ICYMI | 万一你错过了| 당신이 그것을 놓친 경우

The great big Spotify scam: Did a Bulgarian playlister swindle their way to a fortune on streaming service? – Music Business Worldwide – this also shows how topsy turvy the economics of Spotify are

China is quickly becoming the dominant force in startups | Quartz – makes you wonder about Silicon Valley

Opinion | The Tyranny of Convenience – The New York Times – Americans say they prize competition, a proliferation of choices, the little guy. Yet our taste for convenience begets more convenience, through a combination of the economics of scale and the power of habit. The easier it is to use Amazon, the more powerful Amazon becomes — and thus the easier it becomes to use Amazon. Convenience and monopoly seem to be natural bedfellows. – great article by Tim Wu

WPP Vows to Do Better After Weak Results, Nervous Outlook Send Shares Plunging – The New York Times – WPP plans to accelerate a programme to simplify the business by aligning digital systems, platforms and capabilities to provide bespoke teams for its clients as opposed to the different agencies that currently compete with each other to win contracts.

Smart homes and vegetable peelers — Benedict Evans – interesting starting point, but I think that there should be a second layer. Can the intelligence be local (like lighting sensors based on movement and presence in office buildings) or does it need cloud computing? Why can’t smart lightbulbs be at the edge rather than in the cloud. Why does a Nest thermostat need to be in the cloud?

WeChat New Year Data Report 2018 – China Channel

WSJ City | Five signals sent by China’s Anbang takeover – Reining in big spenders (spending capital abroad in an untargeted manner), reduction of systemic financial risk, concern over complex short-term high-yielding wealth products

Wealthy Chinese Women Are Unique in APAC: Agility Research | Jing Daily – interesting dissonance between Hong Kong and Chinese high net worth consumers

Levi’s Invented A Laser-Wielding Robot That Makes Ethical Jeans | Fast Company – the laser and chemical free treatment remind me a lot of the work that Frontline Clothing in Hong Kong have been doing for years in association with their Chinese supply chain partners

Struggling Esprit to close more than 40 shops in Europe | South China Morning Post – it plans to close more than 40 “heavy loss-making” shops in “core” European countries, or make around a 10 to 15 per cent reduction in its controlled space in these countries

Amazon Has Officially Invaded The Advertising Industry | Forrester Research – the bit this misses is that consumers already use Amazon’s search page as a first port of call for things

Huawei distances itself from executive’s comments that rivals using politics to keep it out of US | South China Morning Post  – Huawei did not authorise Yu to make comments about the US on behalf of the company, and does not agree with his views, Chen said. Yu did not immediately respond to a text message seeking comment – Richard Yu is known for going off-piste with media

LittleThings online publisher shuts down, blames Facebook’s algorithm – Business Insider – not terribly surprising, one only had to look at the games companies that built their businesses on Facebook and got eviscerated

Burson Cohn & Wolfe – SixtySecondView – like any other business merger the focus will keep the eye off the ball at a time when the PR industry is seeing exceptionally low growth rates. I have friends and former colleagues on both sides of this in both Asia and Europe; so I hope it works out well.

Samsung says it’s going to stop pumping out features and start making devices good instead – BGR – “We developed mobile phones earlier than China, and we were obsessed with being the world’s first and industry’s first rather than thinking about how this innovation would be meaningful to consumers,” Koh said. “Being the first turns out to be meaningless today, and our strategy is to launch something that consumers believe meaningful and valuable at a right time.” – this reads like a slap in the face to Huawei’s approach on innovation and features

Tea Turns Up Temperature in Fight Against Coffee – WSJ – what tea misses is ritual

Daring Fireball: Berkshire Hathaway’s 2017 Annual Report (PDF) – they know how to play to small town audiences well

Five for Friday | 五日(星期五) | 금요일에 다섯 가지

Things that made my day this week:

A cheap facsimile of the classic Nokia 8110 managed to upstage the launch of of a range of premium Android phones at MWC.  Nostalgia is powerful, but I don’t think what’s going on here. I could see this as a weekend phone allowing consumers to wind down at the weekend and go cold turkey on the app economy.

nokia8110traditionalblack3 png-257014-original

I think that the model of a single form factor based on common reference designs is broken. Apple managed to elevate the build quality of all smartphones as contract manufacturers moved towards an armada of CNC machines and advanced manufacturing. But in the process, the common designs, common components and new baseline in product integrity has homogenised and commoditised Android handsets to such a decree that only scale and advertising budgets are differentiators.

