Archive for the ‘technology’ Category

This is reposted from my personal blog. When the latest Microsoft results came out one of the biggest drags on it was losses made by the company’s online services business; specifically the search product Bing.

The challenge

Google has had an immense head start in building the best search engine (at least for Roman-based languages), it has built up an unrivaled search index, crawling mechanism and a search algorithm that goes through a complete iteration every three years or so according to Steven Levy’s In The Plex. The technical and audience experience challenge that Bing faces is akin to American car companies who woke up to Toyota and Volkswagen having changed the rules of motor manufacture and struggled to catch up as their German and Japanese rivals continued to further improve.

That isn’t to say that Google is unbeatable: Russian company Yandex, Korean company Naver and China’s Baidu have all managed to build search engines that better match the needs of their markets, but no one company has managed to challenge Google across markets.

When I was at Yahoo! we managed to build a search engine that provided a search experience that was as good in terms of relevance of the queries as what Google had to offer. But being just as good is not enough. Google isn’t just a search engine, its a habit, people look there automatically as routine and this is hard to break.

I find it hard to remember a time searching before Google, I remember that I used to use Excite, HotBot and AltaVista depending on what I was looking for and would ‘work’ the search engines to dig up what I wanted. Some time in 1998/9 I was reading a website (I think it may have been Wired or WebMonkey) came across Google and never went back. It was that good. Bob Cringely wrote that in order for us to make that change we have to perceive something to be 10 times better. Bing has never got anywhere close.

Many of the ‘innovations that Bing brought to the table were ‘borrowed’ from Ask.com, like its trade dress. Lycos Europe’s IQ service and Yahoo! both did really interesting stuff with social search; which is exceptionally prescient of Google’s more recent efforts (and involves some of the same people like Bradley Horowitz) but due to user experience and the brands not having ‘permission’ from consumers to be truly innovative; didn’t gain the level of acceptance they should have deserved. Ask had a really hot algorithmic search engine in Teoma (which I still use occasionally), but again it never got the audience it deserved.

So Bing has to work harder and pay more money to gain the traffic that it has:

• Working with mobile carriers like Verizon to be the default search engine on the company’s mobile phones, paying media sites to have Bing as their search box
• Arranging for a PC manufacturer to set Bing up as the default search engine on the internet browser that they supply as part of the software on PCs that they sell
• Arranging for a Bing toolbar to be bundled with free-to-download software (like Adobe Acrobat reader)

The second problem that Microsoft seems to have (looking at both the online services division and the financial performance of Yahoo!) is that it is failing to monetise those audiences it has effectively. There are still too many searches happening with no relevant advertisements displayed alongside the organic results.

From a financial perspective, just how bad is it?

Well according to a Business Insider analysis piece written at the end of April:

Bing is paying about 3X as much for every incremental search query as it generates in revenue from that query.

What does that mean?

It means that for every $1 Microsoft generates from each new search query it buys, it spends $3 to get it.

(And that’s just direct costs–the costs of obtaining and processing the query. It doesn’t include sales and marketing, research and development, and general and administrative costs–all of which are subtracted from the -$2 Microsoft has already lost on every new query.)

And Microsoft is in a rather unique position to distribute Bing as the default search engine on Internet Explorer with many Windows PCs, otherwise gaining access to search audiences could come at an even higher cost.

If Microsoft was to sell Bing, who could buy it?

Hypothetically speaking, if Microsoft decided to sell Bing, there aren’t that many people that have the deep pockets, chutzpah or knowhow to take advantage of it, but here is a list of some of the more likely candidates:

