Solving Android Fragmentation with Trigger.IO Forge and Intel’s Crosswalk

The State of the Hurt

Anyone developing for Android will be all too familiar with the compatibility issues that come up when trying to run apps on older Android devices. Historically, even the act of porting an app from a desktop browser environment to Android has been fraught with issues as the Android WebView lacked many basic HTML5 features and suffered from implementation problems.

While Google did much to rectify this after they started shipping a Chrome based WebView with Android 4.4 the reality is that, at the time of writing, Android 4.4 or higher devices still only account for 44.2% of the marketplace.

The remainder is made up of 7.3% devices still running Android 2.x while a massive 48.5% of devices are still running some version of Android v4.x with various non-Chrome versions of the Android WebView.


Image from:

Finally, if you thought it couldn’t get any worse… even if we assume that at some point in the future we have a majority of devices running 4.4 or higher there is STILL no guarantee that your app will run smoothly across Android versions as it is entirely up to the vendor whether a given device will be receiving regular updates of the System WebView!

Against this background it therefore gives me great pleasure to announce that we have integrated Intel’s Crosswalk technology with Trigger.IO.

Starting from today your Forge apps only have to target a single web runtime for Android which covers 92.7% of all Android devices while retaining full access to all of our Forge API’s.

What is the “Crosswalk” Forge target?

Android versions before v4.4 shipped with a proprietary System WebView which lacked many basic HTML5 features, performed poorly and suffered from compatibility issues.

Android versions higher than v4.4 use a Chrome-based System WebView which, in turn, is based on the open-source Chromium code base which closely tracks HTML5 implementation standards.

Up till now, developing hybrid apps under Android meant that your app would be hosted by whichever System WebView shipped with the device forcing developers to take a “lowest-common-denominator” approach and spend an inordinate amount of effort to ensure rendering compatibility.

What the Forge crosswalk target does is to embed a build of the latest versions of the open-source Chromium WebView directly within your Forge application, allowing your app to execute in a uniform environment irrespective of the System WebView shipped with a particular device.

This means your app will behave exactly the same no matter which Android device it is running on!

How well does Crosswalk perform?

Apart from the tangible benefits that come from being able to target a uniform WebView, the Forge crosswalk target also provides significant performance improvements over all versions of the Android WebView, even those based on Chrome.

HTML5 Conformance

We checked a variety of Android devices against the HTML5TEST Benchmark for HTML5 conformance and found some interesting results:



The older devices showed the most significant improvements over the default System WebView, in all cases exceeding even the iOS System WebView in compatibility scores.


Our biggest surprise from HTML5 conformance testing was the marked improvement we saw even for devices that shipped with a Chrome-based System WebView. Even the latest Chrome-based System WebView scored a couple points lower than Crosswalk. We attribute this to the fact that Crosswalk are able to more rapidly integrate and ship improvements from the upstream Chromium project.


Something else we noticed was that there was some variation in crosswalk scores between devices. Closer investigation showed that, in all cases, the variation came down to the following HTML5 features which required hardware support:

  • WebAudio is only available on ARM devices that support NEON optimizations which were introduced with the ARM Cortex-A8 processor.
  • Newer Video Codecs require specific hardware support for decoding which is not present in some older Android devices. Our recommendation is that if you are in doubt, use H. 264 as support is a requirement for all Android devices on the market.
  • Support for the universally loathed DRM and Media Source extensions are tied to Video Codec support and may therefore not be available on some devices.

Javascript Performance

For Javascript performance we evaluated the same set of devices against the Octane 2.0 Benchmark:

Octane 2.0 Results

The impact of Chromium’s use of Google’s V8 Javascript engine is clearly visible in these results with older devices showing as much as a 60% improvement in execution speed.

Again, the advantages of crosswalk are evident even on newer devices where we see an 8% improvement on identical Nexus 7 hardware between the older Chrome-based WebView on Android 4.4.3 vs Android 5.0.2.

How do I use it?

