boinc
boinc copied to clipboard
BOINC
Android release management
What's the current state of the Android release management? With Rom's departure it seems as if there simply are no further Android releases at the moment, right? If we want to retain Android as a compute and outreach platform we have to fix this. Here are a few questions that I think need to be answered in order to establish a new working process:
- Is there anyone currently responsible for Android releases?
- Who has access to the Berkeley SSL Play Store account and thus could release a new version?
- Should we continue to use that account or create a new community account? Are we entitled to do the latter under the BOINC brand?
- Who can/should take over control over that account
- Is there any documentation on the old release workflow/procedure for someone willing to step in?
- Which client version should we release?
- Which Android app issues are release critical?
- Should we also target F-Droid as a release channel for all free software enthusiasts rejecting the Play Store?
Reserved
(please compile a list of all release critical issues in this comment)
- #1473
- #1597
- #1659
Are you sure this is just a task? I added defect on purpose, it's broken right now!
@davidpanderson / @romw If you guys want (need!) to let the community take over the Android development we need your input and temporary help to get things started. We need at least access to the Play Store account if BOINC itself can't spend time on that anymore.
Please note that this matter has some urgency to it as failing to provide timely fixes to the seemingly abandoned Play Store version will lead to increasing negative reviews that will eventually hurt the whole project.
I don't have the playstore account into. @romw, can you communicate it to me (e.g. by text or phone msg), and I'll forward to Oliver? -- D
On 4/12/2017 2:03 AM, Oliver Bock wrote:
@davidpanderson https://github.com/davidpanderson / @romw https://github.com/romw If you guys want (need!) to let the community take over the Android development we need your input and temporary help to get things started. We need at least access to the Play Store account if BOINC itself can't spend time on that anymore.
Please note that this matter has some urgency to it as failing to provide timely fixes to the seemingly abandoned Play Store version will lead to increasing negative reviews that will eventually hurt the whole project.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/BOINC/boinc/issues/1846#issuecomment-293517723, or mute the thread https://github.com/notifications/unsubscribe-auth/AA8KgYCziRkv36ZUWqLRMsa6U-zO20FNks5rvJN_gaJpZM4MymFj.
Thanks. That would be a first step.
The next step needed would be to discuss the questions in the original post. Some of them need of be answered by BOINC/Berkeley officials I think.
One way of going about this is to transfer app ownership from "Space Sciences Laboratory, U.C. Berkeley" to a new account. Given that BOINC is meant to become a community account we could think about creating a "BOINC Project" or "BOINC Community" account or similar, with access given to a group of trustworthy members.
However, all of this requires that we get access to the SSL account first.
Thoughts?
Related to the above: we should think about registering a domain for the BOINC community project. Apart from moving the official website away from Berkeley (is this intended?) we would be able to define reverse domain IDs and certain email addresses, e.g. one that could be used to register a Google developer account.
Guys, without at least some basic support (just a kickstart) from the official BOINC team the community won't be able to improve the situation. That means, the Android app will die sooner than later. Is this already an accepted part of BOINC's roadmap? If not, please act and enable us to help.
I'll see David tomorrow and talk to him about it.
On Tue, May 2, 2017 at 7:31 AM, Oliver Bock [email protected] wrote:
Guys, without some support from the official BOINC team the community won't be able to improve the situation. That means, the Android app will die sooner than later. Is this already an accepted part of BOINC's roadmap? If not, please act and enable us to help.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/BOINC/boinc/issues/1846#issuecomment-298652951, or mute the thread https://github.com/notifications/unsubscribe-auth/AKXcsmbuy-aGGWbiUzJqJY-WUpsaNwHyks5r1z5EgaJpZM4MymFj .
-- Eric Korpela [email protected] AST:7731^29u18e3
I agree with @brevilo Dreamlab just came out with an easy to use app for people to donate spare cpu cycles. Folding@home is reworking their android app. If app is left as is only the most dedicated will use it until it eventually breaks. No hope of drawing more people into the fold with the app in the state it is.
I checked the DreamLab and wow. Super simple, easy to use, and easily understandable, looks wonderful and modern. All the boinc clients should work like this.
One thing I'm concerned about. The team developed it from scratch rather than use the boinc as a base. It would be more beneficial for the community if the brainpower worked on one solution, rather than reinvent the wheel all the time.
It's a PR project of Vodafone; they're probably not aware of BOINC. Anyway - let's redesign the BOINC Android app to be all those things.
I took a quick look at the listings for DreamLab on the Google Play Stop and Apple Store. My impression is that the UI is simple because the capabilities are very limited. There are only 2 "projects" to choose from; I suspect both are included within the one DreamLab app and there is no automatic way to download more projects or even more worker apps, so DreamLab is not easily expandable.
BOINC supports a larger number of projects, and each BOINC project can (and often does) include multiple worker applications. World Community Grid is an especially good example of a single project with many applications (sub-projects.)
Guys, this discussion is totally off-topic. If you want to discuss app design improvements please open new issue for that purpose.
Thank you.
Hi @TheAspens, any progress on this front? We can fix and improve the Android client/app as much as we want, but that's all meaningless if there's no actual release.
Any update on this? One of the *newer* features that allow for entering any account manager on Android is not on the current play store version. This would allow people to use large managers used for earning Gridcoin like grcpool.com and gridcoinpool.ru that are not on the list. I've seen people decide not to use the app for this reason.
@RoboticMind, new release for Android will be released simultaneously with the Desktop version for Windows, Mac and linux.
I have no such information, @davidpanderson or @TheAspens could know more
Does anyone have build (.apk file) of the current client? If so please send it to me.
@davidpanderson, is there a release branch for 7.16.0? I then could build apk for you.
Almost. @TheAspens wants to document the release management process for Android which should answer the process-related questions I asked. The issue can be closed as soon as that's done.
Thanks
I'm closing this ticket. Main points:
- I'm responsible for Android Release.
- Both me and David have signing keys and access to the Google PlayStore (currently BOINC is not published there but that's the different issue)
The whole release process usually follows all steps from this document: https://github.com/BOINC/boinc-policy/blob/master/Development_Documents/Client_Release_Process.md
All other questions from the original post and the whole thread are outdated.