New York Times, Monday, August 5, 2019

After watching the Westminster Dog Show in February 2018, my crossword theme ideas started going to the dogs. I had a puzzle published in another venue with the title "Top Dogs" — down phrases that started with a dog breed — and then I came upon the phrase "Downward Dog" which I thought would be fun as the revealer for another puzzle, with down phrases ending with a dog breed.

My first grid included "GOOGLE DOODLE" which was rejected based on "Doodle" not being well enough known as a breed itself (as opposed to, say, "Goldendoodle" or "Labradoodle"), so I reworked the grid, and the puzzle was accepted in October 2018.

With the dog days of summer upon us, I think today's publication timing is just perfect. Hope you enjoy!

Crossword constructors live a dog's life. I wonder if that's what inspired Tracy today!

The great goddess Crucivera only rarely sends a seraph to float an idea within your grasp. Even if you can catch it, it's a painful path to publication. The rough timeline for a Monday puzzle, in a typical case:

• Ideation: 3 days
• Gridwork: 3 days
• Cluing: 5 days
• Waiting for Will's yea/nay: 3 months
• Revisions: 5 days
• Queue time to publication: 9 months

It's soul-sucking to wait with bated breath until the day finally comes that you can post on Facebook that LOOK MA, I'M NOT A TOTAL FAILURE, SEE?

(Humph. You make puzzles? You brother saved eight people's lives today. Why can't you be more like him?)

Even if you don't have Asian parents, you too often get scooped because of the long lag time. It's especially rough when someone else's puzzle with the same theme comes out less than a week before yours.

A vast majority of solvers won't be aware of the AVCX crossword since the NYT's solvership is so much bigger. Shouldn't matter, right? But it matters to us constructors.

Will has done a lot over the years to shorten the overall process, which used to be perhaps twice as long overall. I appreciate that he's brought on more people recently to help reduce the 3 months of waiting for an answer. I bet he'll eventually bring that down to 1 month.

Queue time to publication is a harder one. The WSJ is much better about that, averaging about 3 months, but they don't get nearly as many submissions as Will, who gets about 125 a week. Maybe Mike--who edits the WSJ and is close friends with Will--has some advice on how to reduce that painfully long lag to publication.

Tracy's puzzle is solid, with excellent themer choices, and top-notch gridwork — such a hard task with five longish themers. It's a near-perfect gateway puzzle, one I'd gladly hand to a newb. It has to bittersweet for Tracy though, given the circumstances.