What’s new in Firebase (Google I/O ’18)
Articles Blog

What’s new in Firebase (Google I/O ’18)

August 10, 2019


[TITLE MUSIC PLAYING] FRANCIS: Hello, hello. My name is Francis, and I lead
the Firebase product team. Thank you for joining us today. We’re so excited to be here. So in this session,
we’re going to go over an overview of Firebase,
highlight what’s new since I/O last year, and also talk about
ways that Firebase can help you across the different stages
of your app development. So when we started Firebase,
we had a simple vision– to help developers build
extraordinary apps. And we’ve come a long
way since the early days of expanding the
platform from being a realtime database to a broad
app development platform. And while our platform and
community have expanded, we remain focused on helping
mobile app teams succeed. And we do this by
providing a platform to help you solve
the key problems across the different
stages of app development, from helping you build
better and easier with Google infrastructure, like Cloud
Firestore and Functions, to helping you improve the
quality of your app with tools and insights, like performance
monitoring and Crashlytics, and to also help you grow your
business with Google Analytics and Remote Config and the like. So whether you’re
building a brand new app or looking to extend
an existing app, we want to help you solve
these key problems so that you can channel more of your
energy towards your users and running your business. And as you may have heard from
the developer keynote earlier, we’ve seen incredible
momentum over the last year. We now have over
1.2 million apps actively using
Firebase every month, including many global partners
that you may recognize. We appreciate so many of you
are trusting us with your apps, whether you are a sole developer
or working in large teams. We’re committed to helping
developers at companies of all sizes to succeed. So as you may have
also heard earlier, we’ve made a number of
updates over last year. In particular, we’ve
released four new products, and some of these
are built based on our own internal learnings
and tools, but most important and most valuable of all,
based on your feedback. Now, one of the things
that we hear a lot and is top of mind
for many developers is data privacy, especially
with the new GDPR to soon take effect. And so to help
developers succeed under the new
regulations, we have updated our terms of service
to include data processing and security terms,
and have also worked hard to make sure we’re
ready to meet your compliance needs with many privacy
and security standards, like ISO, SOC, so
that you can feel confident you’re building on
infrastructure with world class data security and protection. Now to hear how some of
the Firebase products work and see them in
action, let’s turn it over and hear from some
of our partners. [VIDEO PLAYBACK] – We have over 10 million
users, and a developer team of only three. There’s no way we
could have achieved this scale without Firebase. Sworkit uses Firebase to take
care of all our infrastructure, from user sign-in, to
notifications, to our data analytics in Python. As our app continues
to grow, we’re focusing more on
stability and performance, and now we’re starting to
investigate how Firebase can also help us there. – App qualtiy is key. No one wants to use a buggy
app with a nice user interface. Crashlytics really
saved us a lot of time by servicing
crashes and helping us to pinpoint the root cause. Additionally, it helps
us to prioritize. You see how many
users are affected, and then you can fix the
crash with those most users. – Once you have a
stable app, it’s time to concentrate
on user growth. We did that through linking
Firebase with AdWords. Now we’re able to
run smarter campaigns that drive deeper engagement. We saw a 40%
increase in the usage of Auto Backup, a core feature
in the Google Photos app that helps people keep
their photos safe. – In mobile games,
user retention is one of the toughest problems. When we heard about
Firebase Predictions, we wanted to see if we
could increase retention by targeting the right users. The experiment was
a huge success. By serving the in-game
promotion to users who were predicted to churn,
we saw a 20% gain in retention. – Firebase has been a fantastic
partner, giving us scalable, easy-to-use infrastructure. – All the tools we need
to make our app stable and high-quality. – User insights and
smarter campaigns so we can drive engagement. – And we’ve increased
user retention thanks to the power of Google’s
machine learning. [END PLAYBACK] FRANCIS: So thank you
to all the partners, and all of you using Firebase. [APPLAUSE] So whether you’re building
infrastructure like Sworkit, or improving app
quality like Doodle, or growing product usage
and retention like Halfbrick and Google Photos,
you can see how Firebase can help you through
these different stages of your app lifecycle. So now, we’re going to
switch gears to deep dive and talk about each of
these stages in more detail and talk about some
of the updates. And to start, I’d like to
welcome Alex Dufetel up on stage, our lead