Amazing selection of music from the sound tracks of classic kung fu movies over at Shaolin Chamber 36.

Roni Size talks about the music that influenced him

Frank Herbert talks about creating the science fiction epic Dune

digitalethnography | Field Note Painting Booklet – done by Xinyuan Wang, a UCL social anthroplogist in a lower tier Chinese city. Ms Wang wrote a really good monograph on social media in industrial China.

I like: Josh Tenebaum MIT lecture on building machines that see, learn and think like us

One of the better presentations on the current state of machine learning that I have seen. There is the interesting academic / commercial divide in terms of likely outcomes. Industry isn’t going to create that general purpose artificial intelligence that everyone is writing about. The academic research is very far way as well. In many respects the over promise and under-delivery would be familiar to AI research pioneer Marvin Minsky. Minsky lived through a number of AI booms and AI ‘winters’ when the hype was deflated.

Despite Tenebaum’s reality check on the industry we still see commercial AI concept films created to sell clients on the future promise of a general purpose AI. Agencies and technology companies doing this are only hastening the rise of another AI winter.

Social networks 10 years ago | 社交网络服务10年前 | 소셜 네트워크 서비스 10 년 전

Ten years ago, I was busy helping get a communications agency’s digital offering up and running. Social media was a thing. There was a mix of curiosity and fear in clients. A bit like the web before it, they felt that they needed to do something but often didn’t have sufficient momentum to turn this into action.

Now social media means Facebook/Instagram/Twitter/Snapchat and maybe LinkedIn depending on the client objectives. Back then it was much more diverse. Blogging was still powerful and I spent a lot of 2005 – 2007 helping Yahoo! and then agency clients develop social media and blogging guidelines.

The first panel that I ever spoke at was an event called Blogging4Business held at  the Marriott in Grosvenor Square in April 2007.

It was a time of tremendous growth. Facebook had gone from 28 million users in July 2007 (less than half the size of MySpace) to 200 million users by November 2008. MySpace doubled its user base in the same time but was still left in the dust by Facebook.

Here’s the top social platforms by unique worldwide visitors in November 2008 according to comScore.


Blogger – remember I said that blogging was big? Pyra Labs had launched one of the first dedicated blog publishing tools in 1999. By February 2003, it was acquired by Google. The second iteration of this blog was started on Blogger in March 2004. Blogger’s killer app was that it was much more flexible and scalable than DIY sites like Geocities and Tripod that came before it.  It developed a number of great features added around this time including allowing me to post via email. You’ve got to remember, network access was pretty poor outside the home.

Email was one of the few things that worked well on mobile devices. I started using Flickr as my image hosting which allowed me to send image attachments as well as my copy to my blog. An email could be written on the go, without internet connectivity. The next time that you connect to the internet, the email would sail into the ether and trigger a blog post, which was much more practical than the nascent app economy on PalmOS and Symbian at the time. I sent an email to publish this post using a Palm Treo 650, whilst waiting to fly back from San Francisco airport in August 2005. At the time I was going back and forth between Yahoo! Europe in London and global headquarters in Sunnyvale.

Blogger attracted 222 million unique users on a monthly basis.

Facebook – we all know it now as a dominant ever watchful social leviathan. Culturally Facebook represented a changing of the guard. Up until 2007, web 2.0 ethos had been about data portability. It meant I could easily move my photos from Flickr or my bookmarks from delicious.com. Part of this came from the hippy ethos that was baked into the Silicon Valley world view. Here’s what I was saying about Facebook back in 2008:

I don’t work with Facebook, I can only go on the way that the company presents itself and judge it on its actions. But from this I can make some deductions. Its terms and conditions particularly the ownership of any user data is much more onerous than the likes of Google and Yahoo!. With Yahoo! you grant them a non-exclusive license to your content; you can choose to remove the material when you want. With Facebook, they own your data period. This isn’t about putting their business on a legal footing but serving the audience up with a price on their head, and showing a lack of respect for their audience. And I haven’t even talked about Facebook’s privacy infringing marketing practices, of which Beacon was the latest high-profile example. As the saying goes ‘A fool knows the price of everything and the value of nothing’.

The company takes a Hotel California approach to APIs (your data can enter, but it can never leave). Add to this the control that Facebook is going to put on developers in 2008 – not exactly right neighbourly now is it? Even Apple bows to their influence and had to give developers an SDK (software development kit) for the iPhone. Also remember that developers are the kind of web influencers who can make and break a service: they lead people on and they can lead people elsewhere.

