Posted By Simon Montford on May 18, 2016
I watched the two-hour Google I/O 2016 address so you don't have to. If you missed it and have two hours to spare feel free to click on the Youtube video (below). If you don't have time, or just can't be bothered, then here's an overview of everything that was announced, plus a bunch of screen shots as a picture is worth a thousand words, or so they say.
Google fired a shot across Amazon's bow with the unsurprising announcement that they've been working on an Echo me-too smarthome product called "Google Home" (previously Google Chirp). Privacy issues aside, like Echo it will be an "always-on" device which listens out for the words "okay Google". I know what you're thinking, and no it probably won't appreciate it if you accidentally call it "Alexa". Don't forget robots have feelings too. Amazon’s Echo has been a slow burn, since its sudden unexpected entrance but has since become somewhat of a sleeper hit. The company has sold around three million units at $179.99 a piece, so if you do the maths, that's a pretty big wedge of cash. A smaller version is also available that can be placed around the house for $89.99. Google Home will work with other services, such as Google Play Music, Google Cast and the company's shiny new chat-bot, Google Assistant which I'll come to in a moment.
Next up was Allo, a chat app that will come with end-to-end encryption, a chat-bot, and a whole bunch of emojis. Chat-bots are so hot right now! Google's announcement that they have been developing one of their own was therefore not exactly a shocker. The reason why chat-bots are so cool, and how they differ from the likes of Siri, Cortana and Google Now, is that instead of pithy Q&A interactions, chat-bots use natural language to communicate either via voice or text. This means you can have a free flowing conversation with one and forget you're talking to a robot. One could argue therefore that we now live in an era of ubiquitous computing that renders the Turing Test almost routine and trivial, however my colleague, friend, and verbal sparring partner Dr Tim Willis (PhD in Artificial Intelligence and Natural Language Processing) strongly refutes this but maybe that's because, unlike me, he has yet to be duped by one!
The company then announced a FaceTime rival called Duo, which Google calls "knock-knock". In my view this feature is a liability! Imagine a scenario where you put your phone down face up, perhaps you're in a meeting, at dinner or having drinks at a bar. Suddenly you receive an unexpected incoming Duo call. It's your other half and he or she's decided to surprise you with an erotic video preview that is now being shared with everyone within cited of your phone. Oops.
Moving on to Android N - no official launch date yet. Apparently the new OS has yet to be named, so my vote is for either Nougat. When it is finally released, it will come with Daydream, which is Google's new VR platform. In fact, it appears that Google is doubling down on VR. Apparently 50 million cardboard VR headsets have been sold so far, so launching a fully fledged VR platform makes logical sense if Google is to compete with the likes of Facebook who have stolen a march. Daydream promises under-20-milli-second latency and when released this autumn will come with a plethora of news, entertainment, and gaming content. One major downer is that Daydream will likely only work on new phones that are Daydream-ready. The company showed off a reference design for a VR headset presumably to encourage OEMs to support the platform. Oh and there was also a motion controller with a touchpad, so best start saving your pocket money now so you can afford the cost of entry, otherwise your daydream could very well turn into a costly nightmare.
Another big announcement is the second coming of Wear. Version 2.0 will not only make the Apple Watch look like the flawed product that it is, but will also make those that purchased one weep with regret (Apple Watch remains the most popular smartwatch, but according to MacRumors, its dominance declined from 60% to 50% last quarter as a result of Android-powered rivals). I'm a smug LG Urbane owner, and I can't wait for the upgrade. It promises to enable developers to embed far more complex interactions meaning users will be able to access data from almost any app and make it show up on the watch face. Furthermore, it will also allow the watch to do an increasing number of things without needing to be tethered to a smartphone - about time!
From a personal perspective, one of the most interesting announcements was that Google has developed a tiny QWERTY keyboard. I have mixed feelings on this, because I'm a co-founder of a company that's developed a smartwatch text input system called Tusi, which uses a very different, patent-pending alternative to QWERTY, which we think makes it far superior! On the one hand, it validates our belief that there will be plenty of situations where voice input is either unsuitable (due to privacy, disturbance to others and the sheer awkwardness of speaking into your wrist), or unreliable (due to poor network connection or loud surroundings), and hence touchscreen input is overdue. If you don't like speaking to your watch in public and would like a smartwatch keyboard that's hopefully better than Google's, then Tusi is for you! Download it for free, and let me know what you think, but please do bear in mind it's an early beta.
The keynote then covered Firebase Analytics (get it now, while it's free and unlimited), and this was followed by what for me was one of the most intriguing announcements made during the entire address - Android Instant Apps. This technology offers some of the functionality of an app but only when needed and here's the clever bit, you don't need to download it! Because developers can now "modularise" their apps, only the parts that are required are accessed. This avoids having to install loads of apps that are only used very occasionally, and it also prevents apps from being "siloed" thereby making apps cross-functional. Apple take note!
The event was then bookended by Sundar Pichai (Chief Executive Officer of Google) who returned to the stage to outline the company's commitment to Machine Learning and Artificial Intelligence. He explained how they are investing in their own proprietary hardware called Tensor Processing Units (TPUs) to make TensorFlow even more powerful. TensorFlow was originally developed by researchers and engineers working on the Google Brain Team within Google's Machine Intelligence research organisation for the purposes of conducting machine learning and deep neural networks research, but the system can be used for other applications. Two examples sighted by Sundar Pichai were robotics and medical. Apparently, using deep learning it's now possible for robots to think independently; learning by experiencing and interacting with the world around it as a human would, and autonomously adapting behaviour without the need for human intervention. This is the first time robots have actually been able to demonstrate true creativity, which is a major step forward. TensorFlow is also being used in the medical sector to make prognosis and diagnosis far more accurate. Machine learning and machine vision can now be used to spot and diagnose medical conditions using online diagnosis tools, which is particularly useful for patients who live in low income countries.
In summary, Google deserves a big pat on the back for smashing it out of the park yet again. Smiles all round, providing no one mentions the European Commission's antitrust investigation into Google's search and advertising business, and possibly its Android mobile operating system. One could argue that this is an accolade, because the only reason the EU is going after Google is that the company is crushing the competition. Not only will Google's dominance be causing a few eyebrows to be raised, but I can also imagine it causing a certain Californian rival to shed a bead or two of sweat. It's been another very good year for Google, at the cost of its rivals.
© Simon Montford (WEB3IOT), 2014-2019. Unauthorized use and/or duplication of this material without express and written permission from this blog’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Simon Montford and WEB3IOT with appropriate and specific direction to the original content at web3iot.com.