22, May 2016

Google I/O 2016

Hi all!

While I usually write about comic conventions, there was a technical convention that I was lucky enough to attend this past week. I’ll try not to get too much into the weeds on the various talks, but, if you’re interested, message me, and we can talk about code and nerdiness.

Google holds an annual developers conference, known as I/O, every year. Link here. This year it was held at the Shoreline Amphitheater and surrounding area in Mountain View next to/on the Google campus. If you’ve never been, Google has streets of buildings for their campus. They have giant statues of Androids from every major release as well as random statues throughout the area. You see employees riding the multi-colored bicycles, the self-driving cars, and the giant place marker for the Maps team. It’s quite impressive, especially when you think about how the work going on inside those buildings has shaped the world. In all fairness, you can probably say that about a lot of companies’ headquarters in Silicon Valley. We have a lot going on here.

Anyway, on to the conference!

Day 1: The first day was a bit chaotic. It was incredibly warm (90+ degrees), and the area had little to no shade. 7000+ people packed into the Amphitheater to watch the keynote, which started at 10am sharp. We found some shade on the lawn and stayed there instead of our assigned seats, which were in the sun. Google is working on a ton of new things, and they were happy to showcase a lot of them.

  • Google Home: While Amazon has Echo, Google is working on its own home companion that will allow you to listen to music, control your Nest and other integrated products, and answer questions for you. A bit Big Brother-ish, but it does help advance machine learning.
  • Allo: Allo is a smart chat program that can learn from your usual responses and respond with smart suggestions to people. There are games you can play with it and your friends in chat, it’s integrated with YouTube for video, and it even offers an incognito mode that deletes your chats once you are finished with them just like Chrome’s incognito mode. Unclear if this is flat out replacing Hangouts, but I assume so.
  • Duo: Google’s new video calling service that uses your phone number for video chat. You have a live preview of the caller, which is semi-creepy. The guy demoing it hung up on his kids in front of millions, which was easily the funniest part of the keynote.
  • Android N: There isn’t a name for the new version yet, so you can vote for it here. This version will be integrated with Vulkan, has a lot of compiler adjustments to improve the runtime and reduce the amount of storage needed for an app. You’ll be able to clear all notifications and double tap to go back to the previous app. There will also be picture-in-picture for Android TV.
  • Daydream: With everyone doing VR, it makes sense that Google is as well. Daydream is their product. While it works somewhat with Cardboard, they have a plan to  create a headset and a controller. It will work with Google Movies, street view in Maps, Google Photos, and YouTube.
  • Android Watch: Designers are now creating nice watchbands, so they look more like watches and less hideous. New messaging, ability to write a reminder on the watch, and can track your fitness. Also, your phone doesn’t have to be on for the watch to work.
  • There were a lot of updates to Android Studio for devs, improved analytics to apps, and the open sourcing of part of the machine learning api’s.
  • Android Instant Apps: Possibly one of the coolest things from the keynote was the announcement that you can view something in an app without having to install it. For example, someone sends you a Buzzfeed link, and you don’t have the app. You can open the link, which will open in the app but not actual install anything on your phone. You can have a great experience without bloating your phone.

After some craziness of trying to find food (too many people, not enough chicken), we settled in for the first breakout talk: Android N. The talks were held in pop-up domes around the area. As I mentioned earlier, there were a lot of changes/improvements in the newest version. Constraint layouts, new notif templates, split screen, easy multi-locale for international apps, ability to say exactly what you need access to in your permissions, and Vulkan support were some of the highlights.

Grayson and I split up at this point, and I went to the Games: The Google Advantage talk. There were several big things mentioned. First, there is a donations api that is being added for players to donate to favorite charities in game. It’s an easy way to help the world. Second, Google is launching an Indie Corner, a section of the Play store that will focus only on Indie games and Indie developers. It’s a great way to get your game out there, and you can submit your game to them here. Third, there are a ton of enhancements to the player stats. You can see all kinds of information about your app and figure out the best way to use that information. Fourth, they are introducing video recording and live streaming to YouTube. Twitch has a pretty good handle on this market, so I’m curious if Google will be able to penetrate it.

Next was a talk on Google Play, which was also somewhat games related. There is a new beta program that allows for private feedback and new ways to experiment with icons and descriptions to see which ones users really respond to. The Developer Console has enhanced benchmarks and reports, along with a mobile app. There will be an Early Access section in the Play store for early adopters and those apps that offer beta programs. The presenters also focused on how important it is to keep your app small as the majority of emerging markets have smaller capacity phones. They also went through the algorithms a bit for Play Search, Recommendations, and Top Charts to explain how your app description can impact your appearance in these.

