Reorder A and B in X (virtual channel construction)

[geoknee]

So that it matches the order in the text: A has priority over (is left of) B. Throughout the paper, A is red, B is blue.

Also recolour the "split circle" of X, so that A is leftmost and B is rightmost: this way it matches the ordering of the outcome of that channel.

Original:

Proposed:

[geoknee]

It's clear that (in the text) A has priority over B in the Joint Channel (J) and the virtually funded channel (X)

nitro-paper/nitro-protocol.tex

Lines 234 to 240 in f4c0a76

	\item $A$ and $B$ bring their channel $\rchi$ to the funding point:
	\begin{align}
	\finalizable{\rchi}{\alloc{A:a, B:b}}{A,B}
	\end{align}
	\item In any order, $A$, $B$ and $I$ setup the virtual channel construction:
	\begin{align}
	& \finalizable{J}{\alloc{A:a, B:b, I: x}}{A, B, I} \\

The text and the figures are internally consistent with themselves. To make them consistent with each other, we would have to have A being blue and B being red, which is different to the rest of the paper. The aim of the PR is to have the same scheme throughout the paper.

[NiloCK]

As you explained this, I approve. As I review the changeset, it's hard to connect the dots.

Can these changes be run through pdftex (or whatever builds the paper itself) so that the changes also show in nitro-protocol.pdf?

[geoknee]

Can these changes be run through pdftex (or whatever builds the paper itself) so that the changes also show in nitro-protocol.pdf?

There is an updated main.pdf checked in with this changeset. It won't be easy to find the "diff" for this file -- but you can "view file" an scroll through the whole thing. Since the changes are restricted to figures, it should be enough to check each figure for changes.

[geoknee]

This change is debatable: the split circle is used throughout the paper, including places where it allocated to Bob then Alice, or just to Bob. So the colour-coding does not denote anything about the outcome of that channel, only the participants of that channel.

That said I still prefer it this way. The participants are also ordered A,B so the colouring should be, too.

[geoknee]

It affects Figures 3,5,6,7,8,9,10,13,14,15,16,17,18,19,20,24,25,26

[geoknee]

The changes to this file only affect Figure 25. The change consists of permuting the positions of a2 and b2, as well as the edge connecting them with the joint channel. This happens in 3 of the 4 frames of the figure.

[geoknee]

The changes to this file redefine guarantee (used only in Figure 25 and 26) and guarantee2other (used only in Figure 26) diagrams. In essence the change swaps the outcome of G and G': so that the funding from (A,I) eventually prioritises A first in J (and likewise for B). In other words (during opening) the left guarantee now has output ports 132 and the right guarantee has output ports 231. The necessary follow on change is made to Figure 26 when G has been defunded.

This is a necessary change given the reordering of A and B in J.

[geoknee]

I have just spotted another issue, however, in that L and L' seemed to have switched positions in Figure 26 (w.r.t Figure 25), as well as the ordering of A and B in J. I think we ought to switch L and L' back.

[geoknee]

7fd8d55