• Alibaba: whilst Jack Ma is a super-smart business man; he has enough problems extracting Yahoo! Inc. from his current business rather than acquiring Baidu
• Aol: is in a similar position to Yahoo! in many respects with not enough cash on hand to do the deal and too many things that they need to address internally to even consider a Bing purchase
• Apple: has the money, but there isn’t the incentive or the expertise to take on the integration of the business into Apple and the leviathan of a challenge that whipping Bing into a product that Apple could be proud of
• Baidu: Robin Li and the team at Baidu certainly have the smarts to take on Bing, the finance could be made available via various sources in China. Baidu has smarts in search; particularly non-roman languages which happens to be where the market growth is likely to be. In addition, Bing would be an ideal platform to bring Bing’s concepts of box computing to western consumer markets. Microsoft would have the problem that it wasn’t selling its search arm but empowering the next competitor to challenge it in the mobile and desktop operating system space with a Baidu acquisition. The biggest issue is would this pass muster with US politicians? As you would have a nexus of wounded US pride, high technology and the politicised issue of censorship. Throw in a couple of Google lobbyists for good measure, stand back and watch it burn
• Facebook: could buy Bing and it could be attractive to Microsoft; particularly if the purchase was made with some Facebook equity included. But they don’t need to do it and the shifting of the focus could adversely affect investor sentiment towards the great satan of social
• Google: wouldn’t be able to buy Bing because it would set off antitrust alarms throughout the western world
• Naspers: the South African media company has a reputation for making smart investments in online properties and could add its brand to a larger consortium
• NHN: The Korean owners of Naver understand search and could bring a new fresh approach to non-Roman language search but they just don’t have the revenue to do it on their own. They could still be a wildcard in a private equity-based bid
• SK Group: South Korea’s largest conglomerate has a number of successful online services that search would dovetail into, however they are already committed to investments in developing markets like Vietnam and have been burned by US-based investments in the past like the Helio and lost out on a recent bid for Blockbuster Inc. to Dish Networks
• Tencent Holdings Limited: As a major online property in China, Bing maybe desirable as a way to help combat the march of Baidu. However, it is uncertain that Tencent would have the relevant deep technical knowledge to turn Bing from costly millstone to world-beating search engine. Like Baidu Tencent would also come under US political scrutiny. One thing in their favour would be having Naspers as a major shareholder – so taking some of the sting out of the anti-Chinese rhetoric likely to fly around
• Yahoo!: couldn’t afford it, is struggling to grow the existing business that it had and has dissipated its search talent around Silicon Valley thanks to Bartz & Icahn. It would be a strategic pivot too many and would likely get killed in a shareholder revolt if they could find someone to bank roll the deal
• Softbank / Yahoo! Japan: Masayoshi Son is an exceptionally savvy business man. Bing would have to be about more than business: self actualisation. Yahoo! Japan snubbed Bing in favour of Google recently, a purchase of Bing could be awkward
• Yandex: it is doubtful that the Russian search business could raise enough capital to take over Bing; if it did it would have its work cut out to turn Bing around in terms of market share. But they would have the smarts to increase profitability of what they already have
• Private equity: in theory a company like KKR could make the relevant phone calls and round up funding to buy Bing at the right price. They would have to find a management team, a path to cut costs and find a clear marketing proposition to keep audiences. Finally private equity would want to have an exit; there wouldn’t be that many acquisitive buyers, so they would have to look at some sort of public offering

Why keep Bing?

One of the reasons that I was so surprised that Yahoo! gave up on search was that it is not only a business providing a potentially great contextual advertising point of view. It is also the glue or the mortar that holds online services together. Given that there are so many screens are now part of the internet experience: mobile, television and desktop – the mortar is going to be even more critical in the coming years as the binding agent for a larger eco-system.

With productivity suites moving into the cloud, search becomes the new file system (like Spotlight on the Mac desktop) and Office (or its future cloud sibling) is the core of Microsoft.

More links:
Bing: Should Microsoft Sell It?

Here’s Why The Bing-Yahoo Deal Isn’t Working So Far

Microsoft Is Redoing Bing To Look More Like The New Windows Design

Microsoft Might Dump Twitter From Bing Search Results

Can we please stop pretending that Microsoft’s Bing is doing well

For the first time in nine months Bing doesn’t gain in search share

BING COPYING GOOGLE: Embarrassing But Brilliant

NTT DOCOMO have announced that they will be attending CommunicAsia in Singapore (June 21 – 24) and a couple of things that they will be showing off caught my eye. First of all NTT DOCOMO are pushing the boundaries on technology to try and ensure that we can have world phones that support existing and next generation networks – they will be showing off a six-band prototype amplifier.

Probably of more interest was the idea of roaming mobile payments via NFC. NTT DOCOMO already signed a deal to make this happen with KT (Korea Telecom). The question that crossed my mind was did telecoms carriers have the processing capability that credit card processors like MasterCard, Visa and American Express? I don’t think that the current systems for roaming charges would be sufficiently swift and optimised for the wide range payment values. Cross-posted from my personal blog.

Old Microsoft

One thing that Microsoft was always really good at was doing demonstrations of new technology. Bill Gates historically gave barnstorming performances demo-ing products, even when they crashed. Although this sounds quite twee, as an area of expertise, its really hard to do. If you want an idea of how hard go on to YouTube and look for Apple keynote product demonstrations before Steve Jobs returned versus once he was back at the mother-ship.

