Ged Carroll

Category: innovation | 革新 | 독창성 | 改変

Innovation, alongside disruption are two of the most overused words in business at the moment. Like obscenity, many people have their own idea of what innovation is.

Judy Estrin wrote one of the best books about the subject and describes it in terms of hard and soft innovation.

  • Hard innovation is companies like Intel or Qualcomm at the cutting edge of computer science, materials science and physics
  • Soft innovation would be companies like Facebook or Yahoo!. Companies that might create new software but didn’t really add to the corpus of innovation

Silicon Valley has moved from hard to soft innovation as it moved away from actually making things. Santa Clara country no longer deserves its Silicon Valley appellation any more than it deserved the previous ‘garden of delights’ as the apricot orchards turned into factories, office campus buildings and suburbs. It’s probably no coincidence that that expertise has moved east to Taiwan due to globalisation.

It can also be more process orientated shaking up an industry. Years ago I worked at an agency at the time of writing is now called WE Worldwide. At the time the client base was predominantly in business technology, consumer technology and pharmaceutical clients.

The company was looking to build a dedicated presence in consumer marketing. One of the business executives brings along a new business opportunity. The company made fancy crisps (chips in the American parlance). They did so using a virtual model. Having private label manufacturers make to the snacks to their recipe and specification. This went down badly with one of the agency’s founders saying ‘I don’t see what’s innovative about that’. She’d worked exclusively in the IT space and thought any software widget was an innovation. She couldn’t appreciate how this start-ups approach challenged the likes of P&G or Kraft Foods.

  • LBOs in Silicon Valley

    With the rise of technology sector LBOs it makes sense to reflect on the Silicon Valley use of stock options and going public. In the past the Silicon Valley stock market dream was relatively simple. Hard graft with a possibility of a reward in terms of a stock market listing or a buyout by a larger technology company eager for the new, new thing.

    Now things are different, businesses like Google, Uber and Facebook held out for as long as possible to go public. Technology companies from Apple to Zynga have been punished repeatedly in the market for real and perceived mis-steps. Activist investors charge around Silicon Valley in a similar manner to the way they bullied the S&P index in the 1980s.
    160704 - private equity & the tech sector
    Now technology companies are making up almost half of private equity LBOs. An LBO is a leveraged buy out; its where a prospective owner uses a mix of loans and their own money to purchase a company. The company usually has a steady cash flow that is used to pay down the loans and associated interest. These businesses are generally discounted because they are no longer perceived as being high growth companies.

    The private equity owner looks to either flip the company to another purchaser, or flip parts of the company to pay down the loan. Either flipping or piecemeal sales are designed to raise more value than the original price paid.

    Since these businesses are servicing large amounts of debt, they are vulnerable to fluctuations in their business conditions or interest rate rises. For example, Irish telecoms network Eircom defaulted on corporate bonds in 2012, having been through a couple of LBOs in the previous decade.

    There always has been some LBOs in the Valley, Computer Associates bought up rivals and ran them as part of a conglomerate, with a focus on maximising the business cash flow rather than market share growth. General Atlantic Partners and Cerebus Capital Management had specialised for a long time in LBOs of mature ‘also ran’ business software companies with regular support customer support contracts. But the recent growth in LBOs is unprecedented for the technology sector. More on Silicon Valley here.

    *January, 1 2016 – July 1, 2016.

    More information
    Private Equity Has a Crush on Tech | WSJ – paywall
    What are the major reasons behind Yahoo’s drastic downfall?
    Barbarians in the valley

  • WWDC 2016 – what did it all mean?

    I watched the few hours of keynotes at Apple’s WWDC 2016 (Worldwide Developers Conference). I also read some of the resulting analysis and wondered if we’d been watching the same event.
    Cómo ver la WWDC 2016 en vivo en iPhone, iPad, Mac, Apple TV
    So thought I would think about the event carefully and come up on my take of what it all meant. This is a bit later than I originally planned to publish it.

    Firstly, there was no change in direction for Apple from a strategic point-of-view. Apple has been clear about its direction, it is the ‘how’ which is the mystery. WWDC 2016 was a major unveiling of ‘the how’.

    Over the past few years, Apple has focused on the integration of its devices. The reason why there isn’t one OS*, a la Windows 10, is that the different form factors have different contexts. Cross-pollination of services only takes place where it makes sense, which is why Siri has taken a while to roll out.