Crosswalk is currently in Open Beta and can be accessed by changing your App’s platform_version to v2.3.1beta-3 and using the crosswalk target rather than android:

  1. Open your application’s src/config.json file in an editor and set the platform_version key’s value as follows:
  2. Run your app from the Toolkit using the Crosswalk button or from the command-line with:

This is a Beta release so please don’t be dismayed if you run into issues. We’ll be standing by at to answer any questions you may have!

When will it enter the stable platform branch?

The Crosswalk target is the biggest change we’ve implemented since the release of Android 4.x and the Crosswalk project itself is still under heavy development so we’re going to be holding off a bit on pushing it to the stable v2.2.x release stream until everyone’s had a chance to try and break it!

To track release information you can go to: Forge Beta Release Notes

How much will it cost?

Integrating Crosswalk into Trigger.IO Forge has been a major engineering challenge which we’ve invested a lot of time, money and resources into.

Because of this, we did toy with the idea of making access to crosswalk exclusive to our Go Pro or higher customers as a way of recouping our development cost.

That said, our philosophy has always been to provide all of our customers with access to the best tools for Hybrid App development.

Our experience with Crosswalk is that it falls into that category of game-changing technology which everyone should have access to.

Therefore, we made the decision to make Crosswalk available, at no extra charge, to all of our customers irrespective of the plan they are on.

What about my custom modules?

We’ve been able to maintain binary compatibility with the vast majority of Forge modules and it is likely your modules will continue working “as is”.

If you have a custom module which is behaving strangely under Crosswalk please contact us at and we’ll be happy to assist you with any changes that may be required.

In conclusion

By using the Forge Crosswalk target to build your apps you are now able to take advantage of a unified HTML5 runtime with significant Javascript performance improvements.

Crosswalk also offers a more rapid release cycle than the Android System WebView and continues to bring new HTML5 features and performance improvements to your app even on devices which are no longer supported by vendor updates.

With Crosswalk you are now also able to take advantage of many new HTML5 standards and features which, in the past, had to be avoided due to compatibility concerns, including:

  • 3D CSS Transforms
  • WebGL
  • Date/Time input types
  • Multiple file selection
  • Raw Sockets
  • WebRTC

We feel that the integration of Crosswalk marks the beginning of a new era for our Forge standards-based Hybrid App Runtime and look forward to making the new capabilities developed by the W3C over the last few years available to all of our users.

Happy hacking everyone!

Another splendid year at

Hi Trigger Folk,

Following a turbulent start at the beginning of the year our team have had the time of their lives delivering ongoing product development and world class support for your favourite hybrid app platform.

And what a twelve months it has been… over the course of this year we:

  • Resolved 1787 support tickets.
  • Released 80 module updates including support for major SDK updates from Facebook, Parse, Flurry, UrbanAirship and Apptentive.
  • Released 17 platform updates including support for iOS 8 and Android 5.0
  • Released Forge Live and sped up everyone’s development workflow by an order of magnitude.
  • Released our entire catalogue of native modules as open source.
  • Began building a community of contributors to our open source modules. Thank you for everyone who sent in a push request or three, you know who you are!
  • Introduced you to the future of hybrid app UI’s with our introduction to with sloths
  • Worked with the great folk at PushWoosh to bring you another option for Push Notifications on Android and iOS.

I’d like to say a special thank you to so many of our customers who have been incredibly supportive, patient and just plain awesome during this time. Without you it would never have been possible to successfully transition to become independent, profitable and able to remain this way!

The team will be taking a well-deserved break starting from this Monday, 22 December till Wednesday, the 7th of January 2015. We will however be responding to any critical issues that may come up during this time.

This email wouldn’t be complete without a hint at some goodies Santa has in store for us in 2015 so I’ll just say: Thisthis, some of this and definitely more of this.

All my thanks to our customers, business partners, StackOverflow contributors, module authors, GitHub contributors, contractors and everyone (past and present!) at Trigger. We wish you all a peaceful holidays and a truly satisfying new year!

 – Antoine

Wrapping with for Silky Smooth UI Animations at 60fps

Unless you’ve been living under a rock for the last few months you’ve probably heard some buzz on the interwebs about a new Javascript UI Framework called that promises to change all the things!

