswap 嘅崛起證明咗 Web3 應用點挑戰傳統「守門人」（例如 Coinbase 或 Binance 呢類交易所），用開放、用家主導嘅替代選擇。當然未係完美——用戶繁忙時段都會遇到手續費貴等問題——但證明咗呢種去中心化模式可以有競爭力。
Other DeFi platforms like Aave (for lending) and MakerDAO (for stablecoins) similarly operate with no central bank or company in charge, yet have secured tens of billions in user assets collectively at their peaks, offering loans and generating interest through smart contracts.
其他 DeFi 平台（好似 Aave 做借貸、MakerDAO 做穩定幣）一樣都冇中央銀行或公司管，但高峰時都能儲到用戶數百億資產，靠智能合約出貸同產生利息。
Another sphere seeing real decentralized web usage is digital storage and content delivery. The IPFS network, for example, has been employed to preserve datasets and even entire websites in a censorship-resistant way. One high-profile use was by activists and archivists to create IPFS mirrors of sites that were taken down or blocked.
另一個已經見到實際應用嘅範疇，就係數碼儲存與內容傳播。IPFS 網絡就俾人用嚟保存數據集，甚至整個網站都可以用去中心化方式避開審查。特別高調例子，就係有啲活躍份子同檔案管理員會幫被下架或被封鎖嘅網站整 IPFS 鏡像。
The Open Bazaar project, though now defunct, was a bold attempt at a decentralized e-commerce marketplace (a bit like a peer-to-peer eBay) where buyers and sellers could transact directly using cryptocurrency, without a company in the middle.
Open Bazaar 項目（而家已經停咗），都曾經好大膽咁試過做去中心化電商市場，好似 P2P 版 eBay，用戶用加密貨幣直接做生意，中間完全冇公司抽佣。
And in web browsing, the Brave browser has stepped up as a mainstream gateway to decentralized content.
至於用網絡瀏覽方面，Brave 瀏覽器依家都成為主流通往去中心化內容嘅出入口。
With millions of users, Brave not only blocks ads and trackers (improving privacy), but also integrates Web3 features: it has a built-in crypto wallet, and importantly, it became the first browser to integrate IPFS support natively. This means a Brave user can type an IPFS link or. eth domain and retrieve that content from the decentralized network automatically, instead of going through a central server or gateway.
Brave 有幾百萬用戶，唔止幫你阻隔廣告同追蹤器（保護私隱），仲整合咗 Web3 功能，有內置加密貨幣錢包，最重要係第一個原生支援 IPFS 嘅瀏覽器。即係話，Brave 用家只要打 IPFS 連結或 .eth 域名，可以自動喺去中心化網絡直接攞返相關內容，唔洗經中央伺服器或 gateway。
“We’re thrilled to be the first browser to offer a native IPFS integration… Providing Brave’s users with content that’s more resilient to failure and control,” Brave’s team said at launch.
Brave 團隊喺發佈時講：「我哋好興奮成為第一個提供原生 IPFS 整合嘅瀏覽器……俾 Brave 用戶可以攞到更具彈性、唔怕失效或被人操控嘅內容。」
By making distributed content accessible to anyone with a simple browser update, Brave has effectively put a piece of the decentralized web into the hands of everyday internet users.
只要簡單更新瀏覽器，Brave 就將分散式內容帶到普通網民手中，等於將去中心化網絡變咗日常上網選擇之一。
Social media and communication are also seeing decentralized alternatives gain traction, especially amid real-world controversies.
社交媒體同即時通訊方面，去中心化選擇亦開始流行，特別係現實世界有爭議事件時。
After the tumultuous changes at Twitter (now X) in 2022, many users flocked to Mastodon – an open-source, federated social network. Mastodon isn’t blockchain-based, but it’s decentralized in the sense that anyone can run a server (an “instance”), and those servers interconnect to form a Twitter-like experience. No single company or CEO can
自從 2022 年 Twitter（即依家嘅 X）發生大地震之後，好多用戶都轉投 Mastodon 呢個開源、聯邦制嘅社交網絡。Mastodon 雖然唔係建基於區塊鏈，但佢都屬於去中心化，因為任何人都可以自己開一個伺服器（即「實例」），啲伺服器之間又可以互連，組成 Twitter 式嘅使用體驗。冇一間公司或者 CEO 可以……dictate rules for the whole network; each community has its own moderation.
