Iteration Plan for May 2020

[ahdbilal]

May 2020 Plan

Last month, the App Center team updated various frameworks across our services (Android NDK, Appium, Sparkle) to support the latest versions of each respective framework. The team also made structural changes with our distribution upload client, as a result of which we expect users to get improved and reliable upload speed performance. The improved service will be rolled out to all users this month.

This month, we have the following features planned:

API

• Introduce App-level API tokens [API token per app and per feature #167]

Distribution

• Roll out improved upload speed capability to all App Center portal users
• Improve deep linking to release experience [Link to Release on install.appcenter.ms #140]

SDK

• Package iOS SDK as SPM [App Center SDK as a Swift Package for Xcode 11 #1295]
• Add support for Mac Catalyst [Add support for mac Catalyst #1578]

[phatmann]

Could we please get #125 into this iteration? It is seriously hurting our workflow, and will not take long to implement. Thanks!

[ahdbilal]

Thanks @phatmann for the feedback. I have just pushed for it internally and will share an update soon.

[jasmsu]

Was "Insights into versions affected by same crash group" that was a part of the April 2020 plan dropped? Looks like it wasn't marked as done in April.

[imWildCat]

Anything which the community can help to support mac Catalyst?

[tdekoning]

@ahdbilal any update on ticket #125 being in this iteration plan?

[phatmann]

Could the dev team provide some insight into how issues are chosen for an iteration? Many top-rated issues are ignored month and month, while others are addressed much sooner. What is the criterion for selection? I for one have become increasingly frustrated with the lack of progress on fundamental aspects of App Center and wonder if this is just how it will be indefinitely.

[phatmann]

To provide a little more data, the majority of the most-commented issues are well over a year old.

[zoontek]

Could you take #1575 as well? Currently, the module is unusable with Error Boundaries and graceful crashes.

It would be super useful to catch unhandled Promise rejection (by overriding https://github.com/facebook/react-native/blob/0.63-stable/Libraries/Promise.js#L19) and custom non-fatal error too (by overriding ErrorUtils global handler)!

We are currently using Crashlytics for this, but would love to remove it.

Even without the deep integration, just adding a function that allow to send a custom error with a certain severity would be amazing and I guess it's not that much work 😕

[Mikilll94]

Any update on #125 (Google Play internal test track support)?

[zoontek]

This project is dead 💀

[phatmann]

@zoontek I certainly see why you say this. Month and month we see almost no progress on some fundamental issues. It does make me wonder if MS is wavering on their commitment to App Center.

In May there were 3 commits to master and in April there were 2 commits. There have been no commits in June. That sounds like a project in maintenance mode.

The only 2 posts to their blog in 2020 were to announce the retirement of MBaaS. This is not a good sign.

@ahdbilal Care to comment?

[ahdbilal]

Thanks all for raising these issues. I agree that customer-facing features (#125 especially) have suffered over the last few months. The team has been focused on improving the backend and stabilizing the service during this time. We need to do a better job of providing more clarity on how we are prioritizing our work. I have taken a note of it and shared this as well as your other concerns with the leadership. We will sync back with you soon to share more details on the road forward.

[joshfriend]

Many folks are frustrated because we have used this product for a long time now, by which I mean HockeyApp (which Microsoft acquired and transitioned to appcenter). There were features which HockeyApp already had which were removed when we were forced to transition to AppCenter as HockeyApp was completely shutdown (e.g. linking to specific builds which is still a huge pain point for everyone). Additionally, HockeyApp had very popular feature requests that we had been pressuring to add for years which were seemingly deprioritized by a LOT (caught exceptions for example) after the AppCenter transition in favor of features which we hadn't really been asking for as much.

I fully understand that you have backend work to do, as an outsider i have no idea what kind of issues you may have been left with as a result of the hockeyapp sunsetting! I just wanted to add my perspective (which I think I have seen reflected in other comments on various issues) that long time users of HockeyApp (RIP) are frustrated by what we were familiar with as a very good app distribution and crash reporting tool being turned into a much larger platform which actually does not do as good of a job at performing the duties that HockeyApp used to.

I follow a lot of the issues and feature requests here and lately it has become difficult to tell if there will be meaningful changes made anytime in the near-ish future regarding the features which my team considers important.

I know yall do try pretty hard to respond to issues, and appcenter is still the best distribution platform for test builds that I have found. ❤️

[yvbeek]

@ahdbilal Thank you, I see you've updated the Roadmap document:
https://github.com/Microsoft/appcenter/wiki/Roadmap

What would be the best way for us to keep track of the updates that are being done to the App Center platform?

I think that many of us are worried that App Center is moved to the back burner and that eventually Microsoft will shutdown the platform. I'm not sure if you can share any further details, like how many people are actively working on it. The platform has a few quirks and might have some overlap with GitHub actions, but it would be great if we could continue to use it over the years to come.

[danielwilson1702]

It is not sounding great tbh. Tech debt doesn't really wash with me this project is not that old and crying out for more features, so why put everyone on solving tech debt? Also mid-2021 is oddly specific and an enormous time frame to solve all this tech debt :/