    The first big thing is APFS – a new file system for all of Apple’s devices. This builds on upon the feature set of ZFS which was a file system developed by Sun Microsystems for its Solaris UNIX operating system. Apple had experimented with implementing ZFS in OSX Leopard, but then didn’t follow through. Solaris runs on large enterprise computers where the prevention of data corruption and handling a large amount of file changes simultaneously is very important. Like ZFS, APFS supports encryption, granular time stamping, fast file management and has improvements in data integrity. When it’s fully finished it should make encryption on devices easier to manage and provide the user with more control. It should also help with syncing data across devices and the cloud.

    The interesting thing is how this technology will scale over time handling multiple devices and form factors working seamlessly from a common database. Like many of there other technologies this is an extension of Apple’s Continuity offering and future integration with a wider IoT offering.

    When Steve Jobs launched Mac OSX 10.0 in 2001 he described it as being the OS for the next 15 years. At the time the original MacOS was showing its limits. The UI was colour but hadn’t really moved on that much since System 7.5. The operating system wasn’t multi-tasking. The internet felt kludgy even though it performed well on the hardware at that time. Looking at OSX / macOS now, the operating system it feels fresh. The tweaks and changes under the hood keep the performance hub and the features comparable with the rest of the Continuity eco-system. macOS also doesn’t seem to be seriously threatened by iOS ‘pro’ devices.

    iOS 10 was important to me for its embrace of messenger-as-a-platform. Apple innovates within its own Messages apps with some UI gimmicks. More importantly, notification real estate that was once the exclusive preserve of the Apple dialer. This allows you to accept calls from the likes of Skype, WeChat or Slack from the lock screen. This follows Apple’s model of using it’s own apps to work things out and then open up the function once it is mature. Apple’s own Messages app includes a number of features including:

    • Simple chat bot-like functionality
    • Swipe to read on messages to prevent shoulder surfers from reading messages
    • Messages app takeover emotions
    • More emoji / sticker like icons

    Apple Pay roll-out – continued geographic roll-out makes sense. Apple Pay isn’t about building a rival payment system a la PayPal. Instead, Apple is trying to build more touch points with the user. The level of usage doesn’t matter too much from that perspective. Geographic roll-out to Hong Kong and more European countries makes sense. The more exciting development is two-factor authentication for e-commerce payments on compatible sites using the Apple Pay infrastructure. This is big for shopping on both Mac and iOS-powered devices.

    Thinking differently about intelligence. Unless you have been living under tech industry equivalent of a stone, you’ll be aware of cloud companies like Microsoft, Amazon, Google or Baidu using artificial intelligence techniques to drive device function. Apple hadn’t been as visible in this space up to WWDC. The reason for this is due their rigorous approach to user and device privacy.  There were two approaches to this:

    Having the mobile devices GPU to perform relatively simple neural-network computing. This can learn user preferences or intent over time and be more helpful

    Making Siri more intelligent by looking at the behaviour of users encrypted, salted with false data and aggregated up. Differential Security is the process of acquiring this data. In the second world war, the Allies cracked the cryptography derived from the Enigma machine. But that was only the first part of the challenge. In order for it to be useful the Enigma team used statistics to hide any usage of the intelligence hiding reactive activity in the midsts of statistically expected ‘normal’ behaviour.

    Differential security is kind of similar to this. All the data is encrypted, the phone sends a mix of false data and real data. When Apple looks at aggregated data they can see the false data as being false, but can’t tell which users data is false at a given time.

    Apple’s WatchOS 3 is interesting because of the performance boost it gives the wearable. The difference is really noticeable. The boost in performance is due to Apple having more memory to use than it had originally allowed for. This provides a more refined experience. Much of the UX enhancements were focused on fitness.

    From a developer perspective there were a few things missing:

    • Apple had no new pro-level hardware announcements
    • Apple later walked away from Thunderbolt displays, saying that 3rd parties were now making great displays. This reminded me of when Apple stopped making printers, it felt permanent, though there is a lot of speculation about a forthcoming Apple 5K display – we’ll see
    • Apple still needs to do more work on integrating its Swift programming language throughout its OS’
    • Given Twitter’s peak in growth, Apple didn’t show how Siri would cope in a post-Twitter world

    Finally the two-hour keynote was a love letter to China. At every opportunity Tim Cook mentioned the Chinese market, support for China-specific items like language and called out Chinese apps like WeChat.

    * From a technical point-of-view; tvOS, iOS, and macOS all share underpinnings based on NetBSD and a Mach micro-kernel.

    More information
    Apple Pay supporting banks | Apple Support Documents
    Apple finally opens Siri to third-party developers | TechCrunch
    Apple rolls out privacy-sensitive artificial intelligence | MIT Technology Review
    What is Differential Privacy? A Few Thoughts on Cryptographic Engineering
    Digging into the dev documentation for APFS, Apple’s new file system | Ars Technica
    Apple File System Guide | Apple Developer documentation
    Mac & iOS Continuity | Apple

    More articles on Apple WWDC through the years here.

