The only reason i can see why black would play this is to play d5 at some point. c5 is a possibility, but that would be a bit risky for the king.

Agreed. It seems that the best way for white to capitalize on these weaknesses, especial the dark-square ones, is to break open the center, and use the queen in counjuction with the DSB to penetrate deep into black's camp. My one reservation about attacking is that white has one knight and a rook stationed on the K-side. Also, white's LSB and c3 knight might have trouble entering the attack as well. Of course, they have a fantastic outpost on d5...but for the moment black has it sufficiently defended with his Nf6 and Bb7. However, in the course of the attack white could kick the Nf6 away from his critical defensive post with a timely g5!. I suppose white could start the attack by playing Ne2, then c3, and then finally d4 to open the center. Black doesn't seem to have many ideas here, so I think white has time to set up the afforementioned attack. I'm interested to know what other players (Ion and Marin, this means you...) think about the attacking idea.

16.g5 does look an attractive proposition here, with the idea 16...hxg5 17.fxg5 Nh5 18.Nd5 (with the incidental threat of 19.Nb6ch!).

f5 would have made the game very cluttered and more drawish, goth is right to trey and open things up a bit.

Agreed: 16.f5 Would have locked up the K-side for the nonce, giving Black more time to develop his own plans. It is true, though, that if Black wastes time, White could play h4 then g5, when things open up again... 16.g5, as earlier suggested, might have been preferable...

Seems the whole idea of playing Nb8 earlier was to throw the c pawn forward, was it really worth all that hassle?

this stops Nd4, and it's certainly the best offensive move at this time, wherever possible play moves that defend and attack.

Fortunately it's not as useful as the knight, but white owns the d5 square (it's covered by one, potentially two pawns after c4, whilst black can't threaten the square with any of his pawns.)This is describe as a 'weak square' for black, try to create these for yourself, and avoid giving them to your opponent.

f pawn's hanging again...getting rid of it sets up skewering threats against black's king and queen, and of course, it's always nice to be up a pawn at no disadvantage (there is no drawback, right?).

Don't think so...

This time, there's no loose knight on the f-file...

Once again, white fails to notice blacks error, she's too concentrating on the queenside and is not paying attention to the rest of the board.

And creating one on c3. however, it does stop Nc4 and also the threat of b3, so it's not too bad.

The weakness on c3 is much harder for Black to come at, even though he has a dark square bishop. This position won't be easy to bust open at all...

Isn't c4 playable? Blowing the pawn-chain open, eliminating the hanging pawn on c5, leaving white with a weak d3 pawn and freeing the dark square bishop...

Actually, it's not that bad, although i can see what tag's saying. Black can't knock it back with g6 because of the threat to h6, perhaps white was trying to stop Nd4. Other options for white were Nf3 and Qf2. g5 isn't that effective after 22. hxg5 23. Bxg5 f6.

I wonder if the rook move was intended to induce just this response? Black's bishop is a sorry looking piece right now... The rook hasn't lost much by this manoeuvre...

unfortunately if white takes, black will reign supreme on the c3 square, which is rather close to goth's backyard...

cxd4 is better, for the reasons tag mentioned...

Even 25...Rxd4 looks better than the e-pawn recapture... The open file after 25...cxd4 looks very handy, but is possibly too narrow a breach into White's position to promise decisive results. White would be well advised (after 25...cxd4) to manoeuvre the knight via f3 and d2 to c4. This will involve the preparatory move 26.Rf2 and the Queen later to d1, but she ought to have time for this.