Author Archive
How do we achieve the vision of the Context-Aware Web? Ideally, the website should gather as many context variables as possible through JavaScript (a-la Google Gears) or even the User Agent string. To make that happen, standards and universal adoption are required. What can be done today?
Context Seeder Apps - a term we coined at Handi - are a great compromise. The idea is simple - create a basic native application that extracts all context variables (with user’s permission) from the device and puts them in a GET string. The user is then taken to the parent website which is now more aware than ever before. There is also a possibility for real-time context updates by AJAX if the host platform allows for background applications or embedded browsers.
This is really the browser’s job. However, there are all sorts of privacy and permission concerns - a seeder app is custom to a particular website, has the appropriate branding and user’s trust. Google Gears now offers seeder capabilities to compatible websites on Windows Mobile and our upcoming iPhone projects will also have sleek seeders distributed through iTunes.
Of course, knowing about the environment alone is not enough - the application has to use the data meaningfully. It’s not easy - even Intel is struggling with solving this puzzle. Starting getting ready though - the context is coming at you!
Happy Hacking!
UPDATE: A recent application by the name of PhoneGap supports our approach. Also, catch the current Carnival of Mobilists at MobHappy because the next one is hosted right here at MobScure!
AdMob has been the poster child of the growth of the mobile web. Billions of impressions a month, a huge variety of devices and carriers - a success story from every angle. Brilliant play by the founder Omar Hamoui, no question.
The future of AdMob is less glamorous than its present. Success of this ad network has been always based on the premise that the mobile web is different from the real web. For years, this was true and mobile internet users represented a growing vertical audience. However, the recent stats are alarming - AdMob reports that iPhone users in the US represent 1.4% of its audience, while a Nielsen Mobile White Paper gives the iPhone users 4% of the mobile internet surfing, a difference of 300%.
iPhone users are less likely to go to sites that use AdMob advertising, because they are unaware of existence of a separate “mobile web”. Instead, they go to familiar portals like Facebook, the New York Times or Digg - sites that might or might not have iPhone optimized pages. After all, their iPhones are extensions of the regular Web, with regular web advertising served by the likes of DoubleClick. Furthermore, last time we tried AdMob on the iPhone it periodically served links to WAP sites, which are not supported by Safari and show up as plain text. The old mobile web is going away before it really took off, giving way to optimized interfaces for regular web sites.
It’s not just the iPhone - Nokia Nseries and the upcoming Android platform all have excellent web capabilities. This threatens to turn AdMob into a niche ad provider as it can’t compete with DoubleClick on the desktop, while the reverse is possible. In this light, AdMob’s recent initiatives targeted at iPhone developers are the right move. An even better one would be getting acquired by a big Web ad network.
Release of the 3G iPhone around the world was a landmark event for several reasons. Now, many key problems that have been visible in the mobile ecosystem are gone for the first time. Great UI, unlimited (or close to it) high-speed connectivity, a solid application distribution chain with a link to payment - the iPhone combines them all. At least when it comes to early adopters and application developers, there is no more reasons to complain - we just have to make things happen, living this lifestyle and making great mobile services. Here is the a short re-cap of my first day with iPhone 3G.
It started with a hike to the Grouse mountain. I got bored halfway to the top and started streaming videos of songs that I didn’t have from YouTube, listening to music in my headphones. 3G means that all digital content in existence is available in real-time, anywhere.
On the way down, I played around with a few free apps like Facebook and Twitterific. They were more sluggish and less visually enticing than the corresponding websites, specifically designed for the iPhone. 3G will unite the mobile ecosystem primarily through Mobile Web. Can someone please finally come up with JavaScript APIs for location, camera and so on?
Next on our agenda was an amazing barbeque (happy birthday Enej!) where the iPhone finally died due to a long day of browsing and streaming. I realized that I haven’t made a single call and only sent two text messages in all that time. 3G will dethrone voice and text as the killer mobile apps, complementing them with a variety of better communication tools.
After a quick recharge and a shower I headed to a party in a remote part of West Vancouver. The cab driver (a former USSR sailor, who’s been in the taxi industry for over five years) did not know where the destination was and used the iPhone to navigate. 3G is heavily endorsed by former Soviet Sailors.
