This is so funny that I'm not even sure what to say. You can ask your exact questions about a newspaper but somehow 99% of people manage to read them just fine. I think it's just a you problem that you are looking for an exact algorithm on how to scan a page with multiple sizes of content; in reality, people just look over it all and keep track of what they have or haven't looked at all in their heads.

▲

snackbroken 2 days ago | parent [-]

In a newspaper the answer is simple. You linearly scan the leftmost column to the bottom of the page, then the next column, then the next, and so on until you get to the end of the page. At no point do you ever need to keep track of anything other than "this is how far I've gotten" to make sure you haven't missed anything. Columnar layout make sense in newspapers because both axes are fixed in size, so all you ever do is one long linear scan with wraparound.

	▲	satvikpendem 2 days ago \| parent [-]
		If one axis is fixed, and it is in the case of grid lanes (it's not a fully pannable infinite canvas like Figma after all), you just keep reading the content that's on the current screen, then you scroll. I really don't see how it's any different to, for example as I mentioned previously, a long newspaper with many pages; each "page" is one "screen" worth, analogously. It's like infinite scroll, either vertically or horizontally, where instead of just one item in the list, you have a few. And if we're being really pedantic, Figma users do perfectly fine keeping the context of the content in their minds even in an infinitely pannable canvas. And also, generally newspaper readers do not do what you say, scanning column by column, they instead glance their eyes over all of the headlines then pick which one looks good then they read the article attached to that, it is a free form process. So again, I will contend that this is not a problem for the average reader. I really cannot see where the difficulty you seem to say lies.