Re: [PATCH] upcoming new feature: smooth whole-window sideways scrolling

On Wed, Jan 14, 2026 at 04:34:04PM +0100, Benno Schulenberg wrote:
> 
> Op 13-01-2026 om 07:03 schreef Chris Allegretta:
> > My feedback is simply: this is fantastic.
> 
> Nice to hear.  :)  But... do you wish it to become the default behavior?

Since it's such a big change, I worry if we don't make it flexible in 
some ways it's going to cause people pain to adapt - I actually don't 
care at all about complaints, just if we're causing pain unnecessarily.

One thing which I always tried to err on the side of back in the day, 
was make everything maximally configurable, and it's apparent that just 
does not work here, when there are so many modes affecting text display 
that we have to come up with all these terms to describe what is 
happening (hard wrap, soft wrap, joint scroll, probably more I'm 
forgetting). For the vast majority of users I assume they are either (1) 
not going to edit documents with many super long lines or (2) will 
adjust the screen size of the terminal so that they don't have to deal 
with it. Both of which are perfectly reasonable solutions of course.

Now for the remaining subset of people, I think there are firstly, the 
type like me who frequently create one long line (editing email) and 
will turn on and off soft wrap, regardless of whether this new behavior 
is the default or not. For the remainder, will they actually care if the 
behavior changes? Definitely seems worth putting out a pre-release for 
commentary. I imagine we might hear different responses from folks who 
use it for general purpose editing vs email editing, but then people are 
surprising and seem to react more strongly to minor changes than major 
ones sometimes.

So yeah, long way of saying I think I am in favor of it being the default.

> Let's not get ahead of ourselves and wait and see what people think.  (I set
> CUSHION = 3 just to distinguish my patch from Xylia's.  Maybe CUSHION should
> be 2, to make the behavior more similar to single-line sidescrolling.)
> 
> People that want to experiment with CUSHION can recompile nano.  But...
> I think a fixed, dependable, standard behavior is better in this case.

It is of course ultimately up to you. Xy really initially liked the 
'jumpy' behavior (hence implementing it that way) but I think would have 
been perfectly ok with a larger or settable cushion value. I think she 
will probably keep her own fork/branch going which just overrides the 
default if its not a tunable, heh.

But yeah I am interested in what others think.

Hey other people reading this thread, what do you think? :)

-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQSNpv57+npBirPLI1S8s1bfkQCfpwUCaWfZLQAKCRC8s1bfkQCf
p6HwAKDdzUBUH9Z9a5kl3aWjca0mT9uOeACfbfMw5a9pUx78dP6ZQtTs8SnoH7g=
=b3Xj
-----END PGP SIGNATURE-----