At the party itself, mobile video definitely rules the show. This Mr.Oizo song had a cool clip!!! Let’s enjoy it once again. Have you seen that YouTube video where a guy comes and pokes his friend FaceBook-style? No need to “forward it to me later”, let’s just search for “facebook spoof” and laugh at it one more time. 3G complements social interactions when used properly.
This was just one day from my 3G-enhanced life. The technology is far from perfect - GPS can be imprecise, applications - sluggish and the service itself is far from cheap. But - the new era of mobile is here, in my palm…
Just hours after my post about the diminishing relevancy of the .mobi TLD I got an email from Vance Hedderel, Director of PR and Communications for dotMobi. He was quick to point out that the TLD is doing pretty well, with several brands, papers and services launching their sites in the space. Vance had some excellent insights and I would like to once again state our respect for their team. They do a lot for the mobile web space - it’s just that we see the long-term future of this TLD differently.
After some soul-searching we decided that .mobi will likely become the new call-to-action, similar to the way SMS is the current way of engaging mobile services. Working with one of our partners, we had to consider facilitating inbound SMS messages worldwide. The two ways of doing it are by using a longcode (usually a UK-based phone number) or by getting local short codes for every different country (optimal user experience, but significant overhead when it comes to pricing and legal work). Neither way is optimal. If the users were educated about mobile internet and had a quick way to type in the URL, seeing a .mobi site could indicate that it features an ongoing campaign specifically targeted at the mobile context.
At the same time, we believe that search will become even more important and diminish the value of domain names altogether. It’s an exciting time to be in the industry!
Meanwhile, MobScure has once again been honored by being featured over at Carnival of the Mobilists, held at mjelly this week. Be sure to check it out!
Last week, ICANN ruled in favour of allowing customized top-level domains. It estimates that these new properties will cost upwards from $100,000 - thought the cost will definitely come down with time. Soon, we might be seeing things like .vancouver, .map, .food and .hiking - and the numerous sub-domains associated with these TLDs.
This decision effectively ends the debate about the usefulness of .mobi. dotMobi describes itself as “the gTLD dedicated to users who access the Internet through their mobile phones.” It’s unrealistic to expect that out of hundreds new TLDs users are supposed to only trust .mobi sites for mobile access. Every site on every top-level domain should be mobile-friendly.
I really respect the technology that the dotMobi foundation has developed - we use DeviceAtlas at Handi, for instance. I believe .mobi sites are still useful for hosting services that are purely focused at the mobile context - location-sensitive websites, for instance. However, the battle for large-scale deployment of general purpose content in that TLD seems to be over before it even began.
The other day I was unpacking a brand new N82, courtesy of Richard and Nokia. One of the most visible changes from the N95 model is the out-of-the-box behaviour of the N82 accelerometer - display switches from upright to landscape quite often. GPS got better too - advanced sensors in mobiles are clearly the new norm. It’s apparent that soon our devices will know a lot more about the mobile context than they ever did.
On the other end of device spectrum, my father just bought a brand new DSLR camera and he’s getting better and better at it. It takes great pictures if used properly. I can’t see the mobile ever completely eradicating the DLSR niche market. Or the portable media player market (screen size matters). Mobiles are becoming getting better at doing many things at once, but of course won’t ever be as good as dedicated devices.
That said, I am spoiled. I want all of my devices to be context sensitive. My mobile. My DSLR. My home theater. My fridge and even my microwave. Researchers in academia call this ubiquitous computing and place its arrival to some point decades in the future. However, it doesn’t have to be so complicated.
Mobiles should share the context information collected from all their sensors with every other personal device. The DSLR might not have a GPS, but if I carried a mobile it can derive the location based on the time when a photo was taken. My fridge should notice that I invited a bunch of friends over for pasta by SMS, but forgot to buy the sauce.
Who or what should be driving the specification for this kind of context-sharing standard?
We all agree that Twitter is amazing, Jaiku is great in its own context-sensitive way and Ontario-based Plurk is a horizontal breath of fresh air. Yet, fragmentation of audiences on these microblogging sites is a problem - your friends might not be following you on the one that you prefer, plus interesting people are generally hard to find.