統一規則由個別社群自行訂立，每個社區都有自己嘅內容管理。
By late 2022, Mastodon’s user base exploded from a few hundred thousand to over 2 million active users, illustrating a public appetite for platforms not controlled by one corporation. Similarly, Twitter’s co-founder Jack Dorsey has backed an initiative called Bluesky and the AT Protocol, aiming to create a decentralized social media protocol where users own their identities and can move their social graph between apps.
去到2022年尾，Mastodon 嘅用戶人數由幾十萬急升到超過200萬活躍用戶，反映出公眾對唔受單一企業控制平台有強烈需求。類似咁，Twitter 聯合創辦人 Jack Dorsey 支持咗一個叫 Bluesky 嘅計劃同埋 AT Protocol，目標係建立一個去中心化社交媒體協議，用戶可以掌握自己身份，亦可以喺唔同 app 之間轉移自己嘅社交圖譜。
There’s also Lens Protocol, a blockchain-based social network ecosystem where your profile and relationships are stored on-chain (Polygon blockchain), allowing different social apps to plug into the same user-owned social graph. While these are nascent, they show concrete moves toward decentralizing the social web. For messaging, the Matrix network (used by apps like Element) is providing decentralized end-to-end encrypted chat, which has been adopted even by the French government for internal communications as a self-hosted alternative to WhatsApp/Slack. Each of these examples – Mastodon, Bluesky, Lens, Matrix – is an experiment in giving users more control and portability in their online social lives, as opposed to the walled gardens of Facebook or Twitter.
仲有 Lens Protocol，一個以區塊鏈（Polygon blockchain）為基礎嘅社交網絡生態圈，將你嘅個人檔案同關係資料儲存係鏈上，唔同社交 app 可以連接同一個由用戶擁有嘅社交圖譜。雖然呢啲仍然好初步，但都係社交網絡去中心化嘅實質一步。訊息方面，Matrix 網絡（例如 Element app 用緊）提供咗去中心化、點對點加密嘅聊天，連法國政府都用佢嚟做內部通訊，自行託管，作為 WhatsApp／Slack 嘅替代品。以上例子（Mastodon、Bluesky、Lens、Matrix）都係一次次讓用戶喺網上社交生活有更多自主權及流通性嘅實驗，唔再好似 Facebook 或 Twitter 咁受限制。
Decentralization is also happening at the infrastructure level, often in less visible ways. 去中心化喺基礎設施層面一樣發生緊，只係有時未必咁明顯。
Filecoin, mentioned earlier, has partnered with organizations to store open data sets (like large public info archives) in a decentralized manner, ensuring they remain available even if any one host goes down.
早前提過嘅 Filecoin，已經同多個機構合作，以去中心化方式儲存開放數據集（例如大型公共資訊檔案庫），即使有個別主機掛咗，都可以保障資料持續存取。
Arweave has become popular for storing NFT metadata and even webpages “permanently” – when Wikipedia pages about censorship incidents or news articles are at risk of deletion, activists have stored snapshots on Arweave, which is designed to hold data for 200+ years by economic incentive.
Arweave 近年用嚟永久儲存 NFT metadata 甚至網頁都好受歡迎——例如有啲人擔心維基百科上關於審查事件或者新聞文章被刪，會將網頁快照存去 Arweave。呢個系統透過經濟誘因，目標係儲存資料超過兩百年。
In the realm of domain names, the Ethereum Name Service has over 3 million. eth names registered, including ones by major brands and public figures – hinting at a future where your universal username or website might be a decentralized domain. And consider Bitcoin itself: while not usually framed as “the decentralized web,” it is the original decentralized digital network in production, and in places like El Salvador or amidst financial crises elsewhere, Bitcoin has been used as an alternative financial rail when banking systems faltered. It’s a reminder that the decentralized web can empower not just technologists, but also ordinary people in very real ways – from maintaining access to funds during economic turmoil, to staying connected when traditional platforms fail or censor.
