French Defense is officially SCREWED!!!

  OK, first of all, the French is not totally done for. I just wanted to say that. However, I jound a new, completely unexpected variation.Ok, here Black has three main Variations he can choose from. Lets go with the one I call the Classical Variation.

The second one we can look at is also the most complicated. Lets call this the Mackenzie Variation after Dana Mackenzie who first coined the qh4 move

Against 5...qb6 6. qc2 is also possible. The first person to give deep analysis and leave a desired name it will be named that after them. Thanks!

As you demonstrated, Qh4 by black is no good. White plays Qc2 and Nf3 and chases the queen around. Thus, even if white doesn't win the queen, black has wasted time and accomplished nothing, while has developed pieces with tempo.

So your first game at the top is the only one to even consider. As black, I wouldn't play 5. ... Nc6 there. I'd probably play 5. ... Qb6, threatening f2 immediately. Since you like Qc2 to protect, let's assume 6. Qc2, then I might capture the pawn with 6. ... dxc4 7. Bxc4. Now maybe Bd7-b5 to try and trade black's bad bishop for white's good one. That's how I normally play against the Advance variation in the French, and I see no reason to change that just because white helped me develop my king's bishop by taking on c5 prematurely.

Actually, now that I think about it more, what if black just plays 5. ... d4, creating a passed pawn right from the opening? It can be protected by Nc6, and it's already got the queen and bishop behind it. That might be over-extending, though, since the pawn can't be defended by another pawn, so white might be able to find ways to attack it.

I wouldn't immediatly take the pawn on c5 after all after dxc5 The e- pawn has lost an important defender so maybe Qc7 or Nc6 or Ne7

After black plays Nf6 I don't see how you can claim white is at all better. I'd say black is definitely better with his much better development. 

black never played nf6

Very well, here's my analysis:

4. dxc5 is a harmless continuation after which black has no notable problems. Note that the position after 4. dxc5 is often reached with white to move (after the moves 1. e4 c6 2. d4 d5 3. e5 c5 4. dxc5 e6), and even in that variation white usually has problems getting an edge out of the opening.

In this case after 4. dxc5 Bxc5 black has very easy development, while white has