product manager focusing on ways you can use Firebase
to build a better app. Alex. ALEX DUFETEL: Francis. Hi, everyone. Good afternoon. Thank you for coming. Something that I find
wonderful about app development is just how fast our
industry evolves. The degree of
sophistication of apps today compared to just a few years
ago is really staggering. More than ever,
building successful apps requires from developers to
carefully pick their battles. What do I build versus
what do I grab off the shelf are make
or break decisions for really anyone in the
space, from the startups to the unicorns. And one of Firebase’s
key goals has been, since it’s very
creation, to empower developers to build better, more
successful apps precisely by allowing them to make the
best of turnkey serverless infrastructure. And this is something that
I experienced first-hand. Before joining Google,
I spent five years building and growing a
startup that was powered by many Firebase products. We built a realtime
collaboration app, and initially, we built all
our sync engine fully in-house. And building that infrastructure
was incredibly difficult. And the time we were spending
doing this was time we knew would be better spent on
focusing on the core user experience of our product. So we searched for
a better solution, and that’s how we started using
the Firebase Realtime Database. And we essentially started
playing around with it, and within a week of
prototyping with my team, we fell in love
with the product, and we decided to refactor our
whole app with the Realtime Database. And with hindsight, it was
one of the best engineering and business decisions
that we made as a startup, because it allowed
us to go faster and it helped us focus
on the features that matter to our users instead
of building infrastructure that essentially wasn’t
core to our business. And to me that is what
Firebase really is about. Of course, since
then, Firebase has become a much broader
platform Firebase developers, too, have evolved. Our community is richer
and more diverse than ever, and the typical size of Firebase
teams is also growing rapidly. So today one of
our core focuses is to adapt, to make Firebase
just as powerful, regardless of the size of your team
or the scale of your app. That’s why we build products
like Cloud Firestore or Functions, that essentially
are agnostic to scale. Going forward, you’ll also
see us invest increasingly in tooling to help you better
monitor your infrastructure and test your code,
and in assisting you to build more secure apps. So now let’s talk about some
of these exciting updates that we have for our
Firebase developer tools. Six months ago, we
launched Cloud Firestore, our next-generation database
that stores and synchronizes app data at global scale. Cloud Firestore is a
NoSQL document database that addresses some of the
toughest problems of dealing with app data. Unlike traditional
databases, Firestore allows you to access your data
directly through client SDKs, removing the need to
build a proxy layer between your data and your app. It also allows you to
listen to your data as it changes in
realtime, making building rich interactive
apps much, much easier. And also, it automatically
caches the results of your queries, so
your app essentially works offline out of the box. Now, that’s incredibly
important because users today expect your apps
to work offline, and yet it is still
incredibly difficult to build if you’re building
it from the ground up. Cloud Firestore is also
a strongly consistent, replicated database. It offers a richer
set of queries than its predecessor,
the Realtime Database, and yet all those new queries
have the remarkable attribute of scaling with the result
sets, not the data set. What that means is that you
won’t be refactoring your app and rebuilding your
backend every time you’ve 10x’ed your users. Google is well known for its
industry-leading database systems, like Bigtable and
Megastore, and now Spanner. Cloud Firestore builds
on that knowledge to give you a powerfully
simple database to serve your users
around the world. Adoption of Firestore
has been amazing. 200,000 databases have
been created since October. That’s more than
one database being created every other minute. And if you’ve used
Cloud Firestore already, it won’t come as
a surprise to you that a very large
portion of these users are also using Functions
in their workflows. Cloud Functions, which
we launched last year, lets you automatically run
backend code in response to events triggered by Firebase
features and HTTPS requests. Firebase users
all over the world are using Functions to build
deceptively simple workflows. A very interesting
use case for us is the Hamilton app, in
particular because their usage is so spiky. When they start a new lottery to
sell tickets, millions of users connect at the
same time, and they get millions of requests in
a very short period of time. Cloud Functions
automatically scales without needing any human
action or intervention. And Hamilton also only pays
for the resources they use, instead of having to