By 2008, Facebook was getting its ducks in a row to become an online advertising powerhouse. What we couldn’t see at the time was the way that Facebook missed the boat on mobile and had to consciously work to not fall off the wagon. Back in November 2008, Facebook had 200 million followers.

Facebook was defined in many respects by its rival MySpace. by 2008 they had been rivals over three years. 2008 marked the turning point when Facebook gained the upper hand and surged away. The roots of MySpace was in a digital marketing company eUniverse who had employees with accounts on the Friendster social network.

They decided that they could do a better social network. Friendster had struggled to scale and many times you couldn’t log in. So the bar was set pretty low. Being based in Los Angeles, MySpace used entertainment connections to have bands and celebrities as ‘tent pole’ users on the platform. Digital promotion was really taking off in the music industry so their timing was fortuitous. MySpace was also instrumental in the rise of freemium casual gaming. This drove one of the first booms on the Facebook platform: remember Farmville? And opened the door to freemium mobile game apps like Angry Birds and Candy Crush. A sale to News Corporation saw MySpace take your eye off the ball. In November 2008 MySpace had 126 million unique users, by 2015 MySpace had almost 1 billion active (and inactive) accounts.

WordPress – WordPress.com sprang out of Matt Mullenweg’s open source blogging platform project. WordPress.com offered a half way house between Blogger and having a self-hosted website. It has an annual subscription to contribute towards hosting and a better experience than Blogger. In November 2008 WordPress.com attracted 114 million unique users, WordPress.com claims that this number has now grown to 409 million unique users per month.

Windows Live Spaces was a weird chimera of a blogging platform, photo gallery, a Geocities style guest book and a social platform. You have a pretty good idea from just reading this how much of a mess it was in terms of its user experience. Facebook was pretty awful; Live Spaces made it look reasonably good. It didn’t help that it was originally launched as MSN Spaces back in 2004 and got caught up in the wider online rebrand that Microsoft did in 2006. In November 2008 it was attracting 87 million users per month (presumably due to integration between the Microsoft browser and Windows Live services). Microsoft shut it down in 2011.

Yahoo! Geocities was a pensioner in web service terms by 2008. Geocities was originally launched in 1994, it offered people like you and me the opportunity to create our own web pages with no technical skill. If you became a paid user you could increase the size of your website and get telephone technical support. People would build pages to share holiday or baby pictures in low resolution.


When I was in college, Geocities was useful as other students would publish essays and book reports online, that I could then reference. This was back when you could surf the web or discover content by following a webring. A webring was a pre-Google way of discovering quality subject content once you had landed on a relevant site. As the name implied a webring is a collection of (amateur) websites (usually around a common subject area) linked together in a circular structure. By 2008, Geocities was on its last legs but still drew in 69 million unique users. It was shut down in the US the following year. Geocities still lives on in Japan.

Flickr – I started using Flickr at the end of 2004, it allowed you to have a visual diary a la Instagram. In addition, it provides flexible image hosting. It still does the image hosting for this blog today and allows great searching through the use of labels and tags. Flickr drove access to creative commons visual content. There is a whole novels worth of material about how Yahoo! mismanaged the business and missed opportunities.

When I left Yahoo! in 2006, one of the last things that I worked on was the default installation of flickr on the Nokia N73 camera phone four and a half years before Instagram was even launched. Instagram is basically a less flexible, less community minded Flickr with filters that allowed millennials to convince themselves that their poorly taken snaps was art – a digital version of Polaroid or Lomography. In November 2008, the strain of being inside Yahoo! was starting to show, the founders had resigned in June 2008; but flickr welcomed 64 million unique users. Flickr turned 14 years old on February 10.

The first photograph that I took and posted on flickr for the Pirate Communications Christmas party. This happened right after we’d lost a large part of the BT business, which was the agency’s anchor client at the time.

Pirate Christmas party part one

hi5 – back at the beginning of 2007 there was some data that suggested the user growth at social network hi5 would put it ahead of MySpace, I can remember marketers getting excited about it for about six weeks. By November 2008, it had lost a lot of steam and Facebook tore ahead. hi5 still had a respectable 58 million unique users. hi5 was one of a number of social networks that seemed to be adopted in loyal pockets in developing countries.  The network pivoted into social games to try and ride the casual gaming boom and did some interesting work around game software development kits for the platform.

