马克吐温曾经说过：“历史不会重演，但它确实会押韵”。1994年，《连线》杂志的主编凯文·凯利著就了《失控：机器、社会与经济的新生物学》一书，这是关于社会进化、特别是互联网发展的“先知预言”，后来成为《黑客帝国》主演们的必读物。25年后，八维资本试图捕捉互联网和区块链革命之间的相似之处，以帮助人们更好地了解技术生命周期和区块链行业的未来。
文章系列，敬请期待：
互联网与区块链革命之：
part1:早期的成功产品
part2:颠覆性公司的起源
part3:早期的挑战
part4:新概念，估值和时机
part5:我们是在1994年吗？ 下一步该怎么办？
互联网与区块链革命：早期的挑战（第3部分）
作者：Remi Gai @ 八维资本
编译：八维研究院
八维研究院原创，转载请注明出处
本文是整个系列的第三部分：
在新兴技术的早期阶段，可扩展性，成本和市场教育方面经常存在巨大挑战，这限制了突破性应用程序的开发和大规模采用。 然而，这些挑战通常随着时间的推移而得到解决。我们目睹了互联网的发展，同样期待在区块链中看到类似的进步。
可扩展性和成本：
2017年CryptoKitties发布后的拥堵的以太坊交易
尽管目前的互联网连接速度很快，但在早期阶段，互联网的可扩展性不高，成本却很高昂。 1996年8月7日， AOL（美国在线） 的互联网服务因承载不了互联网用户数而崩溃，并且过了19个小时都没有修复。 在那时，每天都有越来越多的美国人开始上网，而AOL是最大的互联网服务提供商，其他的竞争对手包括Prodigy，CompuServe和MSN 等。互联网本身并没有崩溃，但访问它的能力暂时支撑不住了，对于许多已经开始在生活中依赖网络的人们来说 ，这是一个大问题 。
这与2017年12月在区块链世界中发生的情况类似。CryptoKitties在以太坊上增加了链上交易量，堵塞了网络。大多数用户都无法在区块链上完成交易。除非他们愿意出价高额的Gas费，这在社区里引发了很多混乱和困扰。
以太网网络的堵塞一方面证明了到目前位置区块链技术在更复杂的情景下非常缺乏可扩展性，另一方面，在互联网和区块链技术的早期，网络堵塞和可扩展性不足也证明了这些技术的重要性和潜力。在未来的几年内，我们亟需开发出更好的可扩展性解决方案，来支持不断增长的用户群和需求。
美国互联网和无线移动通信的平均速度
在互联网初期，打造一个初创企业需要高额的网络费用和成本，这与区块链行业的现状也是类似的。那个时候，上网的过程需要一个调制解调器（modem）和一条电话线，所有的在线服务都需要维护一个本地modem网络，以供人们拨入。当时最受欢迎的在线服务- AOL美国在线，每月收取9.95美元，可以无限制访问5小时，而每增加一小时的费用为2.95美元，以今天的标准看，这是非常受限的服务。然而，随着时间的推移，与DSL的宽带连接最终实现，从而提高了互联网速度，到2014年，互联网速度比1999年快了200倍，便宜了90%。同样，我们也见证了无线移动通信速度在过去几年中的增长，从2G技术的低于1Mbps到4G技术的25Mbps，使新的服务通过移动电话（如音乐和视频流）变得可行和实用。随着时间的推移，围绕可扩展性和成本的限制逐渐消除，更复杂的和带宽更密集的应用程序变得实用和可行。
不同项目的交易速度
在区块链领域，我们还处于技术开发周期的早期阶段，我们面对着可扩展性和成本的挑战，这大大阻碍了更复杂的应用场景和大型用户群的使用。低可扩展性意味着开发人员或用户必须支付更高的费用才能超过队列中的其他事务，以减少等待时间。截至2019年2月， 以太坊的平均交易费用约为0.13美元 ，其中更复杂的DAPP（如链上游戏）将很快收取高昂的交易费用。虽然并非每个行动都需要在链上发生，但我们仍迫切需要增加区块链的可扩展性，以支持不断增长的DAPP数量，以及支撑在同一个分布式账本上的所有日常交易（VISA每秒处理24000个交易）。许多项目正致力于通过不同方式扩展区块链的交易速度 ，并在可扩展性，安全性和分布式之间进行可变的权衡。
随着最近EOS和TRON（分别为4,000和750 TPS）主网的推出，过去几个月内我们看到更复杂性能的DAPPs的数量有所增加， 我们期待，随着区块链可扩展性的延展，这一良好的趋势将延续下去。 总的来说，业内正在探索不同的扩展解决方案，如新颖的数据结构（Sharding，Tangle，DAG，Coda等），第2层解决方案（Sidechains，off-cha ins，散列时间锁等），以及更有效的一致性算法（ POS，DPOS，Casper，Avalanche，Hashgraph等）。这些解决方案将逐步增加带宽并降低区块链交易的成本，从而实现新的使用场景和服务 。
1999 - 2010年启动创业公司的成本降低
此外，我们预计启动区块链初创公司的成本也会随着时间的推移而降低。在互联网革命期间，创办一家初创公司的平均成本从1999年的500万美元大幅下降到2005年的50万美元，再到2010年的5万美元。