massively over-provision. Last month, we upgraded our SDK
with a number of new features based on your feedback. We now have client-side SDKs
for calling HTTPS functions. This makes integrating backend
functionality into your app much, much easier. We’ve also added
TypeScript support in our tooling, which was
a very popular demand. And another big improvement
has been the expansion of the ability to unit test
your functions before you deploy them, making sure, basically,
that you can catch problems early on before
they hit production. We’ve also added new monitoring
tools to Cloud Functions. This makes it easier
to test and debug, from prototyping to production. You can also keep tabs on
any errors in your functions and track your functions’
performance, latency, and memory usage. When we say that
Firebase empowers you to build better apps, we
mean more than just scalable infrastructure. We want to help you with tasks
that are equally crucial, such as user authentication. Firebase Auth is currently being
used by hundreds of thousands of apps, and authenticates
millions of users every day. Over 80 million people sign
into an app using Firebase authentication. Keep in mind that most apps
just keep you signed in, so these are new sign-ins. Last year, partnering with
a digits team from Fabric, we introduced phone
authentication, which has seen incredible uptake. And now we keep
working to make sign-in as frictionless as possible. To that end, we just released
passwordless sign-in. This allows your
users to essentially just click on a
button in your app, and then click on a link they
receive in order to sign in. With no more
password to remember, you have one less thing
that can cause friction in your sign-in flow. We also keep integrating
all these additions into Firebase UI, our open
source library providing UI flows for signing in. Firebase UI is available on
iOS, and Android, and web, and it’s being used to sign
in users in tens of thousands of apps across these
three platforms. As you heard earlier today,
machine learning and AI are a huge focus at Google. We feel that machine
learning has the potential to unlock so many
interesting challenges, and we want to make sure
that all developers have easy access to the power
of machine learning. That’s why we’re launching
ML Kit into beta, bringing the power of machine
learning right into your app on Android and iOS. ML Kit provides five
out-of-the-box APIs, including text extraction, face
detection, barcode scanning, image labeling, and
landmark detection. The base APIs let you easily
build ML-powered experiences without needing extensive
background in machine learning. These APIs are both available
on-device and in the cloud. For those with advanced
needs, ML Kit also lets you upload your own
custom TensorFlow-like models, and serves them through Google’s
powerful global infrastructure, so your app can dynamically
retrieve models and evaluate them on device without having
to bundle the model with the app binary. To illustrate this
with an example, let’s take a look
at VSCO, a partner who integrated with ML Kit. VSCO is a photo app that enables
creators to transform photos with powerful editing presets. With over 130 options, it
can be difficult to select the best presets for an image. So using ML Kit
and a custom model, they were able to build a
recommendation feature that automatically recognizes
the content of an image in order to suggest
the best presets. For example, if the
creator edits a portrait, they can automatically recommend
ideal portraiture presets. Importantly, all the
subject matter recognition is done on-device,
protecting user privacy. We plan to heavily invest in
ML Kit over the upcoming months and years. If you’d like to
learn more about it, be sure to check out the
section about machine learning on mobile
devices tomorrow at 3:30. This covers some of the updates
of how it’s easier to build your app if you use Firebase. If you want to
learn more, be sure check out our session on
building planetary scale apps tomorrow at 9:30. And with that,
I’ll turn the thing over to my teammate
Kristen, our PM lead focus on helping you
improve app quality. [APPLAUSE] KRISTEN JOHNSON: Thanks, Alex. Good afternoon, everybody. I joined Firebase along with
the rest of the Fabric team a year ago. It was a bit of a
homecoming in a way, because the goals of
Fabric, even back when we launched Crashlytics
six years ago, were to help developers
build their apps, understand their users, and
grow their businesses, just like Firebase. The investment that
Firebase and Google are making in the developer
community is astounding, and I’m excited to
be a part of it. A critical part of having
been successful app business is having a stable
and performant app. We’ve found through Play
Store research studies that 42% of one-star reviews
mentioned crashes or bugs, while an astounding 73%
of five-star reviews mentioned speed,
design, and usability. Now, the relationship between
app quality and customer satisfaction and
retention is undeniable, and we think Firebase can help. Over the past year,
we’ve been working to make a number of improvements
