The Future of Puzzle Creation with Krazydad

Hi! Can you tell us who you are and about your work?

Hey there! I’m Jim Bumgardner, and yeah, my surname’s funny, and I’m also known as Krazydad of krazydad.com—a long-standing website dedicated to printable puzzles, mazes and more. Currently, I work under the California Office of Data & Innovation and I’m tinkering with the state’s COVID-19 website.

I’m a solo developer who has been keeping krazydad.com updated and maintained since 2005, and everyone can find a puzzle that tickles their fancy in my extensive selection that spans from Kakuro to custom-made sheets designed for kids and the like! We also have a dedicated community that regularly updates and submits their crafted puzzles, so I’m sure you’ll definitely find something that clicks with you when you get into puzzles like we do.

How did you start becoming a developer?

When I went to college, I took on music composition as my focus, and that actually paved the way for my entry into the development industry. Around the 70s, we thought that computers were evil and incompetent, as the media at the time portrayed it like so. 2001: A Space Odyssey had HAL 9000, and Star Trek had all these tech and computers, so I went into the 80s with a very low opinion of computers in general.

But I got into modular synthesis at that point, and a modular synthesizer is technically an analog computer. Since I was into music and synthesizers, I ended up getting sucked into using computers to create music! There was this composer named John Cage, known for his piece 4’33, and he did a lot of music that involved chance operations called aleatoric music. Aleatoric music basically relied on chance to guide the composer—which goes against the normal process where composers decide what happens to a piece.

I then started to work on a piece for a synthesizer and tape to create that, but it needed random numbers to produce what I had in mind. So I bought a little Timex Sinclair computer for around a hundred bucks, and I wrote a basic program that simulated the I Ching Yarrow Stalk method and generated a ton of numbers. I used that to create music and well… that’s how I ended up getting interested in developing. Really, at that point, I got more into development and produced less and less music.

Did it cause you any regrets or doubts about using technology to further your pursuits?

Nope, not at all! There are some very adamant believers that music will always be at the top when humans are doing it all. These believers stick to the idea that there is something unique and special about the human ability that machines can never be able to recreate, ever. Heck, some even say that it’s because of quantum physics that humans can be as musically talented as they are, and I really don’t think so. Sure, human talent is important, and these machines are still ever-pressed to catch up to human ability, but many factors need to be considered. The skill of the programmer and their musical inclinations are just some of them.

As for the gaming side, there are die-hard puzzle solvers and makers who swear on manual construction alone and are irked by the use of computer-generated puzzles. I do understand their point of view, especially since there are creative inspirations that only humans can achieve at a fast pace. Even if I am a puzzle creator who uses computer generation for my work, I do know that technology isn’t quite there yet.

But look at today! The detectable differences between human and machine-generated art are getting less and less, and the gap between raw talent and technological advancement is shrinking. Frankly, I think in less than a century, computers will be able to catch up to human ability. So yeah, I think that computers will eventually win in that aspect, and I don’t regret undertaking this path.

What are some memorable things that you have in mind in regard to developing and tech?

Ah, I have just the story! I’m the mayor of the North Pole. Well, technically! Alright, remember Foursquare? It was an app that basically gamified location tracking—if you kept checking in, let’s say, a Starbucks, and if you checked in the most, then you’d end up becoming the “mayor” of that Starbucks location. And if you had friends, you could also check out where your friends last checked in, there were badges and stuff… it was a very awesome app.

So I was having breakfast with my wife one day, and a thought occurred to me: “Could I become the mayor of the North Pole through Foursquare?”. I checked in the app and sadly, there was no North Pole. But wait, you could introduce new locations. That gave me an idea. I then figured out how the API works, and I made a script where I could introduce a new location. I also found a way to introduce en masse locations that did not exist.

So I checked into the North Pole and made myself the mayor. I burst out laughing when I did it, then I tried checking into the Taj Mahal? I could. So I did. Statue of Liberty? Yes, I can. So in a couple of days, I checked into all these world landmarks in just two days! I was the mayor of the Taj Mahal, Mount Rushmore, the Playboy Mansion, the Metropolitan Museum of Art, and a whole lot more. It even got to the point that I became the mayor of over 200 Starbucks locations, and the former mayors of these locations didn’t like losing their spot. I also distinctly remember one of the employees at the Statue of Liberty was not happy about that! Aside from that, I also made fake accounts of Martha Stewart and Simon Cowell, and would have them check in at a bunch of places, then people would end up flocking to these locations. Fun times!