Orkut – is a classic tale of opportunity squandered. Orkut was ‘inspired’ by extranet group software called InCircle. Back in 2004 it looked like a dead cert. A Google backed social network that had been developed during an engineers 20 per cent time. At this time:

  • Silicon Valley was seen as making a positive difference
  • The internet was still largely idealistic
  • Google could do no wrong. They didn’t need to ‘do’ PR, the positive stories wrote themselves such as ‘Google spots Jesus in Peruvian sand dune‘ and the general media would be absorbed with every Google Doodle

There was no apparent reason why their venture into the social networking space shouldn’t do well. It featured integration of (at the time) great services including GTalk instant messaging and both YouTube and Google Video.  Orkut became a a community of over 300 million registered users.  The problem was that Orkut only took off in very localised areas, if you weren’t a Google employee, Brazilian or Indian you likely won’t have been an active Orkut user for long and this stymied its growth. I suspect that the reason for this was that the initial community was grown on the back of Google’s engineering base. There were groups of users in the Persian gulf states and Saudi Arabia but access was blocked due to groups like ‘Dubai Sex’.

Orkut had numerous security issues in 2006 and 2007. Profiles became infected with the MW.Orc worm which stole users banking details, user names and passwords. A second less malicious worm automatically made a user join a virus related community and then replicated itself on the users profile to spread further. Finally there were session management and authentication issues that posed particular risks for those people using cyber cafes. They allowed access to a users Gmail account. After ten years Google finally pulled the plug on Orkut. In November 2008 Orkut had 46 million unique users.

Six Apart – had it all. Photogenic husband and wife founders, great technology and bold business moves. Six Apart created Moveable Type which at one time was the go to blogging platform. They also  founded TypePad which was a wordpress.com analogue. Vox launched in 2006 was a simpler blogging experience – Tumblr was a good analogue. They bought French platform Blog and Loic Le Meur became their head of EMEA. Around the same time they bought LiveJournal. Vox was eventually closed down, but LiveJournal was sold for a profit. At the time, the smart money would have been on Six Apart to win over WordPress.com, in the end it turned out rather differently. Back in November 2008, Six Apart websites enjoyed 46 million unique users.

Baidu Space – Baidu Space was a social network that grew out of the search engine company.  It launched in 2006 and over 18 months it managed to grow to getting 40 million unique users by November 2008. It was closed in 2015 and all the content was transferred to Baidu Cloud file storage service.

Friendster – prior to its sale in 2009 Friendster had managed to clock up 115 million registered users, which was a minor miracle given how unstable it was and was virtually impossible to log into over a two year period. On a good day it was like using a website though a sluggish Citrix window. Twitter used to have a reputation for poor infrastructure which made the ‘fail whale’ iconic, but that was nothing on Friendster’s instrastructure.  They knew what they had to do , but due to management issues and technology issues couldn’t make it happen. It seems like a miracle that in November 2008 the site had 31 million unique users. Mark Zuckerberg is obscenely rich because Friendster was very unlucky.

56.com – was owned by Chinese internet company Sohu at the time. It was one of the most popular video sharing sites. This was back before Tencent came to dominate the video market alongside Tudou | Youku. It lost a good deal of its traction when it was shut down for over a month in June 2008. By November 2008 it had 29 million unique users per month. 56.com was sold to youth orientated social network Renren at the end of 2014.

Webs.com – is now owned by Vista Print and is surprisingly still going. It was launched in 2001 and was a step up from Tripod, Geocities and Angelfire in terms of allowing you to build and host web pages.  In November 2008 it attracted 24 million unique users. This was on the back fo massive growth from 2003 to 2007. It has now been overshadowed by Squarespace and WordPress.

Bebo was founded at the beginning of 2005. The founders(Michel and Xochi Birch) managed to built a user base amongst teens in the UK and Ireland before selling it on to Aol in March 2008 for $850 million. The Birch’s sold at the top of the market. Aol managed to sell on Bebo, but at a large loss. Eventually it went into bankruptcy administration. It is now a Windows programme that streams to Twitch. In November 2008 is had  24 million unique users.

Scribd – it is optimistic to call Scribed a social service but that’s the category that comScore listed it in. It was launched in 2007 and by November 2008 it had 23 million users. Over the years it has evolved into a subscription service for books, comics and audio books.

Lycos Tripod – Tripod was a rival of Geocities, that has evolved into a small business hosting business similar to Square Space. It was launched in 1995 focusing on college age students and was acquired by Lycos in early 1998. By November 2008 it enjoyed 23 million unique users. That didn’t start the European arm of Tripod being closed down in 2009.

