@MNb: Thanks, now I see what you mean. By the way, why do you think that 1.e4 e5 2.f4 ef4: 3.Nf3 d6 4.d4 g5 5.h4 g4 6.Ng1 is =+?
I have seen that statement by you several times on the forum, but haven't found concrete variations.
At least Gallagher in his article series for the german magazine "Schach" (I think it was around 2000) thought white was OK.
@Dragonslayer: Congratulations!
1.e4 e5 2.f4 ef4: 3.Nf3 g5 4.Nc3 Nc6 5.g3 Nd4!? is very creative.
I think it's hard for white to prove full compensation, e.g. 6.Bc4 (d3 Bd6!?) f3 7.d3 d6 8.Be3 Bg7 9.Qd2 Ne7 10.0-0-0 could be playable, but I'm not sure if white has full compensation.
Regarding the Kieseritsky, Stefan Buecker writes that after 1.e4 e5 2.f4 ef4: 3.Nf3 g5 4.h4 g4 5.Ne5 Nf6 6.Bc4 d5 7.ed5: Bd6 8.d4 0-0 9.Nc3 Nh5 you prefer 10.Ne4. But I can't see what to play after 10. ... f5!, e.g. 11.Ng5 Nd7! 12.Ne6 Qf6 13.Nf8: Kf8: 14.c3 Ng3! 15.Rg1 Ne5: 16.de5: Be5: and it is difficult to see which white piece takes part at the game at all.
I think the main issue in the position is the black pawn on f4. It severely hampers the white position and prevents the Bc1 (and consequently Ra1) coming into play. One move before black kept the pawn with Nh5.
So for me the most logical move is 10.Ne2 trying to eliminate that pawn, when Buecker's analysis at Chesscafe.com seems to indicate a good game for white.
Actually 10.Ne4 f5 11.Ng5 Nd7! is very good for Black (I discussed this with Bücker and tried everything: 11.Nxd6 cxd6 12.Nd3 seems best).
10.Ne2 has its problems too. Instead of Riemann's 10..Re8 Black has at least two better choices.
So it seems 9.Nc3 is under a cloud. However Bücker did a good job of putting 9.0-0 back on the playing field.