  • Throwback gadget: SnapperMail

    Thinking about SnapperMail takes me back to end of 2001, I started to prepare for leaving my job at Edelman. This meant upgrading my home IT set up. I picked up an iBook. The iBook was Apple’s consumer-orientated laptop made from 1999 to 2006. Mine was a second generation ‘Snow’ laptop with a G3 processor, dual USB sockets and a combo drive which allowed me to watch DVDs and burn CDs.

    I used the move to go on the first version of OSX. The move also meant that I got a new email account, my default account to date. It had two key attributes:

    • No adverts, so it looked professional in comparison to having a Yahoo! or Hotmail email address and it wasn’t tied to an ISP.
    • IMAP support which allowed me to use my email account across different devices that syncs across the devices. POP3 downloads the  emails from the server to the device, so is ideal only for when you are accessing email from one machine

    My iBook was my only source of email access whilst I left Edelman, freelanced, and then eventually joined Pirate Communications. My first smartphone was a Nokia 6600, which I used alongside a Palm  PDA – l got this sometime around the end of 2003. The 6600 supported IMAP out of the gate, it was slow, but I was connected.

    The 6600 was eclipsed by Palm’s Treo devices which were a better device. I moved from the 6600 and a Palm Tungsten T3 combo to a Treo 600 smartphone in January 2005.

    The process wasn’t smooth. The Treo was sufficiently fragile that I got a translucent silicon jacket that worked surprisingly well with the keyboard and screen protector to look after the touchscreen. Software wise the Treo 600 was a step back from the Tungsten T3 PDA. The screen was smaller and the software felt sluggish in comparison. I had deliberately chosen the 600 over the 650 because I had previously worked agency side on the Palm account and been a long-suffering device owner so knew how crap they were at bug fixes on new devices. The media didn’t call the former Palm CEO ‘Mad’ Bill Maggs for no reason (just sayin’).
    snapperfish limited
    Unfortunately Palm had not been as progressive in comparison to Nokia with its default email client. The software didn’t support IMAP. Fortunately I used to follow Mitch Kapor’s blog and he had recommended SnapperMail: an app from a small New Zealand company SnapperFish.

    SnapperMail was a compact modern email client. It has a number of features that we would expect now:

    • It supported IMAP
    • It supported SSL client to mail box encryption*
    • it was really easy to use
    • You could work with attachments including zipped files**
    • There was no restriction on the file size of attachments, the only restriction was your email account rather than your email client

    This looks like the kind of technology you would have thought Palm should have done. At the this time Palm were competing against Microsoft Windows Mobile 2003, BlackBerry 6200 series, 7100 series and early 8700 series. Yet the default email client was back in the 1990s.

    *The full-fat application cost US$39.99

    **SnapperMail came bundled with HandZipper Lite which handled the compressed files and JPEGWatch Lite image viewer

    I used this alongside MetrO – a public transit directions app and QuickOffice Pro – to read Office documents as part of my modern smartphone experience. It wasn’t just me that loved SnapperMail, it was praised by Walt Mossberg back when he wrote at the Wall Street Journal.

    SnapperMail won two Palm Source (Palm’s software licence business) Powered Up awards in 2003. It was recognised as Best Productivity and Best of the Best Solution. More on Palm here. 

    More information
    SnapperMail Has Solid Software For Savvy Mobile E-Mail Users | WSJ
    QuickOffice
    MetrO – open source mass transit application
    PalmSource Welcomes Developers with Awards, New Tools; Announces New Licensees | PalmSource press room

  • Google I/O 2016

    Google I/O 2016 happened on May, 18 – 20.  There had been a lot of pieces of coverage about the different products and services released. But I wanted to spend a bit of time reflecting on what Google I/O 2016 told us about their viewpoint on technology.

    Giving apps a second chance

    Google knows as well as anyone that the app moves towards a maturity model where consumers stick with the core apps that they want and then don’t go any further.
    apps
    Data shows that consumers use their top five apps 88 per cent of the time. So why would Google care when it knows that 60 percent of the top apps on the Android platform?