The tile interface

One of the most interesting things that I found about Windows 8 was the tile interface, because of the implied design choices that Microsoft seemed to have made.  Historically, one of the key sales points of using Microsoft software was that there was lots of people who were familiar with it and it consequently improved productivity and reduced training requirements; generally how things work is one boat Microsoft historically hasn’t rocked. There has been some changes like the Clippy help character and the interface of Microsoft Office 2007.

Clippy was removed from future products. As for the interface of Office 2007; well it still cause consternation at the different companies that I have worked at as users have struggled to transfer their skills over. So from a user experience design point of view; tiles took a lot of guts for Microsoft to roll out.

The tiles concept is also an interesting compromise. It move the interface away from 3D eye candy; so reducing processor power required; but still keeps the transitions which are the true magic in modern touch interface operating systems. Which is an interesting trade off in power and performance.

The two dimensional aspect of the Tile interface reminded me of a 21st-century HyperCard stack.

There were a number of aspects of using backwards compatible applications alongside native Windows 8 applications in the Steven Sinofsky demo that looked like a user experience car crash. This difference is even more marked when you think about the difference in philosophy and mindset between the interfaces: from keyboard to touch.

But is it innovative?

One question that I did have about the design choices for Windows 8, was whether Microsoft is being buffeted into these choices by trying to match competitor innovations, or whether this was Microsoft carving a new path.

Let’s look at some of the main features:

• Boot disk on a USB memory stick – in Microsoft marketing speak this is called Portable Workspace; this has been available from the Linux community for years. Apple historically hasn’t gone down this route, mainly because it also wants to sell you the hardware that you boot up on as well. The hardware and the software license are one. In terms of Microsoft’s current business, this isn’t much of an innovation as they already have licenses that allowed consumers to use Windows at home; based on their work Windows licenses: WHERE those machines were not used concurrently – Portable Workspace is a logical progression in terms of licensing terms. My initial take on this, is that Microsoft does not seem to be embracing Intel’s USB replacement Light Peak / Thunderbolt
• HTML5 desktop applications – back in the day Microsoft used to be able to rely on having an army of developers coding the new, new thing for its platform. A lot of that expertise is now working with web technologies. Microsoft’s use of HTML5 desktop applications follows in the conceptual footprints of Konfabulator, Yahoo! Widgets, OS X widgets, Google Gears and pretty much the entire SDK for the HP/Palm WebOS
• App store – Nokia used to have a poor app store experience, but still had one back in the day. Apple have one, Google has one, GetJar has had one for years. As had mobile carriers

Maturation of a platform

Historically, one of the key differences between the Apple ecosystem and the Windows ecosystem was that a PC could be upgraded, with the hardware being much more modular in nature. In reality, there was a lot of work required to get all the pieces to talk to each other and it used to provide hobbyists no end of satisfaction. There is a whole ecosystem of aftermarket board, component and case manufacturers that supported tinkerers and hardcore gamers. The ‘Franken-PC’ was also the entry-level computer for many lower income consumers.

By comparison, Apple computers were considered to be more of an appliance, with true plug-in and play, years before Microsoft described it as a feature of the Windows operating system. It also allowed Apple to take a more holistic systems-based approach to consumer design.

With Windows 8 Microsoft is more closely with component and PC manufacturers to reduce the amount of component configurations and combinations that they have to code for. The operating system also supports system-on-a-chip architectures. Historically compiling lots of different functions on a chip was used as a way to reduce costs in mass-produced consumer electronics. The use of it in PCs implies a different design philosophy, one that is looking to simplify mass manufacture, and by implication that there may not be as much room for differentiation and after-market components.

The PC looks like it is discarding its hobbyist heritage and becoming an appliance; a bit like the same journey that radio went through in the early 20th century, or the motor car over a longer period.

A focus on the middle

I get the sense that power users are being abandoned with Windows 8, it is a world away from a hard day’s graft in a CAD programme, or Excel; based on a hope that it gains the resurgent consumer demand for touch computing devices.

Is touch all its cracked up to be?

One of my colleagues Becky talks about the way her young daughter pokes and prods at their iMac computer screen to try and interact with it. Seeing something like that makes one think that touch is the THE future. I think that touch has a future but is only one part of the future.