第一波成本降低的原因是由于开源Open Source （这意味着不再需要UNIX，Web服务器和Oracle数据库的许可证）和并行计算Parallel Computing（这意味着不需要购买昂贵的Sun服务器和EMC存储）。 然后，第二波成本下降是由于云计算Cloud Computing的成熟 ，由AWS推广开来，云计算提供了网络存储（S3），处理能力（EC2）和根据云上的流量进行自动扩展或缩放的能力。
根据Gartner的说法，部署一个概念验证的企业级区块链的起始成本为275,000美元，最高可达数百万美元。而在公有区块链（如以太坊，Neo，EOS等）上搭建简单DAPPs的部署和交易费用可能会花费$30-85k，而且目前的区块链人才短缺，人力成本也要花费很多钱。
与互联网的发展类似，我们希望，在启动区块链项目时可以不断降低成本。未来，区块链交易变得更具可扩展性和更便宜，企业级的更多全栈平台（BaaS 托管 - Kaleido，AWS，Azure，Oracle等），开发平台（NoOps - Esprezzo），中间件（Omnitude）和开发者工具（Mist，Geth，Remix等）正在出现，而更多的人才也将迁移到区块链领域来。
教育：
曾经，全世界50％的CD都有AOL标志
教育是新技术被应用和接纳，降低进入门槛，帮助“跨越鸿沟”（Geoffrey Moore创造的一个术语 ）的重要推动力 。 在认知鸿沟未被跨越之前 ，技术通常被视为一个异类，一旦真正跨越鸿沟，人们就会认为它是一种安全的消费品。
在互联网的初期，Jan Brandt被聘为AOL的营销副总裁，她的职责是扩大用户群。在市场调查中 ，她意识到人们不知道该如何使用计算机，因为“有人拿起电脑鼠标开始将它指向计算机，就像遥控器一样”，所以“启蒙”教育是很重要的。与其告知用户一种产品（比如电脑）胜过另一种产品（比如电视），还不如告诉他们到底怎么去使用电脑。
AOL实施的营销策略是免费发放试用磁盘CD给人们试用，结果效果出奇的好。人们开始注册并在线上绑定他们的信用卡。AOL对这一战略进行了大量的投资，使CD随处可见，在人们的邮箱里，在新电脑上，在杂志上，通过电影租赁等方式分发出去。因此，在全球生产的CD中，有50%都有美国在线的标志，他们一举超越了竞争对手，并在几年间将其用户群从20万增加到2500万。总的来说，AOL的营销活动是一个很好的例子，它通过降低早期用户中技术水平较低的人群的进入门槛来推动产品的大规模应用。
从Coinbase学习并获取0x程序
为了推动加密世界的发展，我们已经看到了许多关于教育下一代加密技术和区块链技术用户的举措，其中包括由 德勤 或 IBM 举办的大型会议，如 Consensus共识大会，企业教育， ConSensys Academy 等开发人员教育 和大学教育，如 ConSensys Academy区块链教育网络 。
许多公司还试图通过活动和营销活动赠送Token或冷钱包。最近， Coinbase推出了一项 学习和赚取加密程序 ，为完成他们的教育课程的用户提供加密货币奖励。目前，与传统支付解决方案（如 Venmo 或 Paypal ）相比，区块链的使用仍有很多缺陷和摩擦，并要求用户具备一些基础知识（私人/公共密钥，Gas费/交易费等），更不用说与密钥管理和交易相关的不可回撤的风险。我们仍然处于行业的早期阶段，全球采用率不到1％，对普通大众进行关于加密货币，区块链和密钥管理的培训对于行业的发展至关重要。 像 AOL 和 Coinbase 这样的大公司可以通过赠送免费试用的CD和加密货币来给用户提供额外的激励，这将有助于降低用户的进入壁垒并推动技术的普及。
总的来说，不可否认的是，过去几年，区块链行业已经在解决可扩展性、成本和市场教育方面取得了重大进展。虽然我们还处于这一技术周期的早期，但我们相信，区块链将以与互联网类似的方式慢慢成熟，并逐渐变得更具可扩展性、更经济实惠和更实用，最终，区块链技术和产品将会被普罗大众广泛的采用。
后记
马克吐温曾经说过“历史不会重演，但它确实会押韵”。 我们试图在互联网和区块链革命之间勾勒出一些相似之处，以帮助公众更好地了解技术的生命周期。请留下您的想法和评论，希望本系列文章将为行业提供一些有价值的观点。
资料来源：
“互联网如何发生” ，由 Brian McCullough 撰写 。
https://www.linkedin.com/pulse/reply-unending-chasm-how-survive-geoffrey-moore/
https://www.coinbase.com/earn/0x
https://bothsidesofthetable.com/why-has-seed-investing-declined-and-what-does-this-mean-for-the-future-6a9572357130
https://bothsidesofthetable.com/a-deep-dive-into-what-has-really-changed-in-venture-capital-f5d225f7f8
https://howmuch.net/articles/crypto-transaction-speeds-compared
https://ylv.io/how-much-does-it-costs-to-run-dapp-in-2018/