to our app quality products so that they’re ready for the
largest and most complex apps out there. Let’s take a look at some
of these improvements. One of the biggest changes
we made over the last year was bringing Crashlytics
to the Firebase console, first in beta, then in GA. Crashlytics is our
best-in-class crash reporter. And we are excited
to share today that it’s currently
helping app developers keep their apps stable on over 3
billion devices worldwide. We are honored to be able
to help so many of you build a better user experience. A couple months ago we
integrated Crashlytics with Google Analytics
for Firebase so you can see the trail of
events leading up to a crash. We call these breadcrumbs. In fact, breadcrumbs is
one of the main reasons that our partner
WeTransfer decided to try out Firebase
Crashlytics when we launched it in beta in October. WeTransfer is a company
that helps its users share large files easily so that they
can stay in the creative flow while they’re working. They have over 40 million
active users that share over 1 billion
files every month, and it’s critical
that each one of them goes off without a hitch. Now, the WeTransfer team has a
goal, maybe even an obsession, with getting as close to
100% crash-free as possible. Since this isn’t
always achievable, they have a health
metric of 99.5%. And if the app ever
dips below that, they stop everything
and fix crashes. Breadcrumbs shows
them the actions that a particular user takes in
a specific session leading up to a crash. This makes debugging
so much easier, and the team no longer
needs to figure out how to produce the
problem themselves, which can be a lengthy
and frustrating process. As you can see here,
breadcrumbs has been a game-changer for them. And while 100% crash-free
is an audacious goal, we’re proud that Firebase can
help them chase their dream. Another improvement
in Crashlytics that we’re launching today
is an integration with Slack. We want Firebase to work
seamlessly with the tools that you already use. Now when there’s a
crash that’s affecting a large percentage
of your user base, Firebase will automatically
alert your team and the Slack channel of your choosing. We have many more innovations
to come, so stay tuned. We’re super excited about
the future of Crashlytics, so in order to focus our
efforts on improving it, we announced a few months ago
that we’re stopping our support for Firebase Crash Reporting. The product will continue to
be accessible until September, but we encourage you to try
out Crashlytics and upgrade to it today. Of course crashes are important,
but an often overlooked but equally important
part of your app quality is its performance. At last I/O, we launched
Firebase Performance Monitoring, our tool for
getting performance insights from your user’s perspective. Over the last year, we
battle-tested the SDK, and we’re excited to announce
that Performance Monitoring is graduating out of beta today. Along with this
announcement comes a few new exciting features. We’ve heard feedback
that there is a lot of useful data in our console. Users can slice and dice by
things like country, carrier, and other attributes. But we’ve also
heard that it can be tough to find the most
important issues among all of this rich data. Today we’re announcing the
addition of the new Issues Feed where we do the digging for you. You’ll find it at the top of
your Performance Monitoring console dashboard, and
there we highlight issues that you should be
paying attention to, such as slow app start time
or failing network calls. You can also dig into that issue
and find the culprit behind it. We’re also now collecting
more metrics out of the box than ever before. On top of app start time
and network requests, we now have data on
slow and frozen frames. What’s cool here is that
we collect this information automatically for
every page in your app. Between custom traces
that you set up yourself and our out-of-the-box metrics,
Performance Monitoring has your back, helping
you understand your user’s experience
out in the wild. Now, one of our
partners, Swiggy, is one of the biggest food
delivery services in India. They started using Firebase
Performance Monitoring about a year ago. As soon as they dropped
the SDK into their app, they realized they had a big
problem with app start time. And as we all know, first
impressions last forever. Performance
Monitoring helped them diagnose the issues that
were causing the app start time to slow down and monitor
those fixes over time. In the end, Swiggy
was able to make app start time improvements
of 2x on Android, and an amazing 12x on iOS. Overall, their customers
noticed, and were delighted. Also, Swiggy saw a 2.2%
increase in daily orders. And these type of
increases really matter when you have over
350,000 orders every day. Now, there’s one more
piece of the puzzle that I’d like to talk about. At Firebase, it’s
always been vitally important that our products
work cross-platform. In that spirit, today we’re
expanding Firebase Test Lab to include iOS. [APPLAUSE] Now, Test Lab provides you
with the physical and virtual devices that allow
