My philosophy on openings, gambits and so on in general ...

a3 is definitely a better first move than a4

While these rook pawn moves can be exploited by Black, and a strong Black player will generally find positional advantage early, it's hardly the end of White's game. In general such moves are far more likely to give Black an opening advantage, and therefore could be seen as a sort of handicapping move. This is only in terms of theory, however. There are psychological benefits to moves like 1. a4.

First and foremost, your opponent knows it's a bad move, but likely hasn't studied good ways to refute this opening. In this case, the opponent may often simply play a favored opening without really understanding how to exploit the weakness of White's move. Furthermore, if White has a plan after such a move, and Black is overconfident by underestimating White's ability due to opening choice, they are in for a rough game. Lastly, moves like 1.a3, 1.a4, 1.g3, etc. generally do not have extensive amounts of theory to study, so the game is often out of book relatively quickly, which may be of benefit to some players.

So maybe not every opening that has a name is sound. Maybe quite a number of gambits have been refuted. I think not playing them or against them from time to time does a disservice to one's chess study, however, because they give rise to positions that may be rare, but instructive...I mean, hell, given the extremely high frequency of blitz and bullet games on this site, I'd hazard to guess that quite a significant number of the tactics puzzles and captchas may even have come out of such silly openings. ;)

Also, I agree with Lightsss. It's objectively better. 1. a3 is less weakening, gives the king luft, stops infiltrating moves on the kingside from the enemy bishop or knight, and is a move that likely is going to be played at some point by White in a large majority of openings anyway for all those reasons. And if the White player prefers playing Black because they don't handle the initiative well, then it's a pretty solid way to pass on the first move and let Black lead the dance.

I like the guy who said 1.a4 isn't bad but he obviously doesn't play it. I like those wacky openings, but then again I'm a mediocre chess player. I've had 1.h4 work once or twice when Black castled kingside anyway after my opening since he's used to it.

@static_shadow
although lightss may be right in that 1. a3 is stronger than 1. a4, a4 is definitely played more often by GM's than a3 in any given game. I suppose 1...e5 could be met with 1...c4 to make a3 useful? Allowing the center to open up is pretty sketchy though, and I would guess an engine would condemn such a move. If this did work, then d5 openings could slightly benefit from 1. a3. I'm gonna hold that your best bet is a reverse pirc-- and in that case I'd probably rather play nbd2 e4 etc. and get myself a reverse hanham with a4 anyway.

"If it has a name, it's more worthy of play than if it doesn't."

If that was the idea, then I might be inclined to agree....but who knows what names will be made in the future for new openings?

Or is the opening knowledge in chess pretty much complete? I doubt that it is.

Let me erase my earlier statement and say that the best opening ever in chess...is the Barnes Opening. Hey, its got a name.

Hey, it works against really low-rated players! :O

Makes them so mad to lose against it, too! :)