域名方面，以太坊名稱服務（ENS）已經有超過三百萬個 .eth 域名，仲包埋好多大品牌同公眾人物註冊，顯示未來你嘅通用用戶名或網站都有可能係去中心化域名。而講返 Bitcoin，本身可能未必畀人直接聯想到「去中心化網絡」，但佢就係最原始而且已經投入運作嘅去中心化數碼網絡。喺薩爾瓦多、某啲金融危機地區，Bitcoin 曾經用來做銀行系統失靈時嘅另類金融通道。咁提醒咗我哋，去中心化網絡唔只利於技術人，更加可以實質幫到普通人——由經濟動盪時點樣維持資金接觸，到傳統平台失效或者封鎖時點樣保持聯繫，都發揮到作用。
Crucially, big companies are not ignoring these trends.
最重要嘅係，大企業冇忽視呢啲潮流。
Many are hedging their bets by investing in Web3 or incorporating decentralized tech. For example, Coinbase (one of the largest crypto exchanges, inherently a centralized entity) launched Base, its own Layer-2 blockchain network, to help scale and encourage decentralized apps – and they’ve made clear it will become increasingly community-governed over time.
唔少企業已經分散投資，參與 Web3 或者納入去中心化技術。例如 Coinbase（全球最大加密貨幣交易所之一，本身都係中心化機構）推出咗 Base，一條 Layer-2 區塊鏈，幫助擴容之餘亦推動去中心化應用。而且 Coinbase 已公開表示，Base 會越嚟越多由社群管理。
Payment giants like PayPal have integrated support for crypto and even for Web3 identity standards (like allowing users to log in with a wallet).
支付巨頭例如 PayPal，已經支援加密貨幣，甚至開始兼容 Web3 身份標準（例如畀用戶用錢包登入）。
Cloudflare itself, interestingly, operates a distributed web gateway and has experimented with hosting some Ethereum and IPFS nodes on its network, as if acknowledging that the future might involve servicing decentralized content rather than just traditional websites. These real-world moves show a convergence: while startups and open-source communities drive the decentralized web from one side, some incumbents are also embracing elements of it, bringing hybrid solutions to users now.
有趣嘅係 Cloudflare 本身都有做分布式網關，仲試過喺佢網絡上託管以太坊同 IPFS 一啲節點，似乎承認咗未來有可能唔淨係服務傳統網站，仲會兼顧去中心化內容。呢啲舉動反映一個融合——新創同開源社群推動去中心化網絡之餘，部份傳統巨企都開始接納相關元素，為用戶帶來混合式方案。
Will the Decentralized Web Go Mainstream? – Challenges and Outlook
去中心化網絡會唔會成為主流？——挑戰同展望
With so much momentum and hype, a natural question is: when (and if) will the decentralized web “conquer the world”? Is it destined to be the new norm, or will it remain a niche layer of the internet used mainly by enthusiasts?
咁大勢所趨，大家都會問——去中心化網絡究竟幾時（甚至會唔會）一統天下？係咪註定成為新常態？定只會留喺一小撮發燒友網絡界面？
The truth is, a fully decentralized web will likely arrive gradually rather than in a sudden takeover, and it faces serious obstacles along the way.
事實係，完全去中心化網絡更可能係慢慢滲透，而唔係一夜天下。中間一定會遇到唔少重大障礙。
In the optimistic view, we are on the cusp of Web3’s breakout moment. Venture capital funding, developer talent, and user interest in decentralized platforms have all ballooned in recent years. The technology is maturing – for instance, upgrades like Ethereum’s recent improvements and the rise of Layer-2 networks have dramatically increased capacity and lowered fees, making blockchain transactions faster and cheaper than they were a few years ago. Dozens of promising new projects are launching in various industries, from decentralized music streaming to Web3 gaming and metaverse worlds where users own in-game assets. Some Web2 giants are also integrating Web3 features (Twitter experimented with NFT profile pictures and crypto tipping; Instagram piloted digital collectibles).
樂觀看，Web3 已經接近爆發點。呢幾年，創投資金、開發者同用戶對去中心化平台興趣都飛升。技術亦都成熟咗，例如以太坊升級同 Layer-2 網絡崛起，令容量大增、手續費大減，區塊鏈交易變得比幾年前快又平。唔同行業都有新項目湧現，由去中心化音樂串流、Web3 遊戲到元宇宙入面用戶擁有資產。Web2 大企都試緊 Web3 新功能（Twitter 試過 NFT 頭像同加密小費、Instagram 都做過數位收藏品）。
These all suggest a future where average users might use decentralized web features without even realizing – e.g., your favorite game might run partly on a blockchain, or your digital wallet might replace the way you log into sites.