Given the rate at which JS frameworks proliferate these days this could be a somewhat outrageous claim to make unless, in fact, it can and does.

I’ve been following their progress for some time now but rather than giving you my opinion I’ll ask you a few questions and then give you the opportunity to find out for yourself:

  • Would you like your app to feature silky smooth UI animations & transitions running @ 60fps without having to write native code?
  • Would you like the ability to create gorgeous custom user interface components that can compete with anything we’ve seen in native apps?
  • Would you like to easily populate a scrolling list view containing more than ten items without reducing your mobile device to a superheated vapour?
  • Would you like to do all of the above in Javascript without having to fight HTML & CSS every step of the way?
  • (Can you tell that I’m excited yet?)

If your answer to even one of these questions is “Hell Yeah!” then pop open a terminal window and do this:

And this:

Answer the prompts truthfully and then you can go:

Or, if you prefer:

Which should give you something like:


Want Moar?

Keep an eye on @triggercorp for our upcoming tutorial covering the ins and outs of using Forge native modules with

In the meantime, you can check out our sample app built with and here:


What’s a generator anyway?

When you did this:

…you invoked a generator that creates a seed app configured with all the bits and bobs needed to use within a app.

Generators are built using a really cool tool called Yeoman which also allows you to create seed projects for many other kinds of JS frameworks and projects.

You can check the famous-triggerio generator out here:

We’re currently working on creating generators for a number of other frameworks, let us know if there’s a framework you’d like to see us support!

What do hybrid apps have in common with sloths?

Hint: It’s not velocity

Ridiculously fast app development with Forge Live

Today we released v2.1.7 of the Forge platform and with it, a major new feature we’re really excited about: Forge Live


Forge Live shows you the effect of your code changes in real-time on an emulator or device making the the mobile build / deploy / test cycle faster than even the web (you don’t need to refresh!).

See it in action now:


It’s almost halfway through 2014 and there’s never been a better time to be hybrid app developer.

Frameworks such as AngularJS, Ionic and React promise an order of magnitude improvement in developer productivity. Coming up we can also look forward to technologies such as, WebGL and Apple’s Nitro JavaScript engine. Technologies which promise a level of performance that will open up a whole new class of applications that could previously only be tackled by native developers.

That said, one thing that doesn’t seem to be receiving much love is mobile app development workflow.

Frankly, writing apps for iOS or Android feels much like desktop app development did in the early 90’s!

  • Write some code
  • Compile your app
  • Deploy to device
  • Run app
  • Wash. Rinse. Repeat.

While this model is not without its advantages  it’s a shame to be bound by limitations inherent to compiled languages such as Java and Objective-C when hybrid app development is built on the same dynamic languages that make web development as easy as:

  • Write some code
  • Hit Refresh

In fact, using tools such as grunt-contrib-watch to detect changes to your files and then automatically trigger a browser reload the entire cycle can be reduced to a single step.

That’s why we created Forge Live!

Using Forge Live


If you are developing for Android please note that Forge Live is only supported on devices running Android KitKat (4.4.x) or higher.

Step 1: Install Node.js

Before you can start using Forge Live you will first need to install Node.js on your machine. You can either use your favourite package manager or download an installer from:


On newer versions of Ubuntu the packaged node binary has been renamed to nodejs due to a conflict with another package. Unfortunately this has the side-effect of breaking grunt.

The workaround is to install the nodejs-legacy package as follows:

Step 2: Check your Platform Version

If you’ve just created a new app you won’t need to do anything here.

If you want to use Forge Live with an existing app you will first need to check that your app is compiled with Forge 2.1.7 or higher.

You can do this from within the Trigger Toolkit:

Or in your src/config.json file:

Step 3: Start your app

Once your app is on the latest stable platform version you are ready to start your app up in Forge Live mode.

To do this, simply navigate to the Forge Actions page and click on the platform you’d like to work with:

Or with the command-line tools:

Followed by:

Any changes you make to your source code will now be instantly reflected in your running app. It’s that easy!

If you’d like more information about using Forge Live you can also look at the documentation.

We’d love to hear what you think about Forge Live so please feel free to contact us on with any questions or feedback.