So let’s assume that ten years into the future, an open service just like Twitter becomes a standard and even gets a dedicated button on your mobile. Your friends and family are following you, more than that - industry experts and top local bloggers are automatically added to your stream (they’ve got lots to say!). This service even senses “tweets” in your immediate locale, so you can see what people think about events and venues you’re in.
What would life be like with this kind of service?
For one, it would let us have more friends (but not close friends). In a situation when key events in your life are immediately known to your social circle, keeping in touch would stop being an active task. Instead, we would simply provide feedback to events that happen to people that are important to us.
Citizen journalism would start being simply journalism. The term “news story” would refer to a fast-spreading meme, where anyone could be the reporter.
It would be extremely easy to single out people known as connectors by analyzing the patterns of information spreading - everyone would be on the grid.
Every event would have an informal backchannel, where opinions could be exchanged in real-time. This would put more pressure on performers, as even the slightest mishaps would be noticed and shared.
What else?
Every year, leading mobile designers from around the world meet at an event called MEX. The Mobile User Experience Conference is a great way to discuss the challenges that exist in the industry and more importantly, to brainstorm solutions.
This year MEX is happening in London and although MobScure is not attending, we’ve been keeping an eye on the conference proceedings. The MEX Manifesto is released every year, outlining the top trends in mobile. Take a look and in case you have an interest in mobile software development, make sure to tackle at least one of them!
- Content itself will be the interface of the future
- Handsets are no longer just for the hand
- Fragmentation is the enemy of innovation
- Fashion is a stronger motivator than functionality
- The developing world is the new frontier for mobile user experience
- Search requires a radically different approach in the mobile environment
- Intelligent contact lists are the future centres of the user interface
- Mobile payments herald the next generational shift
- Users as individuals: uniquely complex and contradictory
- The potential of smart voice
Read the manifesto (PDF)
“Yeah, sure… Can you please text me your email?” said the girl next to me. She was talking on her mobile for about ten minutes during the SkyTrain ride and needed to wrap up just as we were heading into the tunnel following the Stadium station.
Why is it necessary to invoke multiple means of communication right in the middle of an active conversation?
Voice calls are great. They are real-time and carry a lot of information, in words and intonations. Yet, they are analogue and lack the link to the digital world. When we have to request a file or a URL it’s easier to turn to an all-digital medium like e-mail. Because of cost and usability matters, texting often acts as an intermediary between the mobile and the web. This multi-modal approach works - but it doesn’t have to be that difficult.
John had an excellent idea - record every call and convert its contents into text. There is some use to simply having a transcript of things you’ve said, but the interesting consequences arise from having a computer analyze the conversation in real-time:
- When the mobile hears the word “file” and the contents description it can reasonably well establish what document is being discussed a-la OSX SpotLight (assuming a link to a storage cloud)
- Names spoken in the call can be mapped to names from the address book with reasonable accuracy
- Actions and dates are also quite clear in verbal exchanges (”send”, “tomorrow”, “spreadsheet”).
The beauty of it is, all the technology required is already there - someone simply has to put it all together. Us?
In other news, MobScure received the post of the week award at the current Carnival of the Mobilists (thank you!). Head to Symbiano-Tek for this week’s best mobile blogs from around the world.
A few hours ago my N93 buzzed twice within three seconds, indicating two incoming text messages. They were from my sister and my roommate, who bumped into each other in the nearby resort town of Whistler. Neither of them expected the encounter and described it as a fun, yet awkward moment. Their first thought was to let me know about it.
It’s true, there are times when we care about location - when planning a trip or looking for a sushi restaurant, for instance. However, proximity is far more important in the mobile context since we are often already close to something - or someone - of value. Unlocking that benefit without an action on user’s behalf is huge - and it’s coming very soon.
A hot new service today is BrightKite - “the location-based social network”. Their goal is to help friends and strangers find each other in places they visit - which is exactly the right idea. What sucks is the need to manually check in at every location. Too much work for me!
However, the real killer feature is their integration with the Yahoo FireEagle geo-location agrregator. Once Yahoo hooks the Yahoo Go! platform to it, many web-services, including BrightKite, will instantly know the location of their users - no effort required. And more importantly, who or what their users are close to.
Very soon, you will be running into people you know a lot more often. Will it make such encounters something quite ordinary?