First of all a trip back in time: touch interfaces aren’t new. Some of the earliest computing work on touch interfaces was done as part of American efforts into defence, with light pens being used as part of the SAGE system that utilised computing technology to help detect inbound Soviet bombers and direct US missiles to intercept and destroy them. This was all developed in the 1950s and put online by 1963. SAGE influenced the original research that would eventually begat the ARPANET; what most people acknowledge as the precursor of the modern internet.

What was interesting about the SAGE system was that the light pen operators didn’t have to type that much, so it made sense to keep their hands on the screen.

Now move forward to 1968, and Doug Engelbart’s now famous technology demonstration to the public of the work done at the Augmentation Research Center – which was part of SRI in Menlo Park. Engelbart demonstrated many of the concepts that we now take for granted: desktop video calling, WYSIWYG interfaces, graphical user interfaces, the computer mouse, file system structures and hyperlinks. Engelbart was probably building on research and a body of knowledge that would have come out of the SAGE system on man-machine interaction; from a technological, ergonomic and human behavioural point-of-view. The big thing that came out of that demo was the prominence that was given to the mouse, over a touch interface. This was because, if you are creating content by typing, it is a bit of a pain to move your fingers then on to the screen.

If we move forward to 1983, Hewlett Packard launched the HP-150, this was a MS-DOS compatible computer with a touch screen and keyboard. It didn’t sell that well, Bob Cringely in his book Accidental Empires talks about how engineers not using the touch screen whilst working on the machine before launch: preferring to use the arrow keys on the keyboard instead showed how difficult it was to put the touch screen into human behaviour.

When the HP-150II was launched the following year, the touch interface was only an optional extra; which was rarely chosen by customers.

If we move forward to the present day Apple is taking a two-pronged approach to touch: with a touch screen interface on the iPhone and iPad; where as the Mac product range has track pad-related interfaces. There are some commonality of gesture controls but that’s it. A lot of this is down to user contexts: the iPhone and iPad are about information consumption and doing kiosk-like tasks; whereas the Mac range is about content creation. With content creation; users like to keep their fingers closer to the keyboard hence the track pad solution. Also because this is closer to the metaphor of the mouse; Apple hasn’t adopted a universal look and feel across all the devices.  You can certainly argue that it is narrowing the gap between them but the important bit to emphasise is not their similarity but their differences: what Apple has chosen not to do.

For my sins, I worked for a brief while in consumer banking; part of the work that I did was with front-line contact centre staff. Whilst I was there, we moved from terminal style screens where you moved around by pressing the tab key to a full graphical user experience (GUI) requiring operators to use a mouse as well as their keyboard.

Whilst there was lots of research to indicate that GUIs help productivity, we found in our particular context that that just the opposite occurred. Average call times increased as contact centre operators slowed down their calls to match the speed at which their worked through customer account computer screens. Having to constantly move their hands from keyboard to mouse, together with increased time it took to render the complex screen started to add up during peak hours resulting in longer customer wait times and an increase in levels of customer dissatisfaction.

I am not trying to say that touch is bad, but that different user contexts require different approaches, and like previous technology manias before it where vendors glom on an idea regardless of any context considerations. So I am a bit concerned about the user experience design for Windows 8.

This is cross-posted from my personal blog.

More reading:

10 Things to Know About Windows 8 – Extremetech

Live from Microsoft’s Windows 8 preview event at Computex 2011 – Engadget

Previewing ‘Windows 8′ – Microsoft News Center

Microsoft Brings Touchscreen to PCs & Laptops to Challenge Apple – But Will it Work – ReadWriteWeb

Here is a scan of my notes from Mobile Monday London last night. Whilst I was convinced that this is an area that will be big, there is still lots of problems to be overcome from a profit point-of-view. But most concerning to me is the various safety issues from loss prevention to consumer data protection and the ethical challenges presented by big data.

This is cross-posted from my personal blog.

This is cross-posted from my personal blog. Last night was the spring edition of Mobile Monday London’s famous demonstration nights and there was an interesting mix of products, software and services on display.

Intel demonstrated their application store which sounded similar to others in terms of terms and conditions. One thing I got a sense of from them is that they are more determined to make MeeGo a success to spite Nokia. The demonstrations they had running looked more polished than demos of the RIM PlayBook the month before.

Threedom was an interesting experiment in inclusive design.

QRpedia was less of a business and more of a duh why hadn’t anyone else thought of putting QRcodes more effectively with Wikipedia.

Swiftkey and the later Nuance demonstrations were about information input for the Android platform which made me thin that the user experience design for input must be broken.

