In early 2008, I had been immersed in the agency version of web development for a whopping three years. Javascript and mobile were exploding, and I had begun to look around for a conference to attend to keep up on the new approaches other developers were taking in the wake of these shifts.
In March of that year, Google announced their first developer conference, focusing on “AJAX, APIs & Tools, Social, Geo, and Mobile.” It seemed like exactly what I was looking for, so I learned how to do things like “Get approval,” “Save receipts,” and “Fill out an expense report.”
I flew to SFO, conquered the BART system, and got to see Marissa Mayer and Steve Souders speak. I agonized over which super-relevant technical sessions to attend, as each choice meant I was choosing not to attend several other equally relevant technical sessions happening at the same time. In short, it was an embarrassment of riches, and I was flabbergasted at the depth and quality of the information being shared. I also ate lots of free Google food.
The experience of gathering in one place with literally thousands of other developers and hearing real-world advice from some of Google’s smartest engineers proved to be very, very worthwhile; five years later, I’m lucky enough to still be attending. Over the course of those five years, lots of things have changed. Android has grown to take over a huge percentage of the conference, and the proliferation of hardware giveaways has fed the growth of a decidedly less-developer-y population at the event.
The current incarnation of Google IO can be broken down into two main buckets: the keynotes and the sessions. The keynotes have without question been turned into a bit of a media circus, shifting from in-depth explorations of Google’s unique culture of technology (See: Marissa Mayer’s 2008 Keynote) to fairly straightforward product announcements and cheerleading.
This year’s keynotes were a good example of this: two multi-hour marathons packed with product announcements and X-Games-esque stuntery. We saw the announcements of Jellybean, Chrome on iOS, the (actually really great!) Nexus 7 tablet, and the (really puzzling!) Nexus Q. We saw Google Glass prototypes literally dropped from a blimp into the conference center via skydiver, and an awkward hard-sell of the new event features in Google Plus.
It’s understandable that these are the highlights that feature heavily in the Google-edited video below, distilling three days of conference into a slightly more manageable three minutes and forty-one seconds:
While this video does a good job of communicating the sheer scale of the conference and breadth of its content, what it doesn’t reflect is the depth. The technical sessions that follow these showy keynotes have maintained an impeccable standard of quality – the only noticeable shift over the years has been in the technologies they cover (the extra nerdy among you can observe the rise and fall of various technologies and platforms over time by examining the session schedules across all five years at the following links: 2008, 2009, 2010, 2011, 2012).
These sessions are invaluable for the sort of development that we do as an agency, as there are nuggets of hard-won, real world experience strewn throughout. Examples? This year’s conference featured a session from Youtube’s mobile developers, discussing their approaches to handling HTML5 video browser quirks in production at Youtube scale. Steve Souders injected some data-based sanity into the webfont explosion (many developers build in an on-load delay to avoid FOUT / users perceive delay negatively / is setting your most important content in custom webfonts always the best idea?). These are just the first two that come to mind — every session was packed with insights like these.
There was also a common theme of tooling in several of this year’s sessions. The ever-evolving influence of Javascript and HTML5 coupled with Google’s culture of focus on performance has resulted in an unmatched and constantly-evolving set of developer tools built into Chrome. Some of the Chrome-specific sessions focused on how to bring these tools to bear on the real-life performance problems that developers encounter every day.
It was particularly encouraging to see a renewed focus on UI performance and user experience in this year’s sessions, as these topics sometimes seemed to get short shrift at previous IOs. There were several good discussions of identifying and resolving “Jank” found in front-end experiences, and almost any talk that mentioned a form was touting of the user benefits of the new autocomplete interface built into the Places API.
It’s worth noting that even those with an established familiarity can benefit from the discussion – the reinforcement of technological decisions made is valuable in its own right. We worked on a fairly in-depth HTML5 Canvas implementation earlier this year, and it was encouraging to have many of the approaches and assumptions we made under deadline validated by the engineering firepower at Google.
If you are a developer working with web technologies of any sort, I can’t stress the value of the sessions at I/O enough. It is absolutely worth jumping through whatever hoops need jump-through-ing to attend.
The prospect of free hardware, coupled with the quality of the content and general over-the-top nature of the event have made the process of Google IO registration fairly competitive, so it’s best to keep a close watch on when things might be announced. Peek in on the Google Developers page whenever you make your biannual stop into Google Plus, or just follow the Google I/O Twitter account. Good luck, and hopefully I’ll see you next year’s I/O!
Figure 1: a tweet detailing my Wife’s reaction to watching me select which sessions to attend:
Sarah’s impression of me at Google I/O (in nerd voice): “I wish I could clone myself so I could go to ALL the sessions!”
— adamkempa (@adamkempa) May 20, 2010
