257 pointsby theanonymousone9 hours ago9 comments
  • cf100clunk10 minutes ago
    MicroG is possible to run on GrapheneOS but isn't recommended since GOS has a sandboxing solution for running Google Play apps:

    https://grapheneos.org/features#sandboxed-google-play

  • maelito3 hours ago
    I have never understood why people using MicroG can use apps like SNCF-Connect, while I cannot.

    SNCF-connect is the largest online marchant in France. Since june, I cannot click the "pay" button anymore. It coincides with the new feature of Pay with Google.

    So : is it because MicroG allows calls to Google services on e/OS devices, while my LineageOS refuses ? Or because MicroG implemented a try catch around this call, without any call to Google's servers ?

    • Maken3 hours ago
      I think that's because /e/os also distributes App Lounge [1], which can connect to Google Store and handle payments (most of the time).

      [1] https://doc.e.foundation/os/apps/app-lounge/

    • ForHackernews3 hours ago
      Some Play Services calls are stubbed out as no-ops and others are reimplemented either with fully free replacements (e.g. navigation) or else by calling proprietary Google remote endpoints (e.g. firebase)
  • orphea5 hours ago
    I guess this is submitted because of https://news.ycombinator.com/item?id=50154624 ?
  • jamesnorden4 hours ago
    Why do the most recent commits all say "Merge commit from fork" without any description of what the changes do or what the fork is?
    • forgotpwd164 hours ago
      Lazy merge commits. The descriptions of the changes are in the commits from the merged branches.
      • Pxtlan hour ago
        This is why I always think emperor git wears no clothes. Hiding these kind of intermediate junk commits from history should not require rewriting and breaking the distributed history you all share. imho the correct solution would be a tag-like layer for annotating existing commits with metadata (additional signatures, hide-flags, retroactive comments, etc) or just a full 2nd dag, but whatever: git sucks at annotating history and annotating history is a core workflow in git.
  • mshish3 hours ago
    It's too bad you have to rely on signature spoofing to be able to use this as an alternative (but obviously I understand why). The spoofing looks narrowly scoped on LineageOS, but in general I imagine it creates some risk.
  • xx__yy5 hours ago
    I use it on LineageOS, works a charm
  • ramon1563 hours ago
    I dont get why some apps depend on Google Play functionality. Why can't I download Telegram on /e/OS
    • microtonal3 hours ago
      This about Play Services and not the Play Store. Play Services are a bunch of extra APIs/services that Google does not make available as part of AOSP. The non-nefarious reason is to make it possible to fix bugs and roll out features faster because the OEMs are terrible at rolling out new Android versions. The possible nefarious reason is to make more of Android closed-source.

      At any rate, microG is a reimplementation of Play Services.

      You can directly download the Telegram APK from: https://telegram.org/android

      (Or using the app store that is part of /e/OS and serves as a front for both the Play Store and F-Droid.)

      • realusername3 hours ago
        They could also have it decoupled from the OS with an open-source version, these are two different subjects.

        The real reason to me is that they want to avoid Android forks outside of their control.

        • microtonal3 hours ago
          Yes, which is why I also mentioned the more nefarious reading of it. Making a large API surface use closed-source code gives them control.
    • raffael_de3 hours ago
      you can't download telegram from telegram.org?
    • cowlevel3 hours ago
      Telegram is a Russian spy app, it needs remote attestation you are not a bot because collecting bot data is not valuable.
  • philipallstar6 hours ago
    Is it too late to rename to "MindTheGapp"?
  • p0w3n3d4 hours ago
    Go home! Goober!...

    I mean Google!