2019 Twin Cities Google DevFest

A few notes on the 2019 DevFest run by the Google Developer Group–Twin Cities on February 2.

Venue — the venue moved across the street to the University of St. Thomas Law School building, with a huge atrium that served as the registration desk, keynote and closing address hall, lunch room, schmooze room and afterparty bar. This worked out better than the School of Management building of previous years, since the atrium allowed for a mix of theater seating and round tables right in the center of the action.

App — the DevFest app added new features. In 2018, the app moved out of the Google Play Store to an HTML5 website. Now offline storage allows the user to save sessions to a personal schedule and access it without an internet connection. To use this feature, you need to log in with a Google account. A new button on every page encourages the user to install the app as an icon on a mobile device home screen, so it functions more like a standalone app. Session feedback is incorporated into every description page. Overall, the process of finding interesting sessions, adding them to a schedule, and providing feedback was fast and intuitive. My only gripe was a disconnect between the speaker bios and the session pages. Clicking on the speaker name at the bottom of each session brought up a Twitter page. The speaker bios provided in the speaker directory were different.

Sessions — Google sent Ben Morss (Accelerated Mobile Pages and Progressive Web Apps can work together), Doug Stevenson (Build and scale mobile apps with Firebase and Google Cloud), Martin Omander (Focus on your code, not infrastructure, with Cloud Functions), Lyla Fujiwara (WorkManager the workshop), Sean McQuillan (Designing with Constraints), and Tiem Song (Adding Jetpack to your Android app). In other words, the Android track receives the most emphasis with four advocates, followed by Web and Cloud (one each).

Compared to the past two years, IoT and machine learning fell back from Google’s favor using this metric. I attended all three machine learning sessions this year and only saw the dog/cat (or puppies/basket of muffins) classifier example once, which is an improvement. IoT continues to suffer from uneven tooling, a convoluted development path, and limited hardware support even as Google invests in solving the last problem first: creating a platform for managing devices and huge piles of data. Meanwhile, my LiftMaster garage door gets no WiFi access until Chamberlain Group, Inc. gives me ownership of my garage door and mobile phone position data.

Afterward — The afterparty bar and appetizers were in the same room as reception this year; last year required donning jackets and hats and trudging into the winter night in search of an outside venue. Those who left early, skipping the usual raffle for a Google Home Mini and two Chromecasts, missed the final surprise: a hot dinner buffet of cheese and mushroom fondue, potstickers, meatballs and chicken wings. Agosto (a Tier-1 Google Partner with a distributed video platform product and custom development) deserves a special shout-out for the level of support they gave this conference, as well as ongoing support for the Twin Cities Google Development Group (gdgtc.com), and the Twin Cities Python User Group (python.mn).

Leave a Reply

Your email address will not be published.