Leaderboard

If you’re the image editor for a VN translation, you’ll probably spend at least half your time setting English type. Lots of it. (The other half will be spent laboriously retouching out all the Japanese text you’re about to replace.) Sounds simple on the surface, right? Any pixel monkey can copy/paste from a translation document. But there’s a lot more to good typesetting than just clicking with text tool and banging away on the keyboard. Just like good prose, there’s a certain rhythm to good type. A practiced designer will make numerous small adjustments along the way that allow the to reader glide effortlessly through whatever’s being said. Reading good type should be like driving on a well-paved road. And the one thing all good display type has in common: someone took the time to kern it. The Basics of Kerning I won’t go into a detailed discussion of kerning here — the terminology, the history, the fact that it sounds like something you’d have to pay an escort extra for. If you’re interested in that sort of thing, there are lots of sites out there for you to read. Better yet, buy yourself The Elements of Typographic Style, the best book about type you could ever hope to own. Call it an early Christmas present to yourself. For our purposes, it’s enough to say that kerning means to adjust the empty space between any two adjacent characters, either bringing them closer together or pushing them father apart. And why would you want to do that? Otherwise, you’ll have gaps and crashes in your type that’ll make things feel ever so slightly off. To illustrate, I browsed over to a free font site and downloaded a typeface at random. (There’s a very good reason I did this, rather than using some industry-standard font like Helvetica or Times Roman. I’ll get to that in a minute.) Downloading, downloading … done! Okay, let’s set some type. Here we have a few words set in Font X — name redacted to protect the innocent. At first glance, everything seems fine. But then you look closer and start noticing little things. Like what are these weird gaps between the first two letters of some words? Some are almost as wide as the full space between words. And hey, what about these letters over here that are more or less crashing into each other? That can’t be good, right? Nope. These are problems. They need to be fixed. Kerning Pairs The reason I picked a free font is because most professional typefaces (aka, “ones you pay a lot of money for”) are designed to avoid the majority of such issues. Once a typographer has crafted all the characters for a font, he or she will then spend countless hours specifying “kerning pairs” for it — basically, instructions on how close each letter should sit next to every other letter. (Here’s how close A should be to B, here’s how close A should be to b, etc.) While some letter pairings may look good at default spacing, others will need to pull tighter or push father out to look right. Professional fonts will often contain hundreds of these kerning pairs. It’s mind-numbing work that takes far more time than most amateur typographers are willing to put into a freebie font project. That work still needs to get done, however, but now it’s on your shoulders instead. And, since most fan translation projects use free fonts for budgetary reasons, odds are you’ll have a whole lot of mess to clean up. Congratulations! Thankfully, once you’ve learned how, it’s pretty easy stuff. I work in Photoshop, so I’ll be showing its kerning interface here. If you use another program, it likely has something similar. Here’s Photoshop’s character palette, with the kerning field highlighted: The "0" you see there means there’s no kerning currently being applied to the characters on either side of your text cursor. Make this number negative, and the two letters will start pulling closer together. Make it positive, and they’ll start pushing farther apart. (Photoshop measures this in units 1/1000 ems, but that’s bar trivia you don’t really need to remember. Just know that in most cases, you’ll be entering numbers in the range of -100 to +100.) You can see the results of some sample values below. In Photoshop, you also have the option of “Metrics” (apply whatever kerning pairs the typographer included in the font, if any) or “Optical” (let Photoshop guess what looks good, basically). Play around to get a feel for things, then advance your cursor through your type, letter pair by letter pair, and adjust this value until the two letters are the right distance apart. Rinse and repeat. And what’s the “right” distance? The one that looks right, of course. It’s a subjective thing, and this is where practice and design experience come into play. Like The Sands Through The Hourglass One of the first art directors I worked under offered me this analogy, which I’ve always rather liked: Imagine the negative space between letters as vases lined up in a row. They’re all different shapes, these vases, but you want each to be able to hold an equal amount of sand (or M&Ms or whatever). Kern until your vases all look like they could all hold the same amount. This is an imprecise rule, of course, and you’ll often want to make your “vases” bigger or smaller for visual effect, but it gives a beginner a good baseline approach. So let’s take that approach here. Let’s go through, fix all the obvious gaps and crashes we noted earlier, then make smaller adjustments to even out the text overall. (We call this giving the type an even “color.”) After some quick fiddling, we end up with something like this before and after: It’s subtle, but the "after" type just feels nicer overall. And if your text is a UI element that some poor reader will spend countless hours staring at, you want to make sure it’s as nice as you can manage. Because the longer you spend with something, the more obvious and annoying its flaws become. (Said every roommate ever.) The good news is you don’t need to do this everywhere. It’d be insanity to kern entire sentences or paragraphs of text, especially since the effect is barely perceptible at those point sizes. You only really need to worry about kerning display type — things like buttons, headlines, title screens, etc. If your type is over 16pt, it probably needs to be kerned. The good news is, as you learn the keyboard shortcuts for your particular application/platform, you’ll be able to breeze through a piece of type in a matter of seconds. In fact, a lot of designers find sitting down and kerning type to be mindlessly relaxing, like knitting or playing Minesweeper or making fun of the animations in Fallout 4. Mind Your Gaps So that’s kerning in a nutshell kernel. It’s the absolute easiest way to step up your type game, and it’s quick enough that there’s no reason not to do it. As a bonus, there’s a fun little online game out there to let you practice your kerning skills in hypothetical situations and compare them to a professional designer’s solution. It’s a fun way to kill some time at work while you boost your skills.

