5 ms·

Everyone suspected it for a while. Their active Twitter account refused for months to say when the app would be updated for the iPhone X. Personally, I suspect

Everyone suspected it for a while. Their active Twitter account refused for months to say when the app would be updated for the iPhone X. Personally, I suspect that they painted themselves into a bit of a corner with the tech stack they choose for Inbox. They used a unique C++ to JS compiler to run Inbox in the browser. This worked decently well in Chrome, but the experience in other browsers has been a lot choppier. It's possible the same codebase was also compiled to iOS and this is what caused the very long delay in updating for the iPhone X. I'm sad to see it go. The UX in the official Gmail app isn't quite as good. The bundle workflow they developed for working through your inbox is something I haven't seen elsewhere.

fra 8y ago
When you have a successful product people love, you don't sunset it because the tech stack is too cumbersome. This has to mean that the product wasn't seeing the adoption they were hoping for.
- nathancahill 8y ago
  You might if you find yourself developing duplicate features in parallel in two different codebases.
- oihoaihsfoiahsf 8y ago
  I'm sure it's a little bit of both, right? If 99% of Gmail users had voluntarily migrated to Inbox, I'm quite certain it would not be getting shut down. That doesn't preclude that there is a significant contingent out there that loves Inbox. (BTW the same goes for the old HN hobby-horse, Reader. If it had had a billion users, it would never have gotten shut down.)
- munificent 8y ago
  Love don't pay the bills. If the product isn't helping the bottom line, companies will sunset it. "Free beer" would be the most widely adopted offering a bar could have, but it's not going to be good business for the bar.
  - haggy 8y ago
    > love don't pay the bills Thats not always true especially with Google. Anything that helps tighten the already tight grip they have on users directly affects the bottom line.
  - nickwalton00 8y ago
    That's sad. I would pay to use inbox.
- pound 8y ago
  Well, you're talking about product strategy in context of google..
pennaMan 8y ago
>They used a unique C++ to JS compiler to run Inbox in the browser I would love to read some reasoning behind that. I'm really trying to come up with pro arguments but can't.
cromwellian 8y ago
There was no C++ to JS compiler used, it was Java compiled to Closure Compiler (J2CL) and Java compiles to Objective-C (j2objc) However only business logic is transpiled, the UI layer on every platform is written by hand in the native impedance matched language of the platform.
- nathancahill 8y ago
  Ah, that's right, thanks for correcting me.
ex3ndr 8y ago
J2CL is much much faster than average react js code. Even testing converted port of bouncy castle's RSA and AES was faster than google's crypto.js

fra 8y ago

When you have a successful product people love, you don't sunset it because the tech stack is too cumbersome. This has to mean that the product wasn't seeing the adoption they were hoping for.

nathancahill 8y ago
You might if you find yourself developing duplicate features in parallel in two different codebases.
oihoaihsfoiahsf 8y ago
I'm sure it's a little bit of both, right? If 99% of Gmail users had voluntarily migrated to Inbox, I'm quite certain it would not be getting shut down. That doesn't preclude that there is a significant contingent out there that loves Inbox. (BTW the same goes for the old HN hobby-horse, Reader. If it had had a billion users, it would never have gotten shut down.)
munificent 8y ago
Love don't pay the bills. If the product isn't helping the bottom line, companies will sunset it. "Free beer" would be the most widely adopted offering a bar could have, but it's not going to be good business for the bar.
- haggy 8y ago
  > love don't pay the bills Thats not always true especially with Google. Anything that helps tighten the already tight grip they have on users directly affects the bottom line.
- nickwalton00 8y ago
  That's sad. I would pay to use inbox.
pound 8y ago
Well, you're talking about product strategy in context of google..

pennaMan 8y ago

>They used a unique C++ to JS compiler to run Inbox in the browser I would love to read some reasoning behind that. I'm really trying to come up with pro arguments but can't.

cromwellian 8y ago

There was no C++ to JS compiler used, it was Java compiled to Closure Compiler (J2CL) and Java compiles to Objective-C (j2objc) However only business logic is transpiled, the UI layer on every platform is written by hand in the native impedance matched language of the platform.

nathancahill 8y ago
Ah, that's right, thanks for correcting me.

ex3ndr 8y ago

J2CL is much much faster than average react js code. Even testing converted port of bouncy castle's RSA and AES was faster than google's crypto.js