Back in '91 or '92 the hot new 2nd edition AD&D hardback was the Tome of Magic. As I recall it was loaded with spells and new character options like elemental wizards and such. It was sort of an Unearthed Arcana for spellcasters, though I didn't see it break campaigns the way UA did when it was introduced. One of the new character types the Tome added to the game was the Wild Mage. The best part of the class was the random Wild Surge table, which basically made the class into a ticking wand of wonder time bomb. Lots of stupid effects to mess with the party.
The problem was that no one I gamed with at the time wanted to play a Wild Mage. And as the guy who usually DM'ed, any Wild Mage I made would likely be killed by the PCs as fast as you can say "roll for initiative". It's a basic law of D&D that the PCs' desire to kill an NPC is directly proportional to the DM's love of the character. (I think that rule of thumb is called Elminister's Law.) So anyway I found no good way to get a Wild Mage into any of my games, but a strong desire to use that crazy table.
As far as I can remember, this was the first time I set down with the idea to systematically attack a problem in D&D. For the first ten years of my gaming career I relied almost entirely on poorly understood Dragon articles, best guesses, ad hoc solutions, and hard-earned lessons from sources such as my local con and my killer DM. But this time I really went at it, filling a whole notebook with different applications, amendments and extensions of Wild Surge effects.
Most of it was crap, but I eventually boiled the first draft down into an article I entitled "Get Wild!", which I submitted to Dragon for publication. I have no idea if I still own the manuscript whipped up on my mom's old typewriter over a three-day weekend. The main jist of the article was a set of alternative uses for the Wild Surge chart, mostly things like spellcasting fumbles, potion miscibility fiascoes, and other situations where the natural order of AD&D magic might go awry. If I remember correctly the article also contained a new Wild Surge chart, a percentage chart with 100 different surge effects. The Tome of Magic was the hot new hardback of the moment and my article was geared toward making the rules in it more widely applicable to campaigns, so I naively assumed that my article was going to be a slam dunk.
But instead of "Get Wild!" signaling my entry into the glamorous world of game design, the kindly folks at Dragon sent it back with a politely worded rejection letter. The manuscript was marked up in red where I made a few punctuation and grammatical errors. That same hand had also taken the time to run a red line through every use of the term "magic-user", with a brief remark that under the new edition "mage" was the correct term. And printed at the top of the article, again in red, was the simple note "Too wild!"
Given that "Too wild!" was the only clue I was given for the rejection of my piece, I was surprised as heck when a few years later they published Joel Roosa and Andrew Crossett's "Even Wilder Mages" in issue #202. That article covers some of the same ground as my "Get Wild!", though my article also addressed a lot of situations were a non-Wild Mage could end up rolling on the Wild Surge charts. But I must note that Roosa and Crossett's article was a lot more polished than mine.
Either way, sometimes I like to kid myself that back in '92 I was actually more cutting edge than Dragon could take.