My point with RtL or LtR was never to point one better than the other, but to open the discussion of those in further arguments that are not focused only in culture, but also that those can have a meaning and a proposal; a way to make people discuss them selfs the subject.
Now with page composition, I don't see one better than the other, both have their strong points and their weak points, but I want to bring to discussion the "how we read"(continuing the previews subject).
In the classical panel format of paneling the sequences work as a tempo manager. Biggest frames represent: A) something with relevance and B) something with a extended period of time. While the space between panels represent the time between the panels (which are understood as events).
In the following comic page we see: Panel 1 is big, which mean it has a relevance, in this case of context, also provide a tempo longer than other panels. In Panel 2 the size is smaller which mean its relevance is less and also the time inside it. In Panel 3 is even smaller, which mean its relevance is just to show the driver. In Panel 4 we have a knife with blood(important) with a size of 1/6 of the page and finally between page 4 and 5 we have a bigger space, which mean there was a bigger frame of time between both events (is not big like a hour big, but is big enough to make us think that it didn't happened immediately). In panel 5 we see the knife being holden on the rain.
Now, we can understand this comic, without even read the dialogues, all thanks to the paneling. We see in the Panel 1 and 2 a car (we presume is the same car as no one have showed us another), there is no arm holding a knife coming from the car, so the knife was only pulled out of the car in the Panel 5. In Panel 3 we see a driver and he doesn't seem to be holding the knife, which mean he wasn't the killer (we presume there is a killer by the fact that there is a dead guy, people and a murder weapon) but as its the only car in it and the killer is obviously in the car, the driver is involved in this. Then in Panel 4 and 5 we see the knife that in Panel 5 is pulled out of the car, and to make it more tasty to read they put a Panel that is over Panel 3 and Panel 5, which mean that is something that happened between those panels, and it show us a dead guy, which we will believe it was killed with the knife at the end on Panel 3, Panel 4 was to prove that the knife was the murder weapon and panel 5 was to prove that the murder weapon was pulled out of the car. And this will help to make sense to a next page were the knife might be dropped out of the car, it would make completely sense, and we just understood the whole event of that page without even reading the dialogues, only understanding the paneling, aren't comic books amazing?
But now lets see a "modern" way of paneling.
And now lets dismember this.
Panel 1 is big (like 1/3 of the page) and show us an eye, we presume by our same logic that this has relevance and have a longer tempo. In Panel 2 we see a samurai armor, and is also big compared to a normal panel so we will presume it has some kind of relevance. In Panel 3 is small, very small, and we see the face of a character who is not the same that the one of the first panel as we can see by their attributes. In Panel 4 we have very small panel with a shading something (I don't know how to say it). Then we don't have a panel actually (its a panel in some sense) and only show us a character (the one of panel 3). And finally we have Panel 5 with the foot of the character.
Immediately we face a completely different logic; the armor, eye and foot doesn't make any logical sense, and this is because it has a different sensitivity. It work in a symbolic level, where eyes represent a character looking something (not just looking but watching carefully, with a high intention), the armor has no sense at all, probably its used a symbol of conflict, but this could be told without the armor. The small panels with the face and shades are representations for the individualism, specially to say that the character is having a monologue. The character without panel is to represent complete individualism in the monologue and finally the foots are to testify the dialogue.
As we see, in this "modern" way of paneling, we are not telling the story through showing what is happening, but its a testify of the dialogues. Now, this "modern" example is bad, because there are great works on it, and this is pretty lame (you see that some panels doesn't make sense, the tempo doesn't make sense, and its to focused on monologues). But lets see one that does it awesomely.
I'm not going to say how to read it, you guys should read it and see how is working.
But this problem of doing bad panelings is not just from "modern"paneling, its also for the "classic", as a prove, lets see this one:
If you guys read the tempos, you will see it has a lack on the sensitive and continuity of those.
Then how should we paneling our pages? Should we focus on the use of panels constantly? or should we break them constantly? Should we use image to testify dialogues? or Images to tell the story? And which one do you prefer?
Finally I think that every person who want to "do comics" needs to read Sandman of Gaiman. Its one of the comics that use paneling composition in a bold but smart way and is also one of the greatest stories on comic books.