PWA + Project Fugu

See meeting agenda


slides - minutes

Video hosted by Web Castor on their StreamFizz platform.


Kenneth Christiansen: My name is Kenneth Christiansen, I work at Intel.

I'm also a member of the TAG, but here I'm wearing my Intel hat and introducing the work we're doing on Progressive Web Apps and Project Fugu.

So, you've probably all heard about Progressive Web Apps.

So Progressive Web Apps are these apps that are installable by using metadata from the Web App Manifest. These can also work offline, or you can create custom caching strategies using Service Workers. The idea is that these can be integrated with the host OS to such an extent that they're indistinguishable from native apps.

So for instance, here, I'm showing a Chromebook, I'm going to a website for Pinterest, and as you'll see in the address bar, I get this nice Install icon. I click on it, and do actually allow me to install this application. So just click on that and it automatically turns into this very native-like experience.

But we also have gaps.

There are a lot of APIs that are not available for these developers that are available on native.

So for instance, here I have a picture of an app that is written, is called AutoCAD by Autodesk.

It's partly using WebAssembly for the view, but, for instance, if people sent me these AutoCAD files, I'm not able to actually interact with them directly because I don't have access to my file system.

we really want to fix all those gaps. So this is, basically, Project Fugu.

Project Fugu is a project that was started by the Chromium team, so by Google, but with collaboration today by Intel and Microsoft, two companies also working on Chromium.

Fugu is the Japanese word for a pufferfish, so this is something people eat, but it's kind of dangerous. If it's not prepared correctly, you don't have the right education, you might actually end up killing people and we wouldn't want to do that.

So in the way of this, we're really trying to bridge the gap between native and the web, but want to do it in a way that would keep the core tenets of the web, such as user security, privacy, and trust. But it also takes a long time to get all these APIs to the web, so you might be wondering what is the process for doing so.

Well, the idea is that we start by identifying need and use cases, not only by talking to web developers, but also to partners. So both Microsoft and Google and Intel would be talking to partners and figuring out, like, what is needed on the web.

Then we write an explainer, following the guidelines from the TAG. Here we have a link to the explainer document, also known as the explainer explainer.

Then we will start soliciting feedback, writing explainers, try to get this idea going, interacting with both partners and the community.

At some point, we'll end up with some kind of spec, which will then be implemented in parallel inside of Chromium, and hopefully other browsers as well, going forward.

Then, when we think it's ready to get actual feedback, it will turn into what is called an Origin Trial. It means that specific origins, sites, can opt in to using this API for a short period of time, by getting a token that they'll put on their website.

And the feedback we gather from that will then affect the spec, and we'll end up, hopefully, with a good spec that at some point will be shipped.

So this is a very short introduction, I only had five minutes.

But I have two links here that might be of interest.

With that, thank you for listening.



Platinum sponsors

Rakuten Institute of Technology, Coil
														    Technologies, NTT

Gold sponsor


Silver sponsor

Yahoo! Japan WebCastor

Gift sponsor


Bronze sponsors

Newphoria, JPRS, Kodansha, Hitachi, Shueisha, Media Do, Sony, Igalia

Friday Coffee sponsor


Network sponsors

NTT West, Cisco, NTT Communications

Support TPAC 2019 and get great benefits from our Sponsorship packages.

For further details, contact