https://bitinfocharts.com/comparison/ethereum-transactionfees.html
英文原文：
Internet vs BlockchainRevolution: Challenges in the Early Days (Part 3)
This article is part of the Internet vs Blockchain Revolution Series. If you areinterested in reading the other articles.
In the early stage of emergent technologies, there are often challenges around scalability, cost,and education, limiting the development of breakthrough applications and massadoption. However, these challenges are generally solved over time as we havewitnessed with the evolution of the Internet and we should expect a similar technological progression in Blockchain.
Scalability and Cost:
Pending Ethereum Transactionsafter CryptoKitties’ Release in 2017
Despite today’s fast Internet connection, the early days of the Internet faced challenges in terms of scalability and cost. On August 7, 1996, AOL’s Internet service failed from supporting a high volume of Internet users and went down for nineteen hours. At that point, Americans were increasingly living their online lives every day, and AOL was the country’s largest internet service provider, amongother competitors such as Prodigy, CompuServe and MSN. The Internet itself hasn’t crashed but the ability to access it temporarily stopped, and it was a big deal for many who started to get used to surfing the web in their daily life. This is similar to what happened in theblockchain world in December 2017, in which the of CryptoKitties spiked the on-chain transaction volume on Ethereum and clogged the network. Most users couldn’t get their transactions through on the blockchain unless they were willing to bid a high amount of gas fees, which caused a lot of confusion and distress in the community. While the Ethereum network jam made many of us realize the current lack of scalability of blockchain technology in servicing more complex use cases, both of these failures in scalability during the early days of Internet and Blockchain showed a validation of how important these technologies had become in a few years for the early adopters and the need for better scalability solutions to support the growing user base and demand.