But seriously though, this technically counts as hacking. This was just harmless pranking, but I don’t recommend doing that nowadays since digital stuff is now more heavily guarded and there are more laws surrounding tech in general. I actually posted about this in my blog… the security of Foursquare, I mean. The blog post ended up getting featured in a New York Times interview with the Foursquare creator Dennis Crowley and we had a nice conversation.

What led to you starting krazydad.com?

At the time I created krazydad.com, I was fully immersed in getting as much joy as I could out of recreational programming. Yes, I enjoyed and continue to enjoy programming, alongside math and puzzles, and I consider them all as great recreational activities. I remember in the 80s, I was a big fan of the columnist at Scientific American who took over for Martin Gardner, and older people would probably remember him as Doodney, and they had an article called Computer Recreations. Doodney’s article was Mathematical Recreations, and it was basically a programming column made for entertainment. I remember getting inspired a bit by that.

Then came the time that I was teaching myself C, the programming language. I wanted to try my hand at making the popular apps of the era. So I ended up making a spreadsheet, a word processor and all those things. And every time a new trendy app would come out, I would end up thinking to myself: “How does that even work?” Like, how did computer chess work? How did this and that activate? A lot of them didn’t look intuitive to me at all, as I didn’t know how those would work with conventional programming techniques.

Afterward, the Sudoku craze hit around early 2005. It was already a hit in Britain, and the U.S. was now jumping into it as well. Of course, it managed to pique great interest in me as well! And even back then, there was tons of info online teaching people how to create their own Sudoku puzzle. A lot of enthusiasts back then were focused on making them as hard as possible, and most of them were also approaching it mathematically.

I ended up joining the community. And at some point, created a puzzle generator that would output PDF files so I could print them and solve them all with a pencil since that’s how I like to do it. Time passed a bit and I had several thousands of PDFs that I had no idea what to do with. But I had a website called krazydad that I made to sell screen savers. It was a small business, really, and it wasn’t that fun to have due to the customer support aspect of it.

So I thought that I should put the puzzles in krazydad.com and see what would happen next! Then it rapidly took off, the puzzles taking the forefront more and more and screen savers less and less. And within five years, the screen savers were all gone and everything was all puzzles and mazes. Yeah, krazydad.com, the puzzle website, started from screen savers.

Can you tell us more about krazydad.com?

Sure thing! When it got rolling as a puzzle website, it was all PDF files. From 2005 to 2014, all the puzzles on the site needed to be printed to be played. After 2014 though, I started making them interactive, and it was one of the major milestones of the site. When I introduced interactive features to the site, it started to outpace the usage of printables, though they’re still quite close by the way. Right now, the stats on the site say it’s 52% usage on interactive and the rest is for printables. But before the pandemic, the numbers were pretty even. After COVID hit, that’s when the numbers started to cross more into the interactive side.

The reason why I started introducing interactive puzzles is that I noticed that the audience was changing, and that same audience is quite the beast to tackle. Back when I started, most of my audience was in their 40s who were looking to improve their brain health and other people who wanted to print puzzles for their elderly parents. And over time, I knew that I had to adjust since the shift to digital stuff is becoming more and more widespread, and that’s what compelled me to add the interactives.

As for my selection of puzzles, I think there’s a total of 52 puzzles in there, with several of them buried under sub-menus. I’d also confidently say that about 98% of the puzzles on my site are created with the generator I wrote. There were a couple of puzzles that I borrowed a generator to create during the early stages of krazydad being a puzzle website, but in time, I learned and switched to generating them on my own as well.

Which puzzle was the most difficult for you to create?

I’d say that it’s the Slitherlink. The generator used on it works in a different way than the rest because there’s spatial reasoning involved in it. Most of my other puzzles are created with a general-purpose generator that’s capable of creating several variants that I consider to be types of Slitherlink. I initially used Processing language on it, but I ended up going back to C due to performance issues. So yeah, while I want my constructions to be as uniform as possible, if I need to use something different to ensure a well-performing result, I’ll definitely do it.

How do you monetize your work?

Monetization has been a long journey for me. I primarily use a donation-based system on krazydad.com, and I’ve noticed that there’s a psychological element to it. People who prefer to use printables are more likely to donate over digital users because they have something that they can hold in the form of puzzle sheets, like, there’s a physical product that came from their donation.

