The final 5%

March 15, 2012 by joningold 2 Comments

If you watched the remake of Battlestar Galactica, you’ll know that after two or three years of escaping murderous robots with LED eyes and their sleazy-nightclub-owner-type owners, the last surviving humans were faced with the terrible threat of the Final Five. Five last Cylons who could yet destroy everything. Hard to pin down, hard to defeat, hard to negotiate with…

It’s something that anyone who’s worked on long projects can sympathise with. Projects can be easy or hard, but every project ends with that final 5%: the final 5% that nearly kills you.

I’ll be honest, this post is really about Frankenstein, inkle‘s first interactive novel app, which is out on April 26th. This is a project I’ve been working on full-time since December and designing, in a way, since I wrote My Angel back in 2000. Frankenstein is high-quality interactive storytelling built for a mainstream audience. It’s beautiful visuals combined seamlessly with great writing and a smooth, slick, gets-gone interface. It’s a lovely piece of work.

Almost. Really, very, almost. There are just a few last touches. Just that final 5% to get perfect.

Part of the problem with that final 5% is the injustice. It shouldn’t be hard to finish a project. After all, by the time it’s 95% complete, you finally know exactly what your project is. You know how every part will come together. There are no unanswered questions, no unexplored corners, no hidden problems. The clarity of vision can be dizzying after months, or years, of working on a project where the end is just a distant dream.

I remember, as I came to the end of Make It Good, the unreality quality of it. It was inconceivable that my to-do list on that game had shrunk from a 2,000 line text file to a set of three points, one of which was a bit of pharmaceutical research that – you know what? – I never did. After ten years of work, all the individual pieces were there, and not only that, they fitted together.

(And then the bug reports came in. That was Make It Good‘s final five – all the testers who did things ever-so-slightly off the order I’d thought of, and brought the complex, intricate logic crashing down. The bugs are gone now, I think: I’ve not had a report for Make It Good in two years beyond a few typos, although I get letters from players most months. But there was a moment there when I thought, this is it, it can never work…)

It’s particularly special, here at inkle, because we’re not just in the final 5% for Frankenstein, but also for inklewriter, our web-based writing tool which just needs a little bit more polish, and for an as-yet-unannounced HTML 5 project which is almost wrapped up.

It’s enough to make me want to start something else long and complicated, just so I can remember what it’s like to be free to make things up, unsure if they’ll work, or be worth doing at all. (Actually, we’ve got that covered as well, now that I look at my notepad).

But there’s no avoiding it. Starting something else is just prolonging the problem.

The trouble is, that final 5% looks like a small amount of work, and it actually is, but it’s exhausting and takes all the skill you have to get right because, being the last thing you do, you don’t get as much chance to revise as you did with the rest of the project. There’s no new information coming around the corner: no revelatory moments where you gain sudden new perspectives. The final 5% faces you up against the purest challenge. Here is the project, here is the problem, and either you can do it – or you can’t.

A lot of projects don’t make it through. The final 5% can be the thing that kills a good idea dead and leaves the audience bemused, underwhelmed, or throwing their pads against the wall. For games, it tends to be bad balancing, or a puzzle early on that’s too dull or too hard. (Heavenly Sword‘s cannonball sequence. The dodgy aiming in Uncharted 1. Hell, even Ico had that bit with the bomb and chandelier and the pillar.)

For novels and movies, it’s that missing “wow” factor, that ropey ending, that over-long edit. The final 5% is the polish and tightness that ties an okay story together into a fantastic one. (And it is only 5% – compare Blade Runner and the Director’s Cut.)

For TV shows like Battlestar, there’s a particular final 5% that rears its head when the series is running out its budget. (And Battlestar jumped the shark on its final five so determinedly it’s almost like the writers simply drew lots to see just how badly they could wrap things up. The Lost guys did the same a year or so later.)

For software like inklewriter, it’s what happens when you press the Escape key, or the up arrow, or hit Tab. It’s the two or three places where it runs just a bit slowly.

For IF it’s the place where the user types HYPERVENTILATE and the game understands.

For Frankenstein, it’s the transitions – you’ll know what I mean when you read it.

Of course, the only way to get through the final 5% is the same way you got through the rest of the project. You make a list, and you do the list. You put one foot in front of the other and keep going. You make the best decisions you can and test out what you’ve got. And you keep faith because, in (time of project so far / 19) more days, it will be done and it will be awesome…

And when that happens, your reward will be to start something new. And it’s not like starting stuff isn’t really difficult too.