A new year with

Ho Trigger Corps!

It’s now been just over two months since I took over the role of day-to-day management at and it’s time to take stock of what we’ve accomplished during this time.

This year I’ve had a chance to meet many of you for the first time and I’m pleased to report that we’ve been able to keep the promise we made to maintain the high bar sets for customer support. It’s been 75 days and during this time you’ve opened 420 support tickets and we, in turn, have resolved 425 support cases.

That said, despite all odds, life at hasn’t just been about support.

We’ve also managed to keep product development ticking over with 2 minor platform updates, 2 new native module releases and more than 10 updates to our existing native module line-up.

After open-sourcing our modules in December last year we’ve noticed a steadily increasing stream of pull requests coming in to the repositories that benefit our entire community.As a result it is largely due to your effort that we’ve been able to keep development moving.

Thank you Trigger Corps, You rock.



New Native Modules


Native Module Updates

  • forge.notification v2.3 New method: getBadgeNumber
  • forge.launchimage v2.4 New option: background-size: cover, iOS7 fixes
  • forge.file v2.3 New method:
  • forge.parse v2.3 New methods: setBadgeNumber, getBadgeNumber, registerForNotifications, Added support for Google Cloud Messaging, Update to latest Parse SDK
  • forge.flurry v2.4 New option: debug, Update to latest Flurry SDK
  • forge.facebook v2.3 Update to latest Facebook SDK
  • v2.2 Fix for mute switch position being ignored. (Thanks @mnaughton !)
  • forge.request v2.5 Fix HTTP method and contentType settings being ignored for file uploads
  • v2.3 Fix a crash when selecting all contacts on Android

Read the full changelog here


Core Platform Updates

We’ve pushed out two minor updates to the core platform which brings us up to v2.1.2.

These updates fixed a number of annoying bugs, made some progress towards a better code signing experience for our Windows customers, brought our Android native API support up to android-19 and added Chrome DevTools remote debugging support for Android KitKat devices.

If you’re not already on 2.1.2 do consider updating your projects today!

Read the full changelog here

New Hires

I’m really pleased to announce our new team member, Gaelin Meyer, who has been doing a great job managing Accounts and Marketing for us here at Trigger.

Gaelin has been working in an advisory capacity with several Internet Startups here in Cape Town since 2005, holds a degree in Psychology from the University of South Africa and is busy raising two beautiful children.

Her primary responsibility is to know what you need from us at, so expect to hear more from her in the upcoming months!


Team changes at is now nearly 3 years old and publicly available since January 2012. We’re delighted with how it has grown from a useful product — 10,000 developers signed up, and millions of app builds — to become a profitable business as of September this year.

We’re proud of the self-sustaining community that has built up around, with several people making their livelihood building apps for clients on our platform, and contributions from the community on StackOverflow. Last week we open-sourced our native modules and can already see developers begin to contribute.

At this point, James and I have made the decision to step back from day-to-day work on, and for the company to bring in new blood. Antoine van Gelder, whom many of you already know, will take over day-to-day management and new development.

We’re delighted to have Antoine take this on and anticipate only positive changes for current and new customers — there is no change to the availability of the service, support or plans for new development. James and I will continue to be involved as directors working with Antoine to improve the company and product, although you’ll see less of us in day-to-day communications.



We’ve worked with Antoine for almost as long as has been around. He was instrumental in getting the product to launch, working across our entire stack from the client SDKs, Toolkit, server-side logic and compile servers. In addition to his development work, he has also interacted with many customers through support and through the maintenance of our legacy browser add-on platform.

Not only that, but Antoine has broad experience running all aspects of software companies having built his own startup and run consulting companies in the past. He’s a keen advocate of standards, open source and creating great products that benefit with the world. All that — combined with his technical expertise, knowledge of the stack, and product philosophy — means we are supremely confident that he is the right person to take from here.

With our v2.1 release we felt the company and product are in great shape and could benefit from new inspiration, so the time is right to pull the trigger (pun intended). Anyone who has worked on the same product for several years will know that it can often benefit from a new perspective, so we’re confident that the plan we’ve put together with Antoine is a win all around.

