And here was I, thinking Black was going for an all-out assault on the K-side!  

What about Nf5, with the idea of g4 the next move? If white plays e4, Nf-c4 will do.

Starting from the fact that B's "bad" B is g7's, not the e6's one! It's bad because it is blocked by its own pawns. This doesn't apply to this very position, where the endgame is still potentially far, and e and f black pawns are not blocked. In fact, this move allows black's e4. If 14. Nd4, NxN. And if 14. dxe4, exe4, 15. Nd4? NxN, so white will be forced back to Ne1. Finally, if 14. Ne1 instead, black is advised not to take exd3, 15. Nxd3. But black will go on with his attack on the Kside.

Throughout the game, I was considering e4-e5 and/or g5-g4. Maybe I should have tried it, although the game works out OK.

Here I would consder Qb3, with a tricky threat on b7. Black can't just play Na5, as usual, since Bd2 will exchange it, then Qxb7 forces Rc8, and white has a strong attack. But black has many other answers... like Rb8 or Qd7.

As white, I would have tried to get something more from the queenside that a flying R.

I still "see" e4 for black. "When the advance of a defended pawn implies a threat to a minor piece which can't take the pawn back, it is alway a potential break, and the consecuences must be kept in mind while the rest of the game goes on". In other words: white lost one piece controlling e4, and can occupy it, reinforcing it with f5...

I'm thinking ...h5, but I have a feeling Black really had to play 16...a6 to slow White down a bit. White's position looks fairly tame right now, but it is the sort that could explode into action very quickly.

Black's plans can go throug 16... Bh3, (say 17. b5), BxB. Now white can take the N, 18. bxNc6, which gives black a tempo to take BxRf1, or go otherwise: 18. KxB, and the resulting black tempo can be used for e4.

Hehe, I won't even have considered it. Queenside is the only white active zone. I won't castle into it, hoping to attack. As black, I must not delay my own attack a single move, and a R on f8 is not my first need, so I won't castle either. Just go on with your attack.

What about 17... g4, forcing Ne1...? Black has h5-h4 all the same but white will be more cramped.

17....g4 is probably good too. I am not convinced Ne1 is forced, he could play Nh4 and allow his king-side pawns to be weakened. While I am trying to win his weak pawn on h4 he may break through on the queen-side.

It doesn't look as a defensive move at all, since it allows the opening of the g column.

White's main threat, ... or Black's? Black has two fell threats - the Queen manoeuvre to mate if White doesn't play 20.Bxh3, and ...Nxh4 and ...Rg8ch if he does (after 20.Bxh3 Qxh3). But White simply has to take that bishop.

Yes I meant black sorry.

I stopped playing Reti openings in over the board play because of situations similar to this. I find that if I am not constantly keeping a threat on my opponent, then I am stuck to trying to defend my king. Not that it's not a good opening, just so many weaknesses to overcome in my opinion.

Quick collapse, which can happen all too easily in this type of game. But Charles - you might want to look at your header details. An uncharitable mind might suppose you lost this game...  

Many thanks for your comments everyone!