
※Notes記事では、英語のセッション動画やポッドキャストの内容を(雑に)英語でメモに書き残すことを行っています。本記事は、あくまで動画を見ながら、参考程度に読んでいただくことを想定しています。Notes記事には雑メモ程度のものだったり、書き起こしのようなものもあります。これから実際の動画を見る際には、本記事の内容が少しでもお役に立てば幸いです。(内容において不備、誤字脱字等ありましたら気軽にご連絡いただけると嬉しいです!m(__)m)
本記事は、What’s New in Firebase (Google I/O'19)の記事です。
The purpose of Firebase#
Help mobile and web app development success
MLKit#
Vision
Natural Language
Custom
Object Detection & Tracking API + Translation API
AutoML Vision Edge#
Upload a set of images (Training data set) and traing
Cloud Firestore#
- graduated from beta
Collection Group Query#
- query all sub collections
Cloud Functions#
Scheduled Functions
- like a cron job
Remote Config Event
Cloud Functions Emulator#
- For testing the written code locally without running remote db
Firebase Test Lab#
Supports Wear OS, Android Q, Android App Bundle
Improved games testing
flaky tests
Firebase Performance Monitoring#
Exporting for deeper analysis
BigQuery Integration
Performance Monitoring for Web App
DOM Interative
First content paint
Fabric + Firebase#
Phone Number Auth
Crashlytics
Latest Release Metrics
Slack Jira, PagerDuty integrations
Export Crashlytics data into BigQuery
Future of Crahslytics Beta#
Next generation of crashlytics beta
App Distribution: g.co/FirebaseAlpha
Roadmap: g.co/FabricRoadmap
- March 2020 => End of Fabric
Google Analytics for Firebase#
Filter almost all properties
Create new custom audience builder UI
所感・まとめ・個人的に印象に残ったことなど#
MLKitのモデル学習させるまでやモデル作成までの流れがかなり簡単になっているのでよりとっつきやすくなっていそう。
BigQueryとより多くのFirebaseライブラリが連携しているので、プロダクト導入している場合は、アプリエンジニアでもBigQueryかけたりAnalytics分析とかしていくとできることがかなり増えそう。(やらねば。)
Cloud Functionsとかも触れていないのでRemote Configとの連携とかも含めて、使っていく。
以上です!