you to run tests that simulate actual usage
environments across a multitude of hardware and OS versions. With the addition
of Test Lab for iOS, we help you get your app into a
high-quality state on both iOS and Android before you even
release it into the app store. One of our partners, Spotify,
goes to great lengths to ensure that each and
every one of their users has a great experience when they
use their music streaming app. They have a comprehensive
set of end-to-end tests that they run every night,
and before Firebase, they even built their own
homegrown device farm. But over time,
they found that it was hard to maintain
it and scale it, as well as found that
some of the tests would often fail unexpectedly,
and they weren’t really producing results
that they trusted. Spotify decided
that they no longer wanted to be in the business
of managing a device testing infrastructure, so they
turned to Firebase Test Lab. Spotify uses Test Lab’s
recorded test videos to get immediate insight
into where tests fail. Plus, with Firebase
Cloud storage, they have long-term
access to test assets in case a regression pops up. Spotify loves the scalability
and reliability of Test Lab, and by the sheer number of
devices available in Test Lab, they’ve experienced a 2x
increase in testing throughput. Test Lab has given
Spotify confidence that their tests
will run quickly and reliably so they
can stay focused on what’s important
to their business, bringing the best music
to fans around the world. We’re excited that Test
Lab is now cross-platform, and iOS support will be rolling
out over the coming months. If you’d like to get your
hands on it early and be an early tester, please
go to the Firebase console or use this URL to get
on the waiting list. If you’re curious to check
out a physical device lab and want to see it
in real life, you can go to our Firebase sandbox
over with the other pods, right out those doors right there. And to hear more about how you
can experience the same success that WeTransfer,
Swiggy, and Spotify did, we have another
talk tomorrow that focuses on app quality
at 11:30 on stage 7. As you can see,
maintaining a stable app is the first step to improving
engagement and growing a meaningful business, but
there’s more to this story. To hear more about this, I’d
like to invite my teammate Nalin to the stage. He’s a product lead focused
on helping you grow your users and increasing
engagement with Firebase. [APPLAUSE] NALIN MITTAL: Thanks
so much, Kristen. Good afternoon, everybody. So far, we’ve talked
a lot about how Firebase can help developer
teams build apps and improve app quality. But as you get ready
to launch your app, there’s another important
area that mobile app teams focus on– app growth. And this is a problem
that’s personal to me, because prior to joining
Google, I was building consumer mobile apps like many of you. And for one of
our apps, Giffiti, we were lucky enough to have
a really successful launch. We made it to the
front page of Reddit. We had over 1,000 five-star
reviews in the first 24 hours. Things were going
really great until we looked at our analytics and
realized that users only used our app for about 10 days
and then disappeared forever. So even though we had thousands
of new downloads every day, the number of daily
active users in our app was not growing at all. We learned that retention
and engagement are really hard problems to solve, and that
a successful launch by itself doesn’t necessarily translate
into a successful business. And as we talked
to other app teams about how they’re
solving these problems, we heard one thing
consistently repeated about the growth process. It’s all about experimentation. You can’t afford to rely
purely on intuition. Instead, it’s much more
effective to run an experiment and let data drive
product decisions. So with Firebase,
we’ve built products that help you across every stage
of the experimentation process. First, you observe what users
are doing inside your apps. Then you make a
hypothesis about how to improve their behavior with
respect to your business goals. Next, you test that hypothesis
by trying a new product or marketing expert experience. And finally, you analyze
the user response, and if it’s successful, you
can roll out the changes to all of your users. And at Firebase,
we’re building tools that help you scale this process
so you can iterate quickly, launch new features
confidently, and ultimately grow your business. Let’s take a look at
each of these steps and how Firebase
makes this easier. It starts with Google
Analytics for Firebase, which helps you
observe user behavior. Specifically, we
help you understand where your users
are coming from, what they’re engaging with,
how often they come back, and how much revenue
you’re generating. Now, with Firebase,
you can always see analytics for
each individual app. And last year we
added the ability to see anonymized analytics
for individual users with both de-bug
view and stream view. And now we’re excited
to give you the ability to level up and see
aggregated analytics for all apps in your project. So if you have an iOS version
