This is an allegory: you represent primitive man, moving through the various stages of evolution as represented by levels in a strange stone edifice that appears before you suddenly. The puzzles represent problems along the way of evolution, problems whose mastery defined certain levels of development--they're mostly straightforward (though one is a bit confusing), but there is a real sense of accomplishment in solving them, somehow. Certain stages are by definition mind-numbingly tedious, which reflects the subject accurately--these are problems that involve tedium--but also raise the question of whether there might have been a better way to design those particular puzzles. (For a similar problem, see The Meteor, the Stone, and a Long Glass of Sherbet.) It should be added that the author isn't trying to convey every single aspect of every phase of evolution; rather, you represent an important breakthrough at each stage, and when you're done, you move on and reenter the scene much later, when Homo sapiens has incorporated your discovery and built on it.
That raises the question that I, at least, found most intriguing about this: does this really have anything to do with evolution? It's kind of a silly approximation, after all, since it's apparent quite soon in each stage what the sought-for breakthrough is, and it's just a question of putting together the needed materials or figuring out the key steps. But it could be argued that Smith has designed this with the feeling of discovery in mind: particularly in the last two scenes, you have the sense of a specific need that drives the breakthrough, not a sudden resolve out of the blue to carve hand tools or domesticate animals. The sense of logical connection is less strong in the first one; there is very little sense that you tumble to your discovery because of circumstances, rather than having a twentieth-century computer user push him around to accomplish a certain goal. Perhaps that's inevitable, given the problem at hand, but I would have liked to have seen at least some sort of conjecture as to what sort of circumstances lit up that particular connection for Stone Age man. The game does capture the brutality of this particular discovery well, in that you have no particular reason other than your own satisfaction for doing what you do, and perhaps the apparent purposeless of your solution to the problem reflects the arbitrary kill-or-be-killed nature of the environment--but it still felt a little unsatisfying.
On the whole, I found the second stage most plausible and interesting; it's the only one where you deal with other characters, and though your interactions are limited, the characters have a certain charm. (I found a certain whimsical charm in their names--Wife, Son, Grandmother.) The central puzzle took real thought and felt genuinely rewarding to solve--and, even as a microcosm, it felt, more than any of the other problems, like what really might have happened. The first stage, as suggested, is a little too illogical to really feel like an account of the breakthrough, and the third just doesn't quite make enough sense; you have the sense of the original motivation for your character, but not what inspired him to try this particular approach. (And the realism/tedium element that worked reasonably well in the first part is simply annoying here, because it doesn't feel particularly logical.) Still, all three are worthy concepts, mostly well thought out.
The writing is nothing special, though arguably that makes sense here--too much attention to the scenery or aesthetics would distract from the goals at hand; you're not in the situations in question to check out the sights. Certainly, the writing is adequate for the purposes; it sets the scene and makes clear what you need to do. My main problem with the mechanics of Edifice is that it's possible in the first part to screw up and make the game unfinishable or virtually so, which, quite apart from its problems for the IF player, doesn't really make much sense in the plot of the game. (The nature of the problems was such that the solutions developed over time, after all; it took many failures to make the discoveries required in each scene.) In several key respects, the first scene requires resources that can be easily wasted--and though the urgency of the situation lends a certain logic to the picture (if you don't solve the problem, you'll die), it still doesn't really make much sense as an evolutionary tableau. Another clunky element is the hint system, which rests in a mural in the edifice--clever enough, but the problem is that the mural only gives you the hint once you've gone out and actually done each step and come back and checked the mural, which I found time-consuming and annoying. You can't, in short, play to where you get stuck, then come back and check the mural, because the mural won't keep up with you. Among problems to be fixed for future versions, this one may be #1.
The end is a bit confusing. There is a cataclysmic event at the end that doesn't seem to fit into the evolutionary frame, as far as I can tell, and my guess was that it's the author's device for ending the story. If so, it's intriguing--but the final sentence, even when you've finished the game "right" (in accordance with the walkthrough, anyway), is a bit of a puzzler. I couldn't decide whether it was a comment on the nature of evolution or simply a bug; if it was supposed to be the former, it could perhaps have been more skillfully done. (At least, it might have a sentence somehow distinguishing it from less satisfactory endings.) And while I enjoyed the ending--it had plenty of drama--it did weaken the allegory a bit; certainly, the evolutionary process didn't end where it does in this game, but the conclusion of the game suggests some sort of ending.
Quibbles aside, though, The Edifice is one of the most intriguing games in the competition, in that it tries something completely new--a first-person account of the highlights of a scientific process. In a sense, your actions are vital to the continuation of that process--if you choose not to go on, progress stops with you--and one way to see your role in the game is that of a guiding force at crucial moments in history, an intervention to ensure that the development of man stays on track. That, at least, would explain some of the knottier problems involving questionable motivations or the difficulty of anticipating a particular result without the player's advance knowledge. Part of the charm of Edifice is that the story it tells is sufficiently ambiguous that it can justify a variety of perspectives, including those who don't care for evolution as a self-perpetuating force at all. (The game, whether deliberately or not, provides some grist for the mill of the argument that mere chance could not been sufficient to turn ape into man.)
Though elements of the gameplay are lacking somewhat, anything that Edifice lacks in playability is easily made up in sheer concept; the idea and the charm of its implementation earn this one an 8 on the competition scale.