呢啲都預視咗未來普羅大眾可能用緊去中心化功能而唔自知——譬如你鍾意玩嘅遊戲部分運作喺區塊鏈，又或者你個電子錢包會取代以前嘅登入方式。
Yet, even proponents admit that mass adoption could take time – likely measured in years or even decades, not months. 不過，就連支持者都承認，要大規模普及可能要好多年，甚至要數十年，而唔係幾個月咁快。
The user experience challenge remains a top barrier. For the decentralized web to conquer the mainstream, it has to be as easy and reliable as the current web. That means a grandma should be able to use a Web3 social network or payment app without confusion or fear of losing her data. We’re not there yet. As one industry report noted, “Web3 UX remains significantly inferior to Web2 in 2025 due to challenges like complex onboarding and technical jargon.” Wallet addresses are long strings of characters; interacting with smart contracts can involve scary pop-up warnings; and concepts like “signing a transaction” or “gas fees” are foreign to non-tech users. Until these rough edges are smoothed out with clever design – possibly to the point where the crypto or P2P underpinnings are completely hidden – many people will simply stick with what’s familiar.
用戶體驗問題始終係最大障礙。要打入主流，去中心化網絡比現有網絡要一樣易用夠穩陣。即係街坊嬸嬸用 Web3 社交網絡或者付款 app 都無難度，唔怕亂用會無咗資料。事實上，現時未做到。就連業界報告都有提：「2025 年 Web3 用戶體驗仍然大大落後於 Web2，因為 onboarding 複雜、術語太多。」錢包地址係一大串字元，搞智能合約成日彈出警告，好多「簽名交易」、「Gas 費」呢啲概念，對唔太識科技嘅人嚟講好陌生。除非設計真係聰明到完全隱藏底層運作，好多人都會寧願揀返舒服習慣嘅東西。
The good news is developers are well aware of this, and efforts like simplified wallet recovery, human-readable addresses, and seamless integration into browsers and phones are actively underway.
好消息係開發者已經知道有咩問題，積極搞緊簡易錢包恢復、人性化短地址、同瀏覽器／手機無縫整合等改進。
Regulation and politics also loom large.
監管同政治都會帶嚟巨大影響。
The coming years will likely see intense debates and power struggles over decentralization. From a government perspective, a fully decentralized web is both tempting and threatening. On one hand, decentralization can bolster national resilience (no single foreign company controlling your country’s digital infrastructure) and drive innovation and competition. On the other hand, it complicates oversight – how do you enforce laws on a network with no headquarters, or tax transactions in a system like DeFi? Already, we’ve seen regulators grappling with crypto: some jurisdictions embrace it, others crack down hard.
未來幾年好大機會見到圍繞去中心化有激烈討論同角力。政府層面，去中心化網絡又吸引又危險——一方面可以讓國家唔受單一外國企業控制網絡基建、鼓勵創新和競爭；但另一方面，無總部嘅網絡好難監管——點樣執法？點徵稅（例如 DeFi）？全世界都已經見到監管機構各自有唔同做法，有啲地區擁抱，有啲堅決打壓。
The European Union’s new MiCA regulation is an attempt to set comprehensive rules for crypto-assets, and could provide a clearer legal path for Web3 businesses in Europe. In the US, however, multiple agencies (SEC, CFTC, Treasury, state regulators) are issuing sometimes conflicting guidance , creating uncertainty that could hamper decentralized projects or drive them offshore. China, notably, has banned cryptocurrency trading and mining outright, which puts a damper on some aspects of Web3 there (though they explore state-controlled digital alternatives). Big corporations might also resist or co-opt decentralization.