This made monetizing the digital puzzles tricky, and I tried a lot of different methods, but I did significantly improve the donation rates for the interactives. Aside from the website, I’ve also extended my revenue earning to different areas. I do publishing and selling to newspapers as well, and I’m in a good relationship with The New York Times. NYT publishes my Star Battle puzzles from Monday to Saturday in their print edition. I’ve also sold my puzzles to Andrews McMeel in Kansas City, especially my Sudokus.

But you may have noticed that I still haven’t talked about ads, which is one of the most common revenue sources for most websites. I used to have ads on the site. In fact, it used to be the biggest and primary revenue generator for it! Back in the early 2000s, Google was still in its “Don’t Be Evil” phase, and its ads were great back then. In the past, there were a lot of awful banners and pop-up ads that everyone hated, and Google came along and offered better ones. All text, relevant to the site and not selling you… ehem… male performance enhancement drugs. So I took them on and things were great for a while and they paid really well!

Over time, however, Google became the thing that they wanted to fight back against in the beginning. The ad revenue dropped significantly, and the relevancy of the ads also became a problem. From 2012 through 2018, if you had a website that had PDF downloads and ran Google Ads, you’d get ads that had a big, green button that said download. Of course, the computer literates knew that it was just an ad that would redirect you somewhere and install some malware on your computer. But what if you were someone over your 50s and had little to no experience with computers and the internet? You’d be very confused and end up clicking on this crap that will cause you problems even though you just wanted to download some PDF puzzles to play with. It was nasty, really.

So after that, I stopped using Google Ads, and there was an immense improvement in the overall experience, for both me and my users. As for the loss of revenue, the donations also improved and picked up the vacuum left by removing the ads. Though several times a week, I get unsolicited emails from ad agencies that want to “help” monetize my website. Usually, they just go to the trash, but when I feel quite spicy, I reply with a little rant then into the trash they go.

All in all, I operate at zero income. What comes in goes completely for the site and I get little to nothing for myself out of it. And that’s completely fine by me, as I prefer it this way for both my mental health and the experience of my users.

Do you regret turning your back on ads? Did you ever try to re-introduce it?

All this time, my main goal was to keep the traffic flat on the website. An upward trend would be very much welcome, but I was more than satisfied to keep it flat and not sink. That’s why I introduced interactive puzzles too, because I knew, at some point, the flat would start to sink, and the digital puzzles helped and continue to help keep the traffic flat.

As for the ads, I actually did try a bit of ads after a while. There was a slight improvement, yes, but I really think it wasn’t worth the effort and time. I got lucky that I ended up having continuous organic traffic, and the last thing I want to do is to cause a mass exodus among my users. Ads are one of the most hated things any user can think of, and I really think trying to implement them back will be counterproductive for me.

Donations are fine already, and the site as well. I don’t want to disturb this long-standing peace. But funnily, I did do something with the donation process that actually improved the rates. So when you load my pages, you’re going to get different donation messages, and they’re all at random. And I’ve added many different types, like, some have funny jokes in them with animal images, and some that have heartfelt messages that I personally wrote. I wouldn’t say it’s emotional manipulation or anything like that, but there is also an emotional aspect to improving the donation rates, and it has worked for me. But I still haven’t done research on which ones were the most effective, so I’ll probably get to that.

Do you think puzzle creation is a profitable venture?

That’s tricky, especially nowadays. Will Shortz is probably the closest to being a puzzle constructor among us, but most of his money doesn’t come from the puzzles he constructs. As I said, I operate at a near zero with krazydad.com, and the money I make from my different avenues of income like publishing and syndication is not enough to sustain me completely.

Puzzle creators do what they do because they love doing it, and it’s just happenstance that we can make some money off of what we love. But really, we can continue doing this comfortably because we have a day job, and day jobs aren’t always things that we like to do, but we have to in order to sustain ourselves.

Can you tell us more about how you got noticed by the NYT?

Definitely! So back during the height of the COVID-19 Pandemic, a lot of people were stuck inside their homes. People were looking for things to engage them, from cooking to binge-watching series on Netflix, and puzzles were also high on the list. The newspapers felt it, especially NYT, so they wanted to expand their puzzle offerings since they were pretty much just doing only crosswords and KenKen.

