I start with a general concept, then build up an outline with a few scene ideas, then move on to a specific page breakdown. The page breakdown will have dialog and actions written out, sometimes it's just [they argue] or [they fuck] and I fill in the details later.

I usually have the entire page breakdown written before starting a comic. For this comic I would sketch a few pages, do the dialog to make sure the page breaks lined up, then color that batch, then post and start on the next batch. If you're asking if I made this up as I went along, no - the loop was the concept from the beginning. When I got the the overlapping parts I massaged the dialog and action a bit to make sure things mostly fit with what I had already written.

For the next comic, Crossworld VR, I rough sketched the entire comic with complete dialog before coloring anything. Partially for efficiency but also since it was a commission I wanted the customer to approve the plot before I commit to a huge amount of work.

Some possibilities:
- Waiting for round 2
- Bound partner has misbehaved
- Threesome/Tag team

Multi-answer, or approval voting, is the superior type of voting system. It avoids the spoiler effect and you can safely vote for third parties without "throwing your vote away"

Well, I wouldn't say *always*, but we've got a couple of months at least.

Facial expressions are a weak point of the AI models I've tried. They can do angry, happy, crying, and neutral pretty reliably but a sideways glance or looking down in embarrassment or a conflicted expression are difficult to wring out of it.

Control net can probably help but it's fiddly.

They probably have advanced roombas or something. Oleophobic surfaces too.

This lady does not have a tag because she appears entirely within this comic and is unlikely to return.

There are 10+ bit characters in the Lighter Chains universe that don't have tags that I think would be more interesting.