歐盟最新 MiCA 規例就試圖為加密資產定一套完整規則，令歐洲地區嘅 Web3 企業有清晰法律路向。美國就複雜啲，證監會、期貨委員會、財政部、州監管部門出咗唔同甚至互相矛盾指引，令法規走向好不明朗，可能會拖慢去中心化項目發展，或者迫佢哋搬去海外。中國就直接全面禁咗加密貨幣交易同挖礦，某啲 Web3 面向被壓抑（雖然緊密推動官方數字化替代）。至於大企業，本身可能會抗拒甚至「招安」去中心化。
After all, if the decentralized web truly flourished, companies like Google or Meta could see their dominance erode.
畢竟，如果去中心化網絡真係興旺起嚟，Google 或 Meta 呢啲大公司都會面臨地位被動搖嘅風險。
It wouldn’t be surprising to see incumbents lobbying for regulations that favor their semi-centralized versions of these technologies, or trying to influence open-source projects from within.
好正常會見到巨企遊說出有利自己半中心化技術嘅規管，或者滲透開源項目內部影響發展。
Another challenge is scaling community governance and preventing consolidation.
另一個難題係點樣擴展社群治理，防止權力再度集中。
Even if the tech works, will the decentralized web truly be decentralized in practice? 即使技術做到，實際運作落去會唔會真係去中心化？
There’s a risk that, as networks grow, control recentralizes in subtle ways – for example, if only a few big players can afford to run massive blockchain nodes or amass huge voting power in DAOs, they might start to wield outsized influence (much like mining pools did in Bitcoin’s early days, or how a few validator companies dominate some newer blockchains). The community will need to stay vigilant to ensure no single actor or cartel can quietly take over critical infrastructure. This is partly a social challenge: it requires aligning incentives and perhaps accepting some inefficiencies to keep things sufficiently distributed. It’s worth noting that even Tim Berners-Lee, who strongly advocates for a re-decentralized web, has chosen approaches (with his Solid project) that don’t rely on public blockchains, partly out of concern for issues like these and the commercial hijacking of Web3 buzz.
隨住網絡擴大，權力有機會以較隱晦方式集結——譬如淨係得幾間大機構可以負擔運行巨型節點、或者 DAOs 入面積聚大量投票權，就可能慢慢形成壟斷（好似早期 Bitcoin 礦池壟斷、或者新型區塊鏈畀少數驗證機構控制）。呢方面社群要時刻警惕，唔畀單一玩家或集團悄悄掌控到基礎設施。呢個其實都係社會問題，要靠誘因協調，可能都要接受某程度嘅低效率，確保夠分散。值得一提，Tim Berners-Lee （強力提倡網絡去中心化）做 Solid 項目時都避開咗公有區塊鏈，部分因為擔心呢啲情況同埋 Web3 概念被商業炒作劫持。
So, will the decentralized web conquer the world?
咁，去中心化網絡會統治世界嗎？
It may eventually weave itself into the fabric of everyday life, but likely in a hybrid form. We might see a future where many mainstream apps quietly use decentralized backends for certain features (like storing user data encrypted client-side, or settling transactions on a blockchain for transparency), even if the average user isn’t consciously “on Web3”. Fully peer-to-peer versions of services will exist alongside centralized ones, and users will gravitate to whatever offers the best experience and value. If the decentralized options prove more reliable (no outages), more empowering (users earn value, not just corporations), and sufficiently easy to use, they could indeed displace the old guard in various domains. But expect a period of co-existence: for example, a
終極嚟講，佢可能會以混合型式慢慢融入生活。我哋會見到好多主流 app 私底下已經用咗去中心化 backend 做某啲功能（例如用戶資料用端對端加密儲存、交易透明化落鏈結算），但一般用戶未必覺得自己係「用緊 Web3」。各種服務會有全點對點版本同中心化版並存，用戶自然揀最合心意同最抵嘅。只要去中心化方案證明更穩定（唔怕斷線）、更賦權（用戶賺益唔只平台），又真係夠易用，就有可能喺唔同範疇取代舊有方案。不過可以預期有一段長時期共存——例如……decentralized Twitter alternative might not outright kill Twitter, but it could push Twitter to change or could thrive parallel to it with its own user community.
一個去中心化嘅 Twitter 替代品未必會徹底摧毀 Twitter，但有可能迫使 Twitter 作出改變，或者喺自家用戶社群之中同 Twitter 並行發展。
The forces standing in the way are formidable – entrenched corporate interests, governments wary of losing control, technical hurdles, and plain old inertia and skepticism.
