My software wish: A text-to-comics program
There seems to be a computer program or a website for everything. Some of them are innovative killer applications, others… not exactly useful. The past few days, the only topic in local newspapers and the biggest topic on Twitter has been Google‘s Chrome OS. While I can’t say I understand the whole “Hey guys, we’re making a new OS that’s really going to kick Microsoft’s ass, because it’s free, and it’s based on Linux” philosophy, Google’s innovativeness is somewhat fascinating. Facebook and Twitter are great examples of popular websites we can’t live without anymore — according to a recent survey, 43 % of all Norwegians use Facebook regularly. Saturday I quoted Brad Guigar’s July To Do-list, where one of the points was to not forget using Facebook and Twitter correctly to promote your comic. I guess most webcomic artists out there know how important that tip is, eh?
The reason these applications (and others) are so popular, is because they fill a demand. People want to be able to connect with their friends. People want to share what they’re doing with the world in 140 characters or less. People want a new OS based around the world of the web. And you can probably say the same thing about all popular software applications: Microsoft Word is popular because people need a program to type letters and reports in (of course marketing matters, too: The reason MS Word is more popular than Word Perfect is because people want the same format as everybody else, and Microsoft has the best marketing). Programs who fill demands rule the world. When I created Nettserier, it was an attempt to fill the demand for an easier way for Norwegians to publish their comics online. What I actually created, without being aware of it at the time, filled another demand — the wish for an online comics community in Norway. Of course, Nettserier is no killer application, but it’s still more popular than I would ever have expected when I launched it — because there exists a demand for exactly this kind of service.
So, here’s what I wish someone will come up with next:
I want to see a computer program that will convert written text into comics.
If this already exists, please let me know in a comment. If not, read on to hear more about what I want from such a program.
But to specify: I don’t want the program to turn the script into ready-for-publication comics, of course. But rough sketches. Storyboards. Something I can show off to an artist and give them an exact image of how I picture a scene in my mind.
When I write comic book scripts, I normally use the same standard phrases over and over. It wouldn’t be a big problem for a computer to interpret them and convert them into images. Especially if I don’t need to show those images to anyone except for artists and editors.
For instance, take a look at this rough comic strip script (it’s not an existing strip, I made it up for this purpose, but you get the idea):
Picture 1: Larger picture. A sunny meadow. Frank and Jane are having a picnic. Frank is sitting under a tree, happy. Jane, standing, is annoyed, looking at the ground. There are ants all over the ground.
Frank: Boy, you’re an excellent cook, Jane!
Jane: I just wish there weren’t so many ants here!
Picture 2: Darker. Close-up on Frank, worm’s-eye view, now angry, holding a machine gun and firing towards the ground repeatedly.
Frank: EAT LEAD, ANT BASTARDS!
Picture 3: Same scene as picture 1. Frank is happy again. The ants are gone and the ground is filled with bullets. Jane is shivering with fear.
Frank: There, much better!
Jane: You freak me out sometimes, you know that?
Not a very creative script, but you get the idea. So, I want a computer program that can look at this script and turn it into a rough strip I can show to my artist. Why? Well, perhaps how I vision Frank’s rage in pic 3 doesn’t come out clear enough in the script, and the artist will draw it in an entirely different way.
The software would first look at the script as a whole, count the number of frames and assume that I want this to be a comic strip, so it displays a comic strip to me. I could, of course, click a button to change this to a full-page graphic novel format if the situation requires it.
The first sentence of picture 1 is “Larger picture”. Based on this, the program would make the first frame perhaps twice as big as each of the other pics. If I don’t like the size of it, I can grab the gutter and slide it back and forth until it satisfies me.
The strip takes place in a sunny meadow. The software could have a large archive of stock backgrounds, find an appropriate meadow, and put that in here. I imagine that stock art, both backgrounds, characters and objects, could be saved as three-dimensional objects so that it’s easy to rotate them and so on. The information I give later in the description will help the program figure out what to do with the backgrounds, as well. The sentence “Frank is sitting under a tree” tells the program that there has to be a tree in the frame. If the program has several meadows to choose from, an icon under the frame would indicate that I can click there to change the background. Finally, the single word “sunny” tells the program that this should take place in daytime, and that the weather is nice.