Making moeblogs in IPB4 is suffering now, I used to be able to copy and paste an entire post and now I need to manually embed each image and make sure they don't get messed up while I'm at it. There's a good way to fix those frustrations, looking at adorable MOE will make it all better, no matter what's troubling you!

How do you eat an entire whale? One bite at a time. Preferably with Cholula. How do you edit/translate/whatever a visual novel? One line at a time. Preferably with bourbon. Whether you’re a fan of the final product or not, one of the things that impresses me most about MDZ’s fan translation of Koisuru Natsu no Last Resort is that it got released, period. As in, if you were so inclined, you could download the installer right now, patch the original Japanese game, and go play the thing on your new-fangled Windows Pee-Cee. No demos, no one-route partial patches. The whole damned VN in English, finished on schedule and out there in the world. The project didn’t stall. It didn’t wind up in no-updates-in-six-months-but-we-think-they’re-still-working-on-it hell. It didn’t climb into that white panel van with Little Busters EX, never to be heard from again. The nice man was lying to you, Little Busters EX — there were no cute little puppies in the back. What were you thinking?! The KoiRizo team did nothing particularly special to make this happen. We just ate the whale one bite at a time. The rhythm method By his own account, MDZ worked very methodically on the project, spending an average of 30 minutes every day translating scripts into English. Not when he felt like it. Not when inspiration struck. Not when enough people harassed him with all-caps emails asking why the HELL hadn’t there been any progress updates on the KoiRizo tracker lately. He made it an expected part of his routine, like brushing his teeth or eating dinner. He scheduled regular translation sessions between classes or before heading out in the morning. He did a little bit. Every. Single. Day. There’s a word for that: consistency. That’s what gets things done in the real world, not 48-hour marathons every random.randint(1,6) weekends fueled by Red Bull, Hot Pockets, and intense self-loathing. Consistency keeps you from getting burned out. Consistency lets you make reasonable schedules and estimates, then stick to them. Consistency is like goddamned black magic. Over the course of the project, MDZ had consistency in spades. If he can maintain that approach to life, I have a feeling he’ll be successful at whatever he puts his mind to after college. When I came on board as an editor, I kept a somewhat similar schedule. I resolved to set aside my commuting time each workday for editing. And so for 40 minutes in the morning and 40 minutes in the evening, Monday through Friday, I’d park my butt in a train seat, break out my laptop, and just edit. Weekdays were reserved for my family. If you’re married with kids, you know there is no such thing as free time on weekends. If you’re not married and don’t have kids, please tell me what the outside world is like. I hear they came out with a PlayStation 2? That’s gotta be pretty awesome. Anyway, that’s what I ended up doing. Edit every single workday. For six months. Until it was done. (Six months? That long to edit a medium-length visual novel? Yeah, that long. KoiRizo weighs in at 36,000+ lines. Over six months, that works out to about 1,400 lines a week, or 210 lines per hour. That’s an edited line every 17 seconds or so, with most of the lines needing substantial polishing/rewriting. I have no idea what pace other VN editors work at, but I felt like this was one I could maintain over the long haul. Call it the distance runner’s lope.) Special topics in calamity physics So why all this rambling about whales and consistency? Because I just got back from vacation a few days ago and I’ve been surprised at how long it’s taken me to get my head back into the various projects I’ve been working on (or even writing this blog). And then I got to wondering how often something small like that snowballs into a stalled or even failed project. A missed day turns into a skipped week turns into a skipped month turns into a dead translation. Which then got me thinking about the coefficient of friction. It’s basic physics, which I excelled at (failing repeatedly). In layman’s terms, it’s a ratio (μ) that gives you a sense how much force two surfaces exert on each other and, therefore, how much force you need to exert to get something moving from a dead stop. Wooden block on ice? Low coefficient of friction. Wooden block on shag carpet? High coefficient of friction ... and a senseless crime against tasteful décor. Once you overcome that initial friction, it takes comparatively little force to keep an object in motion. I can easily imagine there’s a coefficient of friction between us and our work, some quantifiable level of resistance that needs be overcome before we get our asses in gear and be productive. And unlike the one in Physics 101, which is constant for any two materials, this one is different every single day. It depends on a bunch of different factors: how interested we are in our projects, how appreciated we feel, what other projects we’ve got going on at the same time, how much sleep we’ve gotten, what else is going on in our lives, whether or not the Mets are currently in the World Series, etc. Let’s call it the coefficient of slackitude. Once we get started on a project and make it part of our everyday routine, we can largely ignore this number. We’ve overcome the initial slackitude and, with moderate effort, can keep things rolling along fairly smoothly. But each time we let things coast to a stop, even for a few days, we’ve got to overcome the slackitude all over again. And since that value is variable, it might be much harder the second time around. In fact, it probably will be. Eventually, we’ll fail to do so. And our project will die. The takeaway So other than the fact that I had no business being anywhere near a physics classroom, what can we take away from my incoherent ramblings? A couple things: The easiest way to make sure your project gets finished is to stick to a regular schedule. Eat the whale a little at a time — every day if you can. Minimize the gaps. Avoid having to face off against that nasty coefficient of slackitude more than once.The easiest way to make sure your project gets started at all is to pick a time when that coefficient of slackitude is low — when you’re excited by the prospect, when you’re well-rested, when you have relatively few competing interests. When you can focus. Use that time to build your momentum, so when your interest wanes or real life intrudes — it always does and it always will — the project is so embedded in your routine that you can just ride it out.We need more finished translations in the world. So pull up a chair and eat your whale. Do it for your team. Do it for yourself. Do it for poor Little Busters EX, drugged and ball-gagged in a basement somewhere, forever wondering when it’ll finally get to see the puppies.