and an Android version of your app, you can now see
how daily active users, revenue, and other metrics are
trending across your business. This means you can now
analyze data on three levels across all your apps, for
an individual app, and even an individual user. And this detailed visibility
into user behavior is critical, because
to drive growth, you want to identify
trends and find segments of underperforming
users to ultimately make a guess on how to
improve these metrics. Both Firebase Predictions
and Google Analytics help you with the segmentation. So with Audiences
and Google Analytics, you can segment users based on
what they’ve done in the past. But with Firebase, you
have an additional way to segment users– by predicting their
future behavior. Instead of looking back on
what they’ve already done, you can ask Firebase
Predictions to look forward to see what they’re
likely to do. Using a combination of
Audiences and Predictions, you can target users to
drive a better outcome. Let’s take a deeper
look at Predictions. So Firebase Predictions,
which we launched in October, applies Google’s
expertise in machine learning to the analytics
data from your apps. Based on past behavior, it
predicts future behavior, and forms dynamic
user targeting groups. Out of the box, Firebase
can predict things like which users
will make purchases and which users will churn. But if you’re tracking
conversion events in Analytics, then our machine
learning models can begin to predict events
specific to your app, like listening to a
song or sharing photos. Since launch, we’ve
made predictions on over 4 billion app users. So Google Analytics and
Firebase Predictions give you two different
ways to segment users that might be underperforming. And Firebase gives
you actions to take so you can target the
segments and improve the health of your business. You can personalize
a product experience with Firebase Remote Config– maybe to help convert
a user by incentivizing them to make a purchase– or send a highly
relevant message with Firebase Cloud Messaging to
bring users back into your app. Let’s see how each of
these actions is used. So Firebase Remote Config
lets you control the behavior and appearance of your app
without pushing updates to the app store. This is really
important because we know each new
version of your app not only takes days away from
valuable engineering resources, but your users also take
forever to update their apps. So if you use Remote Config to
control things like marketing copy, the size and color
of various UI elements, or even the layout
and flow of your app, you can quickly tweak
these values right on the Firebase console
and push updates instantly to your users. Now another powerful
aspect of Remote Config is that it lets you
personalize the product based on Google Analytics Audiences
and Firebase Predictions. Maybe you want to
incentivize users who haven’t purchased by
offering them a discount, or how about changing
the theme of your app to be more receptive
to international users? Remote Config was built to
provide deeply personalized experiences to keep
your users engaged and move them down
the conversion funnel. Another action you can use
to optimize your business is Firebase Cloud Messaging. Cloud Messaging lets you
send the right message to the right users to bring
them back into your app and keep them engaged. Whether you are using
our API or the console, we give you analytics to see
how many messages were sent and how many
messages were opened. One common practice
is to send messages to users who have not
opened your app in some time to improve retention, but
with Firebase Predictions we can proactively
reach out to users who are predicted to churn. This is a powerful
tool because it allows you to intervene
before it’s too late and gives you the best chance
of improving retention. Now as we mentioned
before, whether you are using Firebase Remote
Config or Cloud Messaging, you will want to
test these actions and make sure that
they are indeed improving your business before
you expose all of your users to them. That’s exactly why we
built Firebase A/B Testing. So A/B Testing lets you use
data to drive your product decisions. You define the variants and
the goal of your experiment– maybe a seven-day
engagement or revenue– and our system will
do all the hard work. From randomly distributing
your user base across the experiment to the
complex Bayesian analysis to tell you which variant is
achieving the best results, A/B testing makes
experimentation simple. We even give you the ability
to track secondary goals so that you can, for example,
ensure that a winner is not only improving revenue, but
is also not harming retention. And once you’ve
verified the results, you can roll the experience
out to all of your users. So let’s take a
deeper look at how top publisher Halfbrick used
Firebase to grow their app. They were looking
for creative ways to increase retention in
their game, “Dan the Man,” using a promotion where
users would receive an in-app currency
to win prizes that made the game more enjoyable. Initially they tried using