Now, the next sentence would offer the program some problems. As I only write “Frank and Jane” without further introductions of what they look like (I could do that, but I’m too lazy to do that in every strip), the program is capable of understanding that they are characters, and based on the names that they are male and female, but it will probably input stock characters that look nothing like what I have in mind. No problem — I can click on the character (either directly in the strip or on the character’s name in the script) and pick another design. I imagine that the program could easily hold tons of generic character designs more than good enough for my use. Also, it doesn’t have to be difficult to make small changes to these generic designs. For instance, if I’m writing a seven-volume graphic novel about a young boy with glasses and a lightning-shaped scar on his forehead, the program wouldn’t have something that specific in stock, but I could easily open a wide selection of scars (or acne, or warts, or birthmarks) and drag the desired scar right into my character design. And then, I could save the character as Frank (or Harry, or whatever), and the next time I write a comic about Frank, the program will recognize the name and use the same design for him. And of course, if I write about a character with a name that sounds nothing like a character name (for instance, if my character is named Stapler or something), the program may not understand that it is a character, but then I could just right-click his name in the script and choose “this is a character”.
As the program knows that this takes place at a picnic in a sunny meadow, the characters would get clothing suited for that activity. Here, too, I can click on the clothes and change them if I don’t like them. But what if I don’t like the positioning of the characters? What if I feel that Jane is standing too close to the picnic rug? No problem — I can just click and drag them to the desired position. Of course, they will be bound by gravity unless I specify that they shouldn’t be, so I don’t have to worry about putting them down three feet above ground. I imagine that unless stated otherwise, the starting position of the characters in each frame (and in the upcoming frames, how the scene is rotated) will be based on the order of their dialogue balloons.
Of course, expressions would be a very important part of such a program, and the various degrees of expressions would make it difficult to give the characters only a given set of stock expressions, so that would have to be more complex. In Making Comics, Scott McCloud describes how all feelings can be made by combinations of six primary feelings, and he’s written in his blog about how these primary feelings can be mixed and matched in easy computer programs. Something like the Grimace Project would be the perfect tool for facial expressions in such a program. I’ve told the program that Frank is happy, which possibly could be translated into “80 % joy, 0 % all other feelings”, but if I disagree with these settings I can easily right-click Frank, choose “change expression”, and drag those sliders a bit. That would be perfect for Jane — her “annoyed” face could probably need a little added disgust. Just as important as the facial expressions would be the body language — here I imagine something Spore-like, letting me drag and rearrange those elbows, knees, hands, feet, shoulders and butts until I have the expression I want (but again, the software should make a decision based on what I’ve written in the script, I only need to rearrange these things if I’m not satisfied).
Objects in the frame would be added based on what I’ve written in the script: There’s a picnic here, and there are ants on the ground. Piece of cake. And then it would just have to add the perfectly-sized dialogue balloons, and the first frame is done. And if I write dialogue balloons with way too many words in them (I tend to do that), the program could give me a warning about it and ask me if I want to change it.
For frame two and three the whole thing would be a lot easier. I have already told the program where we are and what the characters look like. In frame three I’ve even said that everything except for what I’ve specified should look exactly like picture one. All in all, this program will turn my script into a visual representation, and it will take me a lot less time than it would take me to sketch it myself.
What’s the purpose, you may wonder? Why don’t send the script directly to the artist, or draw a rough sketch of it? I could do that, of course — that’s what I do today. And many comic book writers do their own artwork, so the market for such a program could be limited. Still, I believe that such a program would be a great asset for comic book writers because it would force us to, even more than today, think visually.
Like I said, I often write dialogue balloons that are way too long. When using a program like this, I would be made aware of that mistake and learn to change it. I also wouldn’t make comics containing things that just don’t work visually (for instance, a small frame with fifteen characters fighting in it) because I would get the visual before even sending it to the artist and figure out that it didn’t work.
The picture description and the content of the dialogue balloons would not be connected, so if Jane points towards a bus and says “Look! A bus!”, the bus won’t actually be shown in the frame unless I also mention it in the picture description. That could perhaps make me realize that I’m doing the classical mistake of letting the images and the words tell the exact same story, so if I choose to show the bus in the frame, I can change Jane’s line to something more interesting.
And how many times haven’t you heard writers and artists fight over how “This wasn’t how I pictured it!”? Well, giving the writer the opportunity to show exactly how one pictures the frame on paper would get rid of 99 % of those arguments. Of course, the artist may still want to change things, but then at least he can give the writer a call and ask him if that sounds good to him first, instead of assuming that this was what the writer wanted in the first place.
Don’t get me wrong: Using such a computer program would be a terrible way of making comics. But it would be a wonderful way of making scripts for comics.
So if you know anyone who’s making something like this, please let me know about it. The market shouldn’t be too limited — there are tons of writers who could use such an application, especially if one manages to get big publishers like Marvel, DC or Egmont involved. And it shouldn’t be too difficult to tweak such a program into doubling as a great program for making storyboards for movies, either.
Obviously, since I’m no expert, everything I describe in this blog will be a lot more difficult and more expensive than I imagine it will be. But I know that I would buy it.