Kineto’s idea of piggy-backing on wi-fi was something that I found interesting. But I didn’t feel entirely convinced mainly because pretty much all the wi-fi networks I come across are locked down.

Insiteo’s indoor location play seemed equal parts science fiction and Big Brother. However if it can help me successfully navigate Selfridges I’d be sold.

Natural Motion’s demo of  virtual reality through a screen using the gyroscope on an iPad was impressive despite my dislike for games.

Parcel Genie’s real-world gifting was an e-commerce twist on the real-would voucher gifting that Asian social networks like Daum have been doing for years.

Mindings is a Chumby-like platform based on Android with a healthcare spin on things, allowing accessibility to social content for older people and e-health applications.

I was reading an interesting article from Charlotte McEleny this afternoon about the next steps for social commerce and it got me thinking about my general attitude to shopping online. With the proliferation of retailers embracing social commerce it seems more of a norm than a rarity when you go onto an online store and can interact with others, read reviews and share experiences etc. I would be interested to know however how many of us feel that social commerce really does make a difference to our shopping habits? At the end of the day, yes I read reviews for products online but I will still send an email round the girls at work or to my mum to get a final opinion before I buy. In fact I’m not sure I find the reviews online massively helpful at all - just confusing – as they all tend to contradict each other and I know far too little about these people to know if their intent in buying matches mine and therefore if our opinion on the product will be the same. Do you trust the views of a complete stranger when making a decision to buy or not to buy?

As Charlotte points out, what is missing in social commerce of today ‘is a human element’ – ‘If we can now watch The X Factor or The Only Way Is Essex together online and comment and discuss what’s going on live, why can’t I take my friends shopping?’. We need more than just a static review on a site – we need proper interaction. Recognising this, Dell’s global VP of online Manish Mehta told new media age last year that the company wanted to take the social aspects of high street shopping and recreate them online by letting people interact and have discussions in real time when buying products from Dell’s website.

Perhaps this is also where the new Facebook shops will prove their worth? One of the first companies to announce a fully transactional Facebook shop was Asos in January. With its new application users will be able to buy directly from a brand without leaving the ultimate online social destination that is Facebook.

So what do you all think? Have you been taken in by social commerce or are you waiting for other retailers like Dell to push the technology even further? Do reviews help or hinder your purchase decision? Any and all thoughts appreciated…..

So I was up early doors and tuned into CNN which is my usual MO when I am staying over in a hotel.

The presenter on CNN was giving people the 101 on the mobile web

And how to use a QR code

But then their production staff manage to obscure the QR code so it wouldn’t read any way.

Here are the notes I made at the session that Mobile Monday ran on tablet devices.

A few observations:

• The irony of a heavy early-adopter crowd gathered to find out more about tablets, let very few of them had a tablet themselves. If I were a tablet manufacturer I would be doing an Oprah-style giveaway in meetings like this
• Rather than Nokia’s three platforms of Wp7, Android and iOS. It seemed to be a two-horse race in terms of the discussions platforms: Android and iOS. Dave Gibbs of BSkyB did call out BlackBerry as a promising platform and there was a real-life PlayBook demonstrated in the drinks session afterward. Microsoft needs to be mixing it up at events like this
• There was a sharp dichotomy between the developers in the audience and the panelists over the viability of HTML5 as an application/service platform. The developers liked it, the panelist felt it wasn’t fully-featured enough and felt that there was a number of consumer acceptance barriers to non-appified rich mobile web services

This is cross-posted from my personal blog.

So I did probably my most painful thing this month and looked at my credit card. I went through my transactions and then realised that one looked a bit amiss on the American Express statement.

I had a flight to and from Hong Kong, but the statement claimed that my flight went into Kai Tak airport. Which would be a bit hard since it has been closed since 1998. It reminded me of the continued power of mainframe computers. The travel industry and the banking sector both use giant computers to crunch data, much of the code on these computers has a heritage going back decades, but they provide a much more secure and reliable platforms than more ‘modern’ machines.

I can quite easily see what happened. Since Chep Lap Kok took over Kai Tak’s mantle as Hong Kong International Airport including its IATA and airline flight codes so apparent no updates were required in a database and Kai Tak airport got to live on as a piece of legacy data.

This is cross-posted from my personal blog.

Facebook have released a video to explain the hows and wherefores of their new location-based service addition to the social network Godzilla. You could argue about whether this is sharing too much in one place or what kind of privacy car-crash Facebook will make out of location data but that is for a another time. Here’s the video.

This is cross-posted from my personal blog.