Re: Proposal: define scroll amount via CSS

Hazel Seanor skrev:
> Hello,
>
> When scrolling on web pages using the space or page down keys, the
> browser takes the viewport height and scrolls down by that amount.
> This is normally fine (and expected behaviour) except when fixed
> headers are used. Because the entire page size is used and not just
> that which is visible to the user, the page scrolls down too far and a
> few lines of text end up positioned behind the header.
>
> The issue is further described in this Bugzilla thread from 6 years
> ago: https://bugzilla.mozilla.org/show_bug.cgi?id=780345
>
> Firefox's solution was to detect when a fixed header was being used
> and adjust for it, but this doesn't detect all fixed headers (possibly
> because there are multiple ways to implement fixed headers) and while
> other browsers may have implemented this or something similar, Chrome
> has not.
>
> It is also possible to account for this with JavaScript, but this is
> far from an ideal solution. It seems to me like something that should
> be able to be manipulated with CSS. As well, as an accessibility
> concern, it should be something that can be fixed across all browsers.
>
> My proposed solution would be to add a new property, scroll-amount,
> which would be set to a size in units.
>
> This addresses a separate concern from CSS Scroll Snap because when
> scrolling with CSS Scroll Snap, you are brought to a predefined point
> on the page. The point of this change would be to allow a developer to
> define how far each button-press would scroll down the page, from any
> point to any point.
>
> Typical usage (with an 80px header) might look something like this:
>
> body {
>     scroll-amount: 100vh - 80px;
> }
>
> I haven't made a github issue because I'm not sure what to tag it as.
>
> Thanks,
> Hazel
To me this sounds like a case of solving problem Y instead of problem X.

What I see as the problem here is that you want to override the
scrollable area. Not the scroll step size.

Also, the scroll step size is an user preference.

As for that solution Firefox is supposedly using, more details would be
nice. Perhaps that is something worth using as a base for standardizing
instead?

Received on Monday, 16 July 2018 15:34:39 UTC