The future

James and I delight in creating new products and businesses which is why we started We love what we’ve built with but over the past year have also become excited about other projects.

Specifically — based on many shared interests — James, Tim, Connor, and I have decided to join Square. We’re super excited about Square’s prospects, their team, and the new projects that we can impact there. Making commerce easy — Square’s mission — is inspiring and has many facets. We hope to iterate fast and be able to share more of what we’re working on soon.

We’re delighted at being able to make this team change while having the product continue to improve and grow sustainably. It’s a situation that few startups are able to reach, and we owe it all to our amazing customers: thank you!

Please do reach out by emailing with your questions. Forge 2.1: open-source modules and collaborative development

Today, we’re really excited to announce version 2.1 of’s mobile development platform: Forge.

This version is hugely significant for us, as it marks a shift towards a more open, collaborative development approach on the platform. As part of this, is open-sourcing the native modules that we’ve developed.

Since we launched our native modules feature at the end of July, we’ve been blown away by our users’ enthusiasm for adding new native functionality to your apps. Already, there have been over 1300 updates to the almost 400 modules our users have been working on, including modules for Urban Airship, Apptentive and keyboard customisation:

Three 3rd party modules

With version 2.1, we’re doubling down on native modules by making it super-simple to publish, consume and collaborate on native modules. Changes include:

  • ability to publish modules into the module index
  • introduce concept of module namespaces, which allows independent developers to collaborate on one codebase
  • dependency declaration and resolution between modules, so you can rely on a library being present without inlining it
  • add changelists and documentation for your module


We’ve wanted to open source our modules for a long time, but before now there hasn’t been much point because it would have been so hard for our users to take our code and set themselves up to develop on it.

With the changes described above, those restrictions are now gone, and we’re excited to announce that we are open-sourcing all of our own native modules under the 2-clause BSD license.

If you want to tweak the form or function of a native module (e.g. customising the gradient of our topbar module, or extending our file module to capture audio as well as video), you can clone our code from GitHub and make the changes you want in your own module.

You can keep that module un-published — meaning only you and your team-mates can use it — or you could publish the updated module so everyone benefits. Of course, we’ll always appreciate pull requests back to our repo so that we can update the original module when it makes sense!

Note that we’re not open-sourcing everything. Our cloud build service, toolchain, and the core of our platform will remain under wraps: this means you won’t be able to do local-only builds that don’t use our cloud build service. But we firmly believe that the huge usability, speed and collaboration benefits you get from using a cloud-based compiler easily make up for that.

How to get involved

If you’re an existing user, our v2.1 update is rolling out now and you’ll be prompted to update shortly. If you’re not at the stage of creating native modules just yet, note that v2.1 includes a small change to JavaScript behaviour documented in the release notes.

Otherwise, head across to our site to create an account, download our lightweight tooling, and get started on your first app.

The native module development process is covered in our documentation, and keep your eye on GitHub — we’ll be uploading and updating our module code there!

New feature roundup: iOS 7 and Android 4.4 support, new accelerometer and Apptentive modules

As well as a host of minor improvements and fixes which are documented in our module changelog, we’ve been hard at work on some big items.

Namely, support for new iOS and Android platform versions and two new modules: one developed by us, another by a 3rd party which we’re delighted to be able to offer.

Here are the details:

Apptentive: in-app feedback from your users

Apptentive provide SDKs for iOS and Android that enable you to get in-app feedback from your users.

Interacting with your users inside your app is immensely important to boost ratings – create conversations rather than critics and intercept negative feedback. So we’re really excited that they’ve created a module so you can use their SDK in your app:

This example shows how Apptentive can be used to solicit feedback. In addition, the Apptentive module provides:

  • App Store rating flow
  • Surveys
  • Custom user information tracking
  • Custom device data tracking

We’re excited for future updates by Apptentive and other SDK developers who integrate with

Accelerometer module

By popular demand, we’ve created an accelerometer module. Here’s an example of how it can be used to detect whether the phone is laying flat:

iOS 7 and Android 4.4 support

We released support for iOS 7 in September as part of our v2.0.1 platform version, with several new module versions. You can read details of the changes here.