At that time, Will Shortz was working on a full half-page puzzle section that would have new puzzles. He was also very addicted to a puzzle called Star Battle during this period, and back then, I was the main supplier of computer-generated Star Battle. He was going through my site and doing my Star Battle puzzles very enthusiastically. He then contacted me out of the blue and asked if I wanted to run puzzles at the Times, and I immediately agreed to it.

But there was something that needed to be done first. Will didn’t like the name Star Battle—there weren’t any stars in it or any battles at all. So he published them under the name Two Not Touch, and it was perfect since it basically described the whole game in three words. Two Not Touch is what I’d call a Goldilocks game—simple and fun yet had the right amount of strategy involved in it. No single way works for every puzzle, just like with Sudoku and the other classic sheet puzzles. There are techniques involved with the more advanced difficulties, and I’m glad it took off as it did.

Before I started doing these Star Battle puzzles for the NYT, Sudoku was the most popular game on my site. But now? I can confidently say that Star Battle has a slight edge over Sudoku on krazydad.com.

Aside from Will Shortz, did you interact with other notable names in the puzzle industry?

Oh yes! Bob Fuhrer. And we all know him as the father of KenKen. And the way I got to know him on a personal level is a funny story. I was aware of KenKen of course, and I felt that I had to get some of these on my site. So I started publishing a KenKen-like puzzle, and I called them Kenny’s. And within a week and a half or so, I get an email from Bob Fuhrer or probably his lawyer that basically said: “Hey, we got a trademark on this, maybe you could use a different name?”

Of course, I happily changed it, I didn’t want to cause trouble after all. So I changed the name to Inkies, and I continue to publish Inkies up to this day. Afterward, I met Bob in a meeting under the Spec LA—which is basically a group of makers, nerds, artists and creative people. We meet up monthly in Spec LA and we do demos for each other, and what’s cool is that it’s very loosely based on the Royal Society.

One of our members knew Bob and they invited him to the meetup, and I greeted Bob there and told him my name and that he sent me a cease and desist before, but in a friendly way! We hit it off from there and we continue to spend time together with our fellow puzzle enthusiasts. Then came a time when he invited me to Will Shortz’s house for dinner, and the CEO of Andrews McMeel was there too. And that’s also where I ended up being syndicated with Andrews McMeel and doing Sudokus for them.

Imagine, from receiving a cease and desist from a guy to becoming syndicated with that same guy’s help! Life truly works in mysterious ways.

Is there any advice you’d like to offer new developers?

With how long I’ve been programming, there are a couple of things I’d like to tell budding enthusiasts:

1. Don’t force yourself to stick to a fixed path

See, in the beginning, I wasn’t even an avid computer nerd. Remember what I said about having a low opinion on computers back then? I took on a music major back in college. And look at me now, creating all these generators and codes and music with a computer. So don’t force yourself to stick to something that you pre-determined, especially if it’s going to hamper improvement and innovation. 

2. Learn different programming languages

Sure, you’re bound to learn different programming languages if you want to become a successful developer. We all have a language that we’re proficient in and love to use, but don’t box yourself into them as other languages may offer benefits and ways out that your favorite can’t give.

Are you playing any games right now?

The last big game that I played was Zelda: Tears of the Kingdom, and that was a few months ago. I actually am not a big gamer, to be honest. Sure, I make a lot of puzzles and mazes, but I don’t keep up with the latest gaming trends and stuff like that. See, I’m not particularly fond of point-and-shoot games like most of the youth and young adults are geared towards nowadays. I love puzzles though, and I enjoy playing Chess from time to time. But really, what I do most of the time for entertainment is cooking and music!

What’s your plan for the future?

I think most of what I’ll do is pool my knowledge and do more research on my own site, like which emotional trigger in donation messages works best in increasing its rate. Aside from that, I’m also working on a guidebook for puzzle creation and generation using Python. To be honest, if I get enough supporters and backers for the book I’m going to create, I might just retire and entirely focus on it! But I don’t think it’s a financially sound model, so I’m still going to mull over my choices.

As for the puzzle industry, I know that I still have a lot to offer and say, so I’m going to continue enriching my website and joining my fellow puzzle enthusiasts in my community.

Where can we learn more about you and your work?

You can use my handy dandy email address, dad@krazydad.com, for any concerns regarding puzzles or anything about krazydad.com. As I say on my website, if there’s a puzzle with a duplicate solution, tell me about it and we can look at your alternate solution together. You’re all welcome to join my Discord server to talk all about puzzles and creation as well!