I can see your pain points and everyone else’s too, @klausagnoletti. We have explored developing Flatpak some time ago, but ran into multiple hurdles that made it extremely complex to support it (see here).
While we have shelved this since then, we will still heavily consider this in the future given the huge support that our Linux community is ready to give. I appreciate users like you who don’t hesitate to be as direct as you need to be. Thank you for using your voice!
Speaking of frankness, your reply was more or less a copy/paste of your previous comments so it really didn’t bring anything new to the table sinceI just more or less read the entire 7 years of threads on the subject. Instead I would find it way more valuable if you commented on the things I wrote that you haven’t (as far as I know) commented on before. Especially the part on involving the community and trying out something new now that it seems like whatever you’ve done so far hasn’t done much of a difference. At least not externally.
Thank you for sharing your thoughts further! I have forwarded your suggestions to our Linux Team so we can assess alternative ways to approach Flatpak support in the future.
I’d really like to know what else insync is working on that is more important than flatpak at this point. flatpak is or has become THE way to distribute apps on linux especially with the rise of immutable distros. It also simplifies the install to a smaller subset of possible issues and test scenarios. Guys, If you can’t figure it out… hire a contractor to do it for you.
It seems to work fine for me. I think it’s usable, but it could be better, but only with involvement of the insync team, due to the proprietary nature.
Tray icon is there and works, but it’s not the correct icon - you get the fallback icon instead (three dots)
Autostart configured from inside insync does not work. It does not set the correct things. (there’s a workaround in the readme)
To fix 1, the application name needs to be correct, we ship it with the reverse dns name of com.insynchq.Insync which the app also needs to use internal, when talking to the system/notification endpoints. So we either need a build with a replaced app name or some way to affect that.
Woaahhh…just entered this forum and was asking @mia if I should open a thread about the flatpak so more people would be able to see it and ask for it, but what a surprise I’ve just when found this one that it’s almost 10 YEARS since it was opened!
Well, I’m one more asking for it cause I think it’s one of the feature request with most comments and people asking for it and the answer is the same I received by chat, that it’s been difficult, that the message will be sent to devs and that they will try again in the future
No wonder why this topic is full of disappointed users that feel ignored but I’m pretty sure we love Insync in general and want to help, but the problem could be not to define an ETA (commit to it), cause it’s easier to leave it for later, that’s the principle of procrastination.
Anyway, thanks anyway @mia for being polite, but I think it’s time to take actions to do something different (if not, that’s the definition of craziness according to Einstein) and I say all this in the best way possible to inspire change cause I really like Insync.
PS: Thanks to all the others that have commented and tried to make this happen before, I’ve read all the posts.
Hello! We have not considered revisiting Flatpak at this time, but rest assured that we will inform our Linux community as soon as this is back in development.
Mmmm…after 9 years…sorry to say it like this, I’m trying to be as polite as possible, but it would be better to accept that you are not going to do it don’t you think? it’s a shame that the top voted feature request are just ignored from 9 to 15 years and going…
This request has 10 years, if they haven’t done it yet, they won’t do it. Some companies simply don’t care about Linux and customers, we just have to accept that. I don’t use this software anymore due to weak Linux support.
Comercial software rejecting Flatpak in 2026 is a joke with Linux users. Rclone is free, open source and is getting the job done.
The weird part is that distributing using flatpak would definitely cause more people to consider insync. And afaik it’s a easy (low cost) thing to do(?). So all gain no pain, let’s be rational
Yeap, began to do my own research and precisely found about Rclone, and I was afraid of it since I’m not a programer or advanced linux user, but saw that now there are some options for UI…but also, since I’m using bazzite (inmutable distro), not sure if there will be issues… anyway, at least there’s light at the end of the tunnel, thanks and yes, agree with you and @birger.sp sadly.
PS: sorry @mia since you do your job pretty well, feel that you have your hands tied, but maybe could change developers mind at some point. Good luck