    The reasons for an expanded app usage include:

    • A proportion of Google’s advertising (like Facebook) is derived from the promotion of app downloads
    • Android devices are reaching market maturity in many markets, growth is likely to come from new uses – at least some of which will be derived from third party platforms
    • Google has staked its ambition in the PC sector on its Chrome operating system being able to run apps from the Android eco-system. In order for that to happen there needs to be a healthy community of developers
    • In the same way that DoubleClick’s ad network greatly expanded the inventory of Google’s advertising business, third party applications offer Google an additional source of usage for its own services. If you want to see the future of Google Apps look at the the way the likes of Baidu and Tencent allow third-party integration with their own tools

    Streaming or ‘instant’ apps is part of Google’s efforts to encourage consumer trial of new apps and enhance relationships with developers. Firebase, it’s new analytics platform for mobile developers helps them have a better relationship with their installed user base allowing them to use data to target notifications and campaigns.

    More faith in wider area networks (WANs) than personal area networks (PANs)

    Android Wear’s updates were interesting. Put simply Google has more faith in data being delivered in a timely manner over cellular or wi-fi networks than it does for inter device transfers over variants of Bluetooth. Both the Apple Watch and Android Wear products suffered from performance lags when the watch was a thin client of a phone. Having a cellular radio on board the phone presents challenges with battery life, but speeds up real world performance.

    The original design failure wasn’t down to network performance, but is likely to have implications for personal area network technology like Bluetooth in its different variants or ZigBee. These technologies are all about scale, lose a scale advantage and it poses a problem for future adoption by others. This can happen in a virtuous way. Apple’s adoption of USB benefited the standard greatly and drove interest in peripheral development for both Mac and PC. Apple’s abandonment of FireWire and the 3.5″ diskette marked their decline.

    Lots to be concerned about from a privacy point of view?

    Google Home moved yet another pair of Android powered ears into our environment. It was obvious from Google’s description of services that a paid marketing model to be the ‘car booking’ or equivalent service of Home could be very lucrative for the search giant. How this device could be used for market research, tracking brand mentions or government surveillance also poses some conundrums moving beyond smartphones to brown goods.

    Android N features file based encryption rather than treating the whole device as an encrypted disk. This raises questions around the comparative ease of access from a privacy perspective. Secondly, SafetyNet allows Google to reach into a phone to remove pre-existing applications without user permission. There is no explanation if they also have write privileges to the phone as well. If so, expect law enforcement and intellectual property owner interest. From the way it reads this would affect apps and content that have been side loaded as well as got from an app store.

    Android is giving the high ground to Apple on privacy presumably because it considers its own customers don’t care about it that much.

    Reference designs in VR to drive adoption and commoditisation 

    Google’s Daydream project looks to provide standardisation in hardware. By going down this route, Google hopes to spur on the sensor market required for improved AR experience and drive uptake. These will likely be a very different experience to the computer workstation powered Occulus Rift. Driving this technology into the smartphone market may combat the current stagnation in phone sales growth.

    More information
    Google I/O 2016 event page
    A16hz on Google I/O 2016
    Everything Google just announced at its I/O conference
    Palm, Apple, Google and the whole mobile device thing
    The Limits of Google
    If Google’s right about AI, that’s a problem for Apple – Marco.org
    ISIS’s Mobile App Developers Are in Crisis Mode | Motherboard

  • 2016 Mary Meeker presentation

    2016 Mary Meeker’s annual presentation on internet trends is a tradition within the technology sector that goes back more than two decades. Meeker used to be a sell side analyst during the dot com boom and was known as a cheerleader for the sector. Unlike Harry Blodget she didn’t come unstuck with the subsequent bust.

    More recently Meeker moved to Silicon Valley and took a job with a VC firm. Hence the reason why the 2016 Mary Meeker presentation is done in conjunction with KPCB (Kleiner Perkins Caufield Byers).

    The key themes explored in the presentation this year include:

    • Mobile – a favourite for a number of years, but with over half of all internet sessions being done on a smartphone or similar it was inevitable that it would take up a substantial amount of the presentation. Mobile is maturing which is shown in the decline in growth rate of the sector this year. Android is picking up market share due to its cheaper handsets but still lagging behind in share of profit
    • Declining global economic growth. Global debt has risen higher and faster than global GDP. Population growth is also slowing and ageing. Meeker thought that India may be the bright spot due to its demographics, but this assumes that it can get over its structural issues and take advantage of its young population. That is probably overly-optimistic because of rising hindu nationalism
    • Online advertising – efficacy still a serious issue to be dealt with. Consumers hate it hence ad blocking.
    • Social: Meeker saw the big factors being video, images and messaging
    • Voice: the rise of voice driven assistants in the home and on mobile devices. The decoupling of China versus the rest of the world is apparent in this new category.

    Here is the latest iteration for 2016

    More on Mary Meeker here.