This with @mrjuniork
's leave (thanks):
John was not part of that group. Quit spreading lies.
Not sure who you're addressing here or accusing, but who's
spreading lies? And who ever suggested John was part of any 'Alpha' group?!... No one's lying that I've seen...
And to go even farther.
Canary= bleeding edge
Beta= almost stable
Official/alpha(term not app=stable
Hence husky using Delta name because Delta would be before Canary. Backwards alphabet like most projects. Idk where you think Alpha would before Canary. Read the GitHub she was using. She was pulling from the main branch and basically re-forking it with a commit or 2 changed from Canary. More than likely small things like package name and update channel. Literally above in John's own post from over a year ago he denounced them using that name. He didn't say anything about him being in it..if he did start it, why use that name and then call them out for using that name? I mean that sounds off. It's a good story though. She is a contributor, or was or whatever she is now. She flipped on being open source with Alpha. That is fact. Shared in a Chinese only TG. I wouldn't trust it. "Confusing."
... Seems you may be responding to @zgfg
's post here, but let me clarify for everyone:
Alpha has been around for a long while, and was originally vvb2060's (Nangong Xueshan's) personal fork. It was not a group effort, and still probably isn't, although vv may well oblige other LSP Devs like @canyie
with experimental support for new Shamiko functions etc; after all theses Devs all contribute to that now.
'Alpha' experimental builds go back to 2018 with 'Lite' and 'MTK' builds going back at least to May 2020.
These were clearly test builds for fixes later to be merged into official Magisk with feedback/testing provided by Devs native Chinese Telegram community, and Magisk would be completely different if it hadn't been for them. However, until recent times, Alpha and other forks have gone relatively unnoticed despite their impact.
Alpha was always basically the test fork for vv's fixes, and her changes have been at the forefront of Magisk development for some years now... So many major innovations in official Magisk have been developed in vv's Alpha fork and in her lesser known Lite, MTK and other branches...
Just to mention a sampling I posted more than a year ago: MTK compatibility, single package delivery, 64bit support, sepolicy fixes, several isolated process fixes, fixes for UID, MicroG, process names, file based encryption, mount rules for Sammy, auto-close issues bot, reboot menu, multi-user and shared user id compatibility fixes, hidden apps, Chinese doc translations, offline restore app, unsupported environment checking, fix for apex path (a big one), permission fixes, module Installation fixes, Kotlin code cleanup, MagiskHide stopping fix, Renamable App / single package Magisk archive (developed in cooperation with John).... And countless more recent commits for various devices/compatibility, new functionality and major refactoring of magiskinit injection, sepolicy rules and so on...
mentioned, users here have been testing / discussing @vvb2060's innovations for a good while; Existing Magisk is already very much @vvb2060's, and she is now not just a contributor but an official Magisk Dev...
Far from 'pulling/changing a commit or two, likely small things like package name and update channel', Alpha has added/tested many major innovations and compatibility fixes before they were ever merged in Canary... Alpha is a prime reason John has made comments like:
I had already posted about vv's Alpha, and Lite (test branch for whitelist MagiskHide hidelist for bypassing isolated process leak detections banks had started using) Magisk forks but switched to Alpha for daily use in May 2021 simply because (with his new Android Security job) John was no longer updating official Magisk and this was a way to test fixes in the melting pot without any Canary builds available...
After posting about my Alpha use experience (I said there is "presently no better substitute for Canary builds with regular updates that I'm aware of"), several members here like @zgfg
also moved to Alpha in or after August 2021. This was because John had not produced Magisk updates for many months pending the result of Google's internal review of his open-source project as a new 'big tec' employee (see https://topjohnwu.medium.com/state-of-magisk-2021-fe29fdaee458
), but was now again making commits/fixes in Magisk channels as he, along with vvb2060 and other Devs, was developing new Zygisk Magisk hooking functionality... This was a fairly long process including replacing much MagiskHide/HideList code with the different if simpler DenyList, and still no Canary builds were forthcoming until October 24...
Alpha allowed many of us to test John's official commits (vv was largely focused on testing and contributing to John's changes at this time also) and early Zygisk / Denylist during those 3 months without the need to build personal Magisk 'snapshots' ourselves... (Of course there would also have been official CI builds available, but Alpha had the advantage of filtering out most broken test builds.) This is the period @zgfg
was referring to in his post; effectively, as you put it, 'Alpha was before Canary' at least in publishing test builds publicly for some 3 months...
In October, with Zygisk in 23010, most of us moved back to official Canary for daily use/testing.
Lastly, John has never 'denounced' Alpha (or vv) for using the Alpha name; he simply stated:
... It seems that vv was quite agreeable too, and that Alpha discussion group you mentioned even hosted a contest to choose a new name... In the end it seems changing the package name from Magisk to Alpha was mutually acceptable however (so it's just Alpha, not Alpha Magisk)...
Hope this little (recent) history lesson helps.
Edit: Please be careful if responding to this.