That is the worst approach I can imagine. The problem has been known since July! How much longer should we wait? ![]()
Apple already fixed the issue, we are stuck waiting for them to release the fix
unfortunately they are being extremely slow releasing a critical OS fix, I wish there was some way to get this out there sooner, but this is sadly the situation
Thank you. That gives me hope.
Undoubtably Apple will come out with an update tomorrow that fixes all of this just to prove me wrong.
But, I wonder at what point it might make sense to consider perhaps simplifying some of the design of Discourse so these kinds of bugs have less of an impact.
To me, this feels a lot like the ever-present scrolling bug that just bit me again while browsing a thread on a Discourse-powered site. (The one where, if youāre scrolling on iOS and hold your finger down at just the wrong point, Discourse will keep repeatedly loading older posts throwing you potentially many dozens of posts away from where you were a moment ago.)
Yes, the design is extremely clever, and it looks awesome if itās working right. But it also feels rather brittle and breaks down quickly if the browser doesnāt handle everything just right. Or even if your internet connection hiccups at the wrong time, perhaps making a post you were reading moments ago suddenly inaccessible when you scroll back.
Anyway, perhaps not a thought that applies specifically to iOS 26 issues, as it appears Apple hopefully has a fix ready for imminent release. But more of a philosophical question to consider in future development.
As mentioned earlier, this isnāt really a case of us doing anything special or complicated⦠and it has nothing to do with infinite scrolling. Apple introduced a bug that breaks a basic browser height calculation when the keyboard is open. It impacts many sites, including some of their own.
No improvement with iPadOS 26.1 Release Candidate.
I completely agree with @awesomerobot.
The scrolling issue is purely on Apple and WebKit (which is at the core of ANY browser on iOS). It is quite embarrassing to have released Safari so broken, at that level, affecting hundreds of millions of devices worldwide and thereās not even one workaround any of us could suggest would fix it. To my understanding, they are miscalculating the viewport on many different events such as after a keyboard-up event.
With the RC released yesterday, we can probably expect a public release on Monday the 3rd.
No, there isnāt any fix, but there is one trick to survive in the Discourse world ![]()
Thatās how I can write and fix my typos when using iPadOS 26 and DiscourseHub, and Safari works same way:
thatās insane haha. great debugging!
Unfortunately there are lots of bugs, not just 1 bug at play here. I think it is probably worth probably splitting out the iPad OS bug with the composer, this is certainly something we could work around, cause the situation is temporary and triggered by Scroll / Focus, which are 2 things we can work around.
We probably want to split this one off.
Our App seems to be fixed in 26.1 RC, so letās hope we are all done with this mess.
Iāve installed the iOS 26.1 Release Candidate on an iPhone 16 today and the problem is still occurring.
Which problem specifically?
@j.jaffeux has been working on a workaround for āadd inviteā¦ā and ācreate personal chatā¦ā scroll to the wrong spot. We are going to double check if it also resolves the iPad composer positioning.
Can you try again on meta? Is the issue still present for you on iOS 26.1 ?
We deployed one more positioning fix here
Iāll tell you when 26.1 is published. Iām on 26.0.1.
But one improvement has happened in deed. When I clicked reply, the layout and position of the cursor were off, as usual with 26. But when I closed virtual keyboard and re-opened it, everything went normal, and stayed that way ā until I changed app and came back, then the screen was off again, but portrait/landscape did the trick.
So, things are better now.
(Does CDCK send an invoice to Apple after this issue is fixed? I think you guys shouldā¦)
What model device are you on? We had to put a āsleepā in our code to work around the issue, the slower the device the less likely we slept enough.
Me? Quite basics, iPad is ordinary 10th gen and iPhone is 15+.
Rumors are gossiping the 26.1 update will come in one weeks time, so we will see.
Seems better on 26.1, but I am running into other problems with embed-heavy posts causing Safari to crash. Is this a known problem?
Safari the new IE 6 ![]()
I have not heard of this, some repro would help tons here (it can just be raw text, we can paste it in somewhere and determine what is causing the issue.)