The last talk of the day was on Vulkan. Vulkan is a new low-level graphics api that offers higher performance but requires you to know more of what you’re doing. You have a lot of control over buffers and threads, and the presenter kept saying how easy it is to mess up. However, the rewards worth it as your games will run faster and perform better. This was a highly technical talk, so let me know if you want more details.

Day 2: The weather was cooler, although I was still sunburned from the first day. So, we spent more of the day exploring instead of panicking about talks we couldn’t get into.

The first talk I went to was the Realtime Database using Firebase. Databases are what I have a lot of experience with in my real job. Firebase, which is a whole suite of products, offers a cloud database platform that handles synchronization and offline mode surprisingly well. It uses JSON and is really like a tree of lists and objects instead of the usual SQL tables that I work with.

Afterwards, I went to another Google Play talk, but it included most of the things I’d heard about in my previous Google Play talk. They went into instant apps a bit more and showed off some of the Firebase stats that are available for apps.

I attended a couple of VR/AR (virtual reality vs augmented reality) talks in the afternoon. First was the Daydream talk, where the team discussed challenges of VR and their plans/attempts at making a controller that was responsive and felt natural. It’s kind of like a Wiimote with a circle touchpad at the top, which is quite different than the Oculus controllers. Also, #SpeakerTree. The second talk was around Project Tango, which is more computer vision. Grayson worked extensively with this in his Camerascript app, so I was more familiar with mapping the world around us. The new phone being developed by Lenovo will have the depth sensor needed to truly experience AR, and it’s really impressive what can be done. The presenter demoed several cool AR app such as previewing furniture at size in your home to see how things would really look and a dinosaur experience where you can see them at full scale and explore. There are a lot of educational and fun opportunities available.

Day 3: Tired, sore, and exhausted, we headed back to Mountain View for a third day. While 90 on Wednesday, it was now in the upper 50s with a stiff wind. And I thought the weather in Texas was bipolar.

My first talk was on Machine Learning. As you know, I’m a huge sci-fi fan. Terminator: The Sarah Connor Chronicles was one of my favorite shows, and they actually spent a lot of time on this subject. Computers can already beat us at chess, so one was taught to play Go, a game that is considered incredibly intricate and strategic. The AI dominated, and the second game shocked the world. Read up on it if you haven’t already. This talk was more of a discussion as the team said where they thought AI was going, the next problems they are trying to solve. They explained how medical and speech advancements are next, but they are still very limited. For example, they cannot transfer knowledge to another domain. The AI that plays Go cannot apply what it knows to anything else. Also, the system is a fixed size, so it cannot expand to learn more. When a system can translate English into English, we will enter the next phase of machine learning because the system will be able to understand context, emotion, and language at the level we do. It’s fascinating stuff, and we have a long way to go before we create Skynet.

The next talk was back to games. Netmarble, the top Korean mobile game company,  was there, explaining how they used tools and tips from the new stats to dominate the Asian market. Apparently, it’s very difficult to port for Western audiences because Koreans spend 4-5 hours a day gaming on their phones, while we only spend 1-2 hours.

I attended a talk on RecycleViews, which was more for Grayson. They talked about the code changes in views and how they better handle animations, children, etc. The last talk I attended was on VR and Cinema. This talk by Jessica Brillhart was one of my favorites. She’s one of Google’s filmmakers and explained how the storytelling process changes for VR movies. She talked about frames, editing, and how it’s about building a world, not stitching together single frames. You try to guide the viewer towards an experience and a story, but they don’t have to follow your directions. You have to remind them that they can move around, turn to see the world. The viewer has to discover things for themselves, be the appropriate height, and have a human experience in order to truly enjoy the movie. She’s made a ton of movies that she’ll be posting to her YouTube channel. One of them star the creators of Myst, Robyn and Rand Miller. Fun fact: Myst is her favorite video game.

Well, I think this post has been long enough. Google I/O is a great experience. You spend three days learning and meeting fellow tech people. There’s great food, parties, and a chance to try out new tech. It’s a bit pricy, but something you never forget.

Thanks for making it to the end! Enjoy my photos!

~Emily

The Keynote

The Keynote

Some of the domes

Some of the domes

Code bus!

Code bus!

Inside one of the domes

Inside one of the domes

Self driving car

Self driving car

Inside another dome

Inside another dome

Project Loon

Project Loon

Another talk

Another talk

VR

VR

Androids at the entrance

Androids at the entrance

Android shrub

Android shrub

Grayson in front of the I/O sign

Grayson in front of the I/O sign

Me in front of the I/O sign

Me in front of the I/O sign

Another view of the self driving car

Another view of the self driving car

Chaos

Chaos

Leave a Reply

Allowed Tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>