Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Rough ETA/timeline/milestones? #26

Open
josephrocca opened this issue Dec 2, 2023 · 2 comments
Open

Rough ETA/timeline/milestones? #26

josephrocca opened this issue Dec 2, 2023 · 2 comments

Comments

@josephrocca
Copy link

josephrocca commented Dec 2, 2023

I'm wondering if there are any rough estimates for when this will land in Chrome stable? Even something super rough like "hopefully before the end of 2024, but maybe not" would be useful to know.

Also, would the initial stable release support the "safe exe/swf/etc for the web" use case - i.e. no webserver, just a file you can e.g. share via email and double-click to open?


(I'm extremely excited for this proposal, thank you so much for your work in pushing it forward. I think the world desperately needs a "safe, portable exe" and in this humble little repo you're actually making it happen. This is going to be a really big deal.)

@b1tr0t
Copy link

b1tr0t commented Dec 5, 2023

Hi Joseph, happy to hear you're interested! I'm curious about the use cases you had in mind for it, can you share more?

In 2024 we'll be doing a limited test program focused on ChromeOS. We're not committed to ship beyond the limited test program right now and whether it moves forward will depend on developer response, feedback from the test program, and how interested standards and other browsers are. 2025 is the earliest I'd expect the feature for GA beyond ChromeOS, assuming everything goes well with the test program.

PS I agree with you about the possibilities! We'll need to see how developers respond and how they feel about IWAs relative to the alternatives.

@josephrocca
Copy link
Author

josephrocca commented Dec 6, 2023

Thanks for the update!

I'm curious about the use cases you had in mind for it, can you share more?

I have quite a few use cases, but perhaps the most illustrative one is that I run a web-based coding/no-code type platform where people (without much/any coding experience) can make little apps/experiences/games, and I allow them to download their creations so they can run them offline, share them as a file with friends, host them on their own, and just generally be "independent" from the online platform. Currently, though, I can only give them a .html file with all resources bundled/inlined, which has a lot of limitations since it's not considered a secure context when they open it up in their browser, and inlining large resources (e.g. videos, images) as text/base64 is a hack that doesn't really scale. Also, IIRC some platforms are reluctant to open .html files due to security concerns (e.g. phishing). And there are a bunch of related/tangential issues - e.g. localStorage/IndexedDB persistence. Overall .html files just aren't viable as a "safe, portable executable".

see how developers respond and how they feel about IWAs relative to the alternatives.

I'm actually curious what the alternatives are here? We had a sort of "golden age" with .sfw files a ~decade ago brought on by gaming, but that file format was unfortunately a dead end due to fundamental limits of Flash, security issues, and the fact the it wasn't an open format. Things like Electron are a non-starter / non-solution here, because they're not safe/isolated at all. AFAIK a .html file is the closest we currently have, but unfortunately falls far short of what we need for a new "golden age" of local(-first) apps.

Also, there's a somewhat subtle point to be made here in regard to asking developers whether they want IWAs: The ambition of this proposal should, I think, be aimed at changing user behavior as an ~upstream cause of developer behavior change. Once users know that double-clicking a wbn is a safe as clicking a link (or even safer), this suddenly makes it much easier for developers to ship local apps, because users are willing to download and run them quite casually. Said another way: The "market" for local apps is currently tiny (in part) because users are (correctly) trained not to run stuff they downloaded from the internet.

The average developer arguably shouldn't be interested in IWAs because there's no extra value to them over something like Electron until user behavior adapts to the idea that wbns are safe it's fine to casually download and run them. Currently, users will be just as weary of wbns as they are of Electron exes. So I think there'll be an initial, slow chicken-and-egg type growth of IWAs as user behavior adapts to the idea that they can download and run local apps safely. The full potential of IWAs probably won't be visible for several years, and I think almost certainly won't be visible in developer surveys even a year or two into their widespread existence in browser engines.

(To be clear, I understand there is more utility to developers in IWAs than just "a file format that users aren't weary of running", since IIUC there are usage modes of IWAs that don't even involve a local wbn file that the user clicks on. This is just my perspective as someone who thinks that aspect - a safe, portable, offline, 'executable' - is a really big deal.)

Apologies for the wall of text! Thank you again for you and your team's work on this!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants