On Fri, May 6, 2022 at 2:38 AM Marcel Taeumel ***@***.***>
wrote:

> The issue is to *not break older images*.
>
> We do not do that. They just open fine. They run just fine. You can do
> 99.9999% of what you could do before. Please argue at the level of Morphic
> window dragging. Otherwise we cannot make any progress on this matter. You
> are way too generic for this matter here.
>

Ah, that's a good point. SO if it only affects interactive behaviour in an
image which has an IDE I could live with the "you have to execute code to
fix things" position. I'll not live happily, but I'll live :-)

> ------------------------------
>
> On a different matter, the way somebody introduced OS event pumping during
> regular method invocation is a much more serious deal breaker. (I think
> that only the image should do that.) It was pure luck that that has worked
> until recently. Just take a look at the fix in DisplayScreen >> #restore
> and then tell me again that we have to go back to the seriously slow
> --metal backend from before just to make the calls to ioProcessEvents in
> checkForEventsMayContextSwitch work.
> ------------------------------
>
> You should but you cannot always change the VM in isolation. It will work
> 99.999% of the time. But not always.
>
> —
> Reply to this email directly, view it on GitHub
> <https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/627#issuecomment-1119435768>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ADY5VUCOOSAYYCNHQO57DYTVITSC3ANCNFSM5URRNFFQ>
> .
> You are receiving this because you commented.Message ID:
> ***@***.***>
>

--
_,,,^..^,,,_
best, Eliot

—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you commented.