Tagged – Tagged is the cockroach of social networks. It isn’t amazingly successful but is durable. In marketing circles they were known for their deceptive email marketing practices. Tagged users had email contacts taken from their address book and contacted. It resulted in a class action law suit that was settled in 2010. They seem to be still going as a social network site. In November 2008, Tagged had 22 million unique users.

imeem – was a hybrid of Spotify, Vevo and last.fm. Launching in 2005, it allowed had an advertising supported free music model. You could share music, stream tracks and playlists. It also provide widgets that you could embed on a website, Facebook or MySpace profile showing your favourite tracks. They were notable for being able to get licences from all the major US music labels for free streaming on the web.  The economics of it didn’t work, whilst it had 22 million unique users in November 2008, it was sold a year later in a fire sale by MySpace Music.

imeem were smart enough to have good mobile apps on iOS and Android right from the beginning. They allowed the purchase of DRM-free MP3s, which was a pioneering concept at the time.

Netlog – was a Belgian based social network aimed at a youth market. It was founded in 1999, but didn’t really start to take off until 2002. In 2005, they started expand beyond Europe. There were a couple of things notable about it:

  • You could send invitations to friends via Facebook, though the user experience was a bit of a mess
  • They geo-targeted and age-targeted content to users so you would only see content from people in the same region and age group as you

Registered users peaked at 94 million in November 2008 it saw 21 million unique users.

Honorable mentions

Cyworld – Korea had a unique web eco-system. Part of the reason for this was that the Korean government mandated a couple of security standards. Cryptographic links using Microsoft Active-X – I know, I know but they were at least thinking about security for sign-in and banking details, which is more than be said for most governments. Secondly Koreans needed to provide their national identity number. Part of this was due to the fact that Korea was a relatively new democracy with weaker free speech laws. Lets talk about the name ‘cy’ means between people in Korean, rather than alluding to cyber space.

Cyworld started in 1999 by a bunch of students who built the social networking site as an offshoot to discussions they were having about a research project. Most of the members abandoned it after graduation, but one of them became CEO at the end of the year.  It wasn’t particularly successful until a ground up redesign dubbed Minihomepy was rolled out in 2002. This pioneered virtual goods that pre-dates the stickers available on Asian OTT messaging services such as Mixi, KakaoTalk, LINE and WeChat.

Cyworld was bought by SK Telecom – a Korean fixed line and mobile carrier. SK had a distribution footprint including app space on their handsets, a popular online portal Nate and Korea’s most popular instant messaging client NateOn. Over 2003, Cyworld tripled in size from 2 million to 7 million users.


Cyworld peaked out in 2011 at 25 million users. It’s decline due to the rise of the smartphone. Kakao Story and KakaoTalk provided a superior experience on mobile devices and became ubiquitous in Korea with 55 million registered users (just about every man, woman and child in Korea). Facebook entered Korea in 2009, its progress in the country was slow, though it was useful for those Koreans who had international connections. A hacking attack which compromised the Cyworld and Nate user base precipitated a mass exodus from the platform in 2011. There is no evidence that Facebook was involved in the hack; but they did benefit from the fallout.

In November 2008 had 19 million users.

Oprah time: Master Switch by Tim Wu


Tim Wu is an American lawyer, professor and author on internet matters. He is main claim to fame is coining the phrase ‘net neutrality’ back in 2003. He is well known as an advocate of the open internet.

One needs to bear all this in mind when thinking about The Master Switch: The Rise and Fall of Information Empires. In it Wu posits a natural lifecycle for the rise and maturation of platform and media companies; which he called the ‘long cycle’.

Wu uses the following companies as examples:

  • Western Union’s telegraph monopoly
  • Western Union-owned Associated Press’ relationship with the nascent newspaper industry
  • AT&T in telecoms
  • The early film industry and the rise of Hollywood studios
  • Apple’s history – though this point is less nuanced because Apple has cycled a number of times between open and closed systems – a nuance that Wu doesn’t fully pick up on

In Master Switch, he points about how these companies have moved from open systems to closed systems in order to maximise profits and resist change. Wu then uses these cycle to argue that a repeat of history was under way with the modern internet. In 2010 when the book was written; this would have ben the rise of Google, Facebook and Amazon.

Ed Vaizey announced that ISPs should be free to abandon net neutrality in the UK and it was abandoned in the US when FCC chairman Ajit Pai was appointed by Donald Trump.