阻礙呢個發展嘅力量好強大——現有大企業利益、擔心失控嘅政府、技術障礙，仲有人類嗰種慣性同懷疑態度。
Many people like their convenient, curated web services and aren’t actively searching for an alternative. Bridging that gap will require killer apps that offer something tangibly better than what exists, not just something more principled in theory. It might also require crises that expose the weaknesses of centralized systems (much as the Cloudflare outage did, or data breaches have done) to jolt public opinion. In the end, a fully decentralized web is as much a social revolution as a technical one, touching on questions of who owns the internet and how we want our digital society to function. Those questions will not be settled overnight.
好多用戶都好鐘意現有嘅方便、被精心打理嘅網絡服務，並唔會主動去搵替代品。要縮窄呢個差距，唔單止要有殺手級應用，真係做到比現有服務更好，而唔只係理論上更有原則。仲有可能需要一啲危機，揭示中心化系統嘅弱點（好似 Cloudflare 停擺事件、又或者資料外洩），先至會令社會輿論有轉變。最終，要成就一個完全去中心化嘅網絡，不單止係技術革命，仲係社會革命，會觸及到「到底互聯網屬於邊個」同埋「我哋想數碼社會點運作」等等問題。呢啲問題，唔會一夜間有答案。
What’s certain is that the genie is out of the bottle.
可以肯定嘅係，潘朵拉盒子已經打開咗。
The innovations driving decentralization are unlikely to disappear; they have captured too many imaginations and solved too many problems.
推動去中心化嘅創新好難會消失，因為佢哋已經吸引咗太多人嘅想像，亦解決咗唔少問題。
Big players may slow it down or shape it, but even some of them are embracing parts of it. We may well look back a decade from now and marvel at how much more control individuals have over their digital lives – owning their data, their money, their online communities – without having to trust giant conglomerates.
大企業可能會拖慢或者主導變化，但其實佢哋當中都有啲已經開始接受某啲去中心化元素。十年之後回望，我哋好可能會驚訝發現，個人對自己嘅數碼生活擁有幾多自主權：管控自己嘅數據、資金、網上社群，唔需要再完全信任大型企業。
Or we might see a more moderate outcome, where decentralization underpins critical parts of the internet (like identity, finance, and content storage), making the whole ecosystem more robust and fair, even as certain applications remain centralized for convenience or compliance. The most likely scenario is a web that is more decentralized than today’s, but not entirely anarchic: a middle ground where decentralized and centralized systems interoperate, and users can choose the level of control or trust they prefer.
又或者，結果會溫和啲，去中心化只係負責支撐網絡上一啲關鍵部分（例如身份認證、金融、內容存儲），令整個生態更穩陣同公平；同時有啲應用為咗方便或符合法規，依然會保持中心化。最有可能出現嘅情況係，將來嘅網絡會比今日分散得多，但未至於完全無秩序：去中心化同中心化系統互相配合，用戶可以揀自己想要幾多控制權或者信任層級。
Final Thoughts
總結
The push for a decentralized web is, at its core, a push to reshape the power dynamics of the internet.
推動去中心化網絡，最根本就係要重新塑造互聯網上嘅權力結構。
The events of recent years – from infrastructure outages knocking major websites offline, to controversies over data privacy and censorship on big platforms – have laid bare the vulnerabilities of an overly centralized online world. The vision of Web3 offers an alternative: an internet that remains consistently available, that treats users not as products but as stakeholders, and that upholds the original promise of the web as an open, democratic space for information and interaction. It’s an ambitious vision, bordering on utopian, yet grounded in real technologies that are already coming online.
近年發生嘅事——由網絡基礎設施崩潰搞到主要網站停擺，到大型平台關於數據私隱同審查嘅爭議——已經赤裸裸咁顯示咗過度中心化嘅網絡世界係幾咁脆弱。Web3 嘅願景，提供咗一個唔同嘅選擇：一個持續可用嘅互聯網，將用戶視為持份者，而唔係商品，同時實現網絡原本開放、民主、資訊交流嘅承諾。呢個願景好大膽，有啲接近烏托邦，但其實已經開始落實喺一啲真正用得著嘅技術上面。