their own churn models, but like most publishers, didn’t
have the time and resources to dedicate to this. After learning about
Firebase Predictions, they wanted to see if our
churn prediction could drive the results that
they were expecting. So using Firebase A/B
Testing and Remote Config, they were able to
use the console to set up an experiment
with three groups– a control group where no
one received the promotion, another group where users
received the promotion after completing level 3,
and finally, a group that would only see the promotion if
they were predicted to churn. What the system found was a
20% increase in retention rate when showing the
promotion to users who were predicted to churn– really amazing results. So we just saw how Firebase
helps with every stage of the experimentation process. Google Analytics
lets you observe what users are
doing in your app, and it helps you find
user segments based on past behavior, while
with Firebase Predictions, you can find interesting
segments based on what they might do in the future. Once you’ve found
segments to improve, we let you target those
users with various actions. You can personalize the product
experience with Remote Config, or send highly relevant
messages to the right users with Cloud Messaging. And most importantly, you can
do this in a safe, low-risk way by using A/B testing, only
rolling out experiences to users when the data shows
you’re achieving the desired results. If you are interested in
learning more about how Firebase can help
you grow your app, come check out our session right
here tomorrow morning at 8:30. And there’s a lot of
new features coming in Google Analytics,
so I encourage you to attend that as well. That’ll be 5:30 PM tomorrow. Now I’d like to hand
it back to Francis to talk about what’s
next for Firebase. [APPLAUSE] FRANCIS: Thanks, Nalin. I am super excited about the
future of app development and the possibilities ahead. Now, as you’ve seen from some
of the updates we just shared and what you’ve heard
from the developer keynote earlier, across Google,
we’re working hard to advance the technologies
to enable developers like you to shape the
future of app development. And the most exciting part
for us is what all of you will build. Now, last year, I met
a team of developers who built a sign
language translation app to help the hearing disabilities
to better communicate. It’s truly inspiring,
because not only did it help their quality
of lives, but also for the community around them. And it’s inspirations
like this that motivate all of us in
Firebase to work hard in helping you succeed. And so looking forward– looking ahead– we’re
going to continue to deepen the integrations
across Firebase to help you solve these key
problems more effectively. Furthermore, we’re
going to be expanding the integrations across
Google, from streamlining the development experience on
the web to helping you scale better through Google
Cloud Platform, and bringing the advancements
of ML to help you to do more. Our mission for
Firebase is to help you succeed and be
the best platform available for
everybody in the world. While we’ve made good progress,
we are just getting started, and we draw our motivations and
inspirations from all of you. So please share your
feedback, and help us shape the future of Firebase. On behalf of the
Firebase team, I want to say thank
you to all of you, and I look forward to
meeting many of you out in the sandbox after this,
or through the online forums. Thank you so much. [MUSIC PLAYING]

Only registered users can comment.

  1. Google is not helping anyone just wants to be in the source of everyone.
    Example: Unity game I added Google Leaderbord, it works when it is installed as apk, but it does not work when downloaded from Play Store, so there is no problem with the codes. I told google that, but he did not fix it. Google removes the product but does not support it. I am a Play Store customer, but I can not get support. The Amazon playstore is non-productive but helps.

  2. Nothing *NEW* at all, except that ML Kit thing.
    So, If you watched Google I/O '17 last year, you just can skip this video and write your code now.

  3. A word of advice: Stay away from Cloud Functions as they have horrible support and performance. Only JS/TS, several seconds latency, cold start every time, their github issue list is a joke…

  4. Why did you change the way Firebase works with recycler view? Now I can't update to latest version of appcompat because firebase has all these changes and is complement with the same versions!

  5. It's not nothing new like the other comments say, it's a lot new and all of these people are amazing. Thank you guys for a great presentation! =)

  6. Really needs a c# library for windows development. I have an app and I want to make a windows version, but there just isn't a good way to interface with it from c#.

  7. Instead of giving this talk, his time would have been better spent fixing that damn shaky camera mount that's ruining all the videos in this room.

Leave a Reply

Your email address will not be published. Required fields are marked *