Scribes Versus Authors

Volume #1 Issue 8 June 1988

At first glance, scribes and authors appear to be similar. They are both writers; each one sits at a table and writes things on paper. But there is a profound difference between the two. A scribe puts words down on paper; an author puts ideas down on paper. The scribe’s energies are directed towards the technology with which he works. The scribe worries about the craft of writing: the pen strokes that form the letters, the arrangement of words on the page, and so forth. The author lives in the world of ideas. The author creates, develops, organizes, expresses, and writes ideas. The author worries little about the quality of his scribery. Mark Twain’s penmanship is almost unreadable, yet he is generally regarded as the greatest American author. Thus, while both workers produce writing as the concrete result of their efforts, they concentrate on very different things.

I use the terms "scribe" and "author" as ideals rather than realities, as poles on a spectrum rather than discrete categories into which I place people. Throughout history, most writers were some mixture of the two. There was a statistical distribution of writers, lopsided toward the scribe end of the scale.

A similar situation exists with programmers. The population of programmers occupies a spectrum with scribery at one extreme and authorship at the other extreme. As with writers, this distribution is lopsided towards the scribery end of the scale. In other words, most programmers are more concerned with the technology than with the ideas. They take greater pride in memorizing technical aracana than expressing profound ideas.

In the early days of the computer games industry, the scribes reigned supreme. The first successful games platform, the Atari 2600 VCS, was a weak machine needing a strong programmer. With 128 bytes of RAM and 2K bytes of ROM, there wasn’t much resource for programming. Worse, the display was written to the screen by software rather than hardware, meaning that the program had to change graphics register in real time, just in front of the electron beam scanning across the television screen. Thus, impressive graphics displays required program timings that were excruciatingly tight. The programmers who could make the VCS display interesting things were a special breed, intensely involved in the technology, scribes through and through. The tight-ness of the programming environment left little room for authorship. All the brilliant ideas in the world were of no value in an environment that afforded little opportunity to express those ideas. Authors held no status in the games community of 1980; scribes dominated.

All that has changed in just eight years. The most lucrative target machine is now an IBM PC clone with 512K of RAM, a 16-bit 8088 running at 5 MHz, and 330K of disk capacity. This machine is about two orders of magnitude more powerful than the Atari 2600.

The greater power of this environment has changed the priorities of the community. No longer is it necessary to sweat bricks just to get a decent display on the screen; it is now almost easy to get good graphics up. In fact, the days when designers would gasp at the graphics stunts created by their competitors are over. Any graphic trick produced by a given designer can probably be replicated by almost any other competent designer. The phrase "graphics guru" is seldom heard these days.

Similarly, the emphasis on tightness of coding has almost disappeared. I still remember the great pride I felt over the tightness of the coding of Eastern Front (1941). Here was an entire wargame that fit into 16K of space! Perhaps my proudest achievement was the code that changed the color of the trees with the passage of the seasons, from bright green in spring to green in summer to red in fall and brown in winter. The entire feature, code plus data, cost me sixteen bytes of space. Those were the days.

Today, however, I just don’t give a damn about how much space I use. My last project, Trust & Betrayal, marked a major milestone in my career: it was the first game in which I failed to use all the RAM available to me. I targeted the game for 512K of RAM, and when it was all done, it only consumed about 380K. In my current project, I treat RAM the way Genghis Khan treated prisoners. Instead of painstakingly writing one subroutine to cleverly handle 25 different but related situations, I just set up a huge CASE-statement with 25 instances. Why bother?

The same logic applies to execution time. I once knew the timing requirements of every instruction in the 6502 instruction set by heart. I knew all sorts of great tricks for shuffling stuff around the index registers to obtain extra speed. Nowadays, I have one simple trick that always works: if it runs too slowly, precompute the results and stuff them into a huge table. Why bother?

In just eight years I have seen my scribish skills lose their power. Once I could impress the ignorant natives with a few simple tricks, a flashlight or a box of matches. Now the damn natives have VCRs and ghetto blasters; my tricks don’t get me very far.

We’ve seen the effects of this change in the social makeup of our industry. Where are all the hot scribes of yesteryear? Whatever happened to Bob Bishop and Rick Maurer, Carol Shaw, Tod Frye, and so many others? The answer, sadly, is that these great scribes have been abandoned by the computer games industry (although they would say that it was they who abandoned it). However you see it, their skills are no longer needed by the industry, and they and the industry have parted ways. In the span of a decade, the scribe has gone from nerd to wizard to washed-up old tramp.

Authorial skills are now the talent that this industry prizes. The ability to create, develop, organize, and express a fun idea through the medium of the computer is rapidly becoming the dominant factor in the eyes of many publishers who must choose between competing proposals.

The process is not yet complete. We continue to see games hit the market that are clearly the work of scribes. More embarrassing, some of these games become great commercial successes! But the advances in hardware that have so dramatically changed the situation in just eight years are still at work. An IBM PC clone with 512K of RAM is not the last word in home computers. In eight more years, we’ll be working with 32-bit machines with megabytes of RAM. Talent in scribery will be even less important.

I recently met an aspiring young game designer bursting with talent and possessed of that lean and hungry look so indicative of future success. This woman has no training in computer programming; indeed, she has never taken a single college course in technical, scientific, or mathematical subjects. Just a few years ago I would have gently steered her away from computer games. But the times have changed. She has learned to program in HyperTalk. Sure, we all know that her HyperTalk scripts won’t be as fast as our high-performance C code. But if our hot stuff takes 1 millisecond to execute, and hers takes ten milliseconds, what difference will the user perceive? And if our elegant subroutine fits inside 1K of RAM, and hers needs 10K, what difference will that make on a one-megabyte machine? With each passing month, the machines get bigger and faster, and HyperTalk gets better, and my untrained, techno-wimp young friend grows more powerful.

Game designers, take heed. The times, they are a-changing. If you are a scribe, you’d better start changing, too.