Average Internet and WirelessMobile Telecommunication Speed in the US
The high network fees and costs for launching a startup during the early days of the Internet and the current state of the Blockchain industry is also similar. Back then,the process of getting online required a modem and a phone line, in which all the online services maintained a network of local modems for people to dial in.The most popular online service at the time, AOL, charged $9.95/month for 5 hours of unlimited access with each additional hour costing $2.95, which by today’s standard, was very restricting.However over time, the broadband connections with DSL eventually came, scaling up the Internet speed, and by 2014,the Internet speed was 200 times faster and 90% cheaper compared to 1999. Similarly, we have also witnessed an increase of wireless mobile telecommunication speed over the years, from less than 1Mbps with 2G technology to 25Mbps with 4G technology, enabling new services to be viable and practical through the mobile phone, such as music and video streaming. Over time, the limitations around the scalability and cost were gradually removed, allowing for more complex and bandwidth-intensive applications to be practical and viable.
Transaction Speed of Various Blockchain Projects
In the blockchain world, we are still early in the development cycle of the technology, in which the challenges around scalability and cost are preventing the practicality of more complex use cases and large user base. Low scalability means that the developers or users would have to pay higher fees to outbid the other transactions in the queue to reduce wait time. As of February 2019, the average transactions fees are roughly $0.13 for Ethereum, in which more complex dapps, such as on-chain games, would quickly rack up high transaction fees. Although not every action needs to live on-chain, there is still an urgent demand for blockchain to be scalable to support the growing number of dapps, and all the day to day transactions (Visa processes 24,000 transactions per second) living on the sameledger. Numerous projects are working on different means to scale the transaction speed of blockchain, with variable trade-offs between scalability,security, and decentralization. With the recent mainnet launches of EOS andTRON (4,000 and 750 TPS respectively), we have seen an increase of more complexdapps being released in the past couple months, and we expect this trend tocontinue as more scalable blockchains are being deployed in the space. Overall,the different scaling solutions being explored such as novel data structures(Sharding, Tangle, DAG, Coda etc.), layer 2 solutions (Sidechains, off-chains,hashed-time locks etc.), and more efficient consensus algorithms (POS, DPOS,Casper, Avalanche, Hashgraph etc.) will gradually increase the bandwidth and reduce the cost for transactions on the blockchain, enabling new use cases andservices over the years.
The decrease of the cost of launching a Startup from1999-2010
Additionally, we expect that the cost of launching blockchain startups will also decrease overtime. During the Internet revolution, the average cost of starting a startupalso drastically reduced over the years, from $5m in 1999 to $500k in 2005, and$50k in 2010. The first wave of cost reduction was due to the appearance of open source (which meant no license for UNIX, web servers and Oracle databases) and horizontal computing (which meant no need to buy expensive Sun servers EMCstorage). Then, the second wave of cost decrease was due to the maturation of cloud computing, popularized by AWS,providing web storage (S3), processing power (EC2) and the ability to scale upor down depending on traffic (auto-scaling) on the cloud. According to Gartner,deploying enterprise blockchain proof-of-concepts has a starting cost of$275,000 and can go up to several millions of dollars. Depending on theblockchain (Ethereum, Neo, EOS, etc), the deployment and transaction fees of simple dapps on public blockchains can cost you $30-85k (interesting breakdown here), and current blockchain talents are in shortage and expensive ($150k on average).Similar to the evolution of the Internet, we also expect a cost reduction inlaunching blockchain projects as the blockchain transaction speed becomes morescalable and cheaper, more full-stack platforms on the enterprise side (BaaShosting - Kaleido, AWS, Azure, Oracle, etc), development platform (NoOps -Esprezzo), middlewares (Omnitude) and developer tools (Mist, Geth, Remix, etc)are being released, and more talents migrating into the blockchain space.
Education:
At one point, 50% of the CDs produced worldwide had an AOL logo on it
Education is an essential driver for the adoption of new technologies, lowering the barrier to entry, and helping crossing the chasm (a term created by Geoffrey Moore). Before the chasm,technology is usually treated as a foreign object, and once it has trulycrossed, it is perceived as safe to consume. During the early days of the Internet, Jan Brandt from AOL was hired as VP of marketing to growits user base. During market research, she realized that people didn’t know howto use the computer as “someone took a computer mouse and started pointing it at the computer like the remote control”, so it was important to go back to the basics. Selling consumers on the virtue of one online service over another wasnot as important as educating consumers on what an online service was. The marketing strategy implemented by AOLwas to send out trial disks and CDs for people to try out, and the resultsproved to be extraordinary, with people signing up and providing their creditcards. AOL re-invested so much into the strategy that CDs were found everywhere, in people’s mailboxes, on new computers, in magazines, given away with movie rentals, etc. so that at one point, 50% of the CDs produced worldwide had the AOL logo, leap frogging the competitors and growing its user basefrom 200k to 25M clients over the years. Overall, AOL’s campaign is a good example of a successful initiative indriving adoption by actively lowering the barrier of entry for the less tech-savvy crowd of early users.
Learn and Earn 0x program fromCoinbase
To draw a parallel with the cryptocurrency world, we have seen many initiatives in educating the next generation of crypto and blockchain adopters, with large conferences, suchas Consensus, corporate education,taken by Deloitte or IBM, developer education, such as ConSensys Academy, and universityeducation, such as Blockchain EducationNetwork. Many companies are also attempting to in give out tokens or cold wallets for free through events and marketing campaigns. Recently, Coinbase launched a learn and earn crypto program,offering cryptocurrency rewards for completing their educational courses.Currently, interacting with blockchain still has a lot of frictions, and require you to have some foundation knowledge (private/public keys,gas/transaction fees etc.) compared to traditional payment solutions such as Venmo, or Paypal, not to mention the irreversible risks that come with keymanagement and transactions. We are still early in the industry with less than 1% of worldwide adoption, and educating the general population about cryptocurrency, blockchain and key management will be very important for the industry to grow. Large companies such as AOL and Coinbase are well positioned to offer extra incentives through giveaway CD trials and cryptocurrencies to actively help lower the barrier of entry and drive further adoption.
Overall, it’s undeniable that the Blockchain industry has been making significant progress towards solving the challenges around scalability, cost, and education over the past few years. Although we are still early in this technological cycle, we are confident that Blockchain will mature in a similar fashion as the Internet,gradually becoming more scalable, affordable and practical for mass adoption.
-----
The Internet Revolution facts are based on thebook “How the Internet Happened”,written by Brian McCullough. MarkTwain once said “History doesn’t repeat itself, but it does rhyme”. We areattempting to draw some similarities between the Internet and BlockchainRevolutions, to help entrepreneurs and investors better understandtechnological life cycles.
Sources:
Book “Howthe Internet Happened”, written by BrianMcCullough.
https://www.linkedin.com/pulse/reply-unending-chasm-how-survive-geoffrey-moore/
https://www.coinbase.com/earn/0x
https://bothsidesofthetable.com/why-has-seed-investing-declined-and-what-does-this-mean-for-the-future-6a9572357130
https://bothsidesofthetable.com/a-deep-dive-into-what-has-really-changed-in-venture-capital-f5d225f7f8
https://howmuch.net/articles/crypto-transaction-speeds-compared
https://ylv.io/how-much-does-it-costs-to-run-dapp-in-2018/
https://bitinfocharts.com/comparison/ethereum-transactionfees.html
八维研究院出品的其他研究
跨境金融科技平台Pandaq正式发布｜八维资本 X TrueChain
《互联网vs区块链革命》系列之一：早期的成功产品
《互联网vs区块链革命》系列之二：颠覆性公司的起源
硬核技术：证券型通证需要专属的区块链吗?
企业级区块链：过去、现在和未来
Medium:https://medium.com/@8DecimalCapital
八维资本（8 Decimal Capital）是一家跨境的国际专业区块链投资机构，管理美元和Token两支基金，总资金规模超过六千万美金。团队合伙人来自一线VC、PE，基金顾问包括前瑞士银行中国区CEO陈庆，前红杉中国创始人合伙人张帆。八维专注于区块链赛道投资，同时对上下游产业资源进行战略布局。八维投资了底层技术、隐私、存储、Security Token、稳定币等赛道的顶尖项目，如EOS、Icon、IoTeX、Zilliqa、Bluzelle、IRIS Network等，同时布局了交易所、媒体、咨询、技术社区等战略协同资源，如0x、Fcoin、Kyber、Bibox、Mobilecoin、Dorahacks、星球日报、币世界、WXY等。在美国Token Metrics榜单中，八维投资回报位列50家国际基金的前三位，也被36氪、清华x-lab等机构评为中国Top10区块链投资机构。