More recently, we were very excited about Android 4.4 (KitKat) which was released last week. Especially since this new version of Android will have its WebView powered by Chromium. That’s great news for developers since it means that, in apps built for Android 4.4, you can use:

To start building for Android 4.4, you will need to:

  1. Rebuild with our latest platform version – 2.0.4
  2. Update tabs module to 2.5
  3. Update ui module to 2.1

What’s next?

Follow us on Twitter or Facebook to see our latest news and updates.

Sign up now to take advantage of these new features or get in touch with us at anytime at

Guest post – how CardSmith build for mobile using

This guest blog post was written by Scott McClure, developer at CardSmith, currently building the mobile version of their access, tracking, and verification application using He also manages the build and release processes for the company. 

About the app

PockeTracker is a mobile application that allows users to verify admittance to facilities and events, take payments on location, and track attendance. It supports an offline mode and the ability to scan ID cards with magstripe, barcode, or contactless card readers. Version 3.0 of the application was redesigned from the ground up using


We had several specific requirements for the tools needed to build the next generation of our mobile application.

  1. We wanted to write the majority of the code in a language everyone on the development team was familiar with.
  2. We needed to be able to target both iOS and Android platforms.
  3. If we were going to use a hybrid model approach we would still need direct access to the underlying OS. Direct access was particularly important because the application needed to integrate with several different magstripe, barcode, and contactless card scanners, and keyboard emulation would be insufficient to accomplish this. Also, we were going to need to support an offline mode with large amounts of data being cached locally. We felt for performance reasons that this should be implemented natively.
  4. Any tools required for building and packaging the application needed to integrate seamlessly into our current build process. fit all of these requirements and we have been very pleased with the final results.

I’d like to share with you how we accomplished the last requirement to integrate with our current build process.

Build Process

For all of our applications, we rely on Subversion for source control and TeamCity for build management. With PockeTracker, we now also use the standalone build API provided by Source code check-ins are automatically detected by TeamCity, then checked out by the build agent and sent via HTTP to the standalone build service.

We could have just as easily used the command line tools instead of the standalone API, but using the API gave us one less build agent dependency to maintain.

Currently, our build tasks include:

  1. Check out the latest source from the trunk
  2. Run our JavaScript modules through the Require.js optimizer
  3. Update the build number in the config.json file
  4. Zip the source folder
  5. Send the zip to the Trigger build service
  6. Monitor the build for a success or failure
  7. If successful, download the package and publish it

Checking out

This isn’t really a post about how to set up a TeamCity build configuration, but TeamCity makes it very easy to monitor a Subversion repository for changes and check out the updated source onto a build agent. It simply requires you point TeamCity to your repository, add your login information, and create a build process trigger.

Updating the build number

Once a change is detected and the source downloaded, we need to update the version number found in the config.json configuration file. Using a custom C# build runner plugin, and taking advantage of the JavaScriptSerializer class we are able to accomplish this fairly easily:

Zipping the source

Once the build number is updated, we use the NAnt runner to zip the contents of the ‘src’ folder:

Push, monitor, and download

Out last steps are to push the zip file to the Trigger standalone build service, monitor for a successful build, and download the package. Again, we used the C# build runner.

To push the zip file along with some configuration settings, we prepare a web request. Since we are currently only building for Android, we only need to include the required email, password, and ‘and_*’ fields.

Next we send the POST request and get the response:

If the response is OK, we periodically call the service with the ID that was supplied to monitor the build progress:

If the service returns a “SUCCESS” message, then we just grab the package URL and download it. Otherwise, we write the return message out to the console so the build log will show the reason for the failure:

The downloaded package is then published internally for acceptance testing.’

Closing thoughts

Of course, there are many improvements we would like to make to the process. Automating our QUnit tests for the main application and JUnit tests for each Android native module will be our next priority. Another priority will be to add in support for our iOS native modules once those are completed. I hope that this example of our experience with the standalone build API will help you with your own build processes.

Sign-up for now to get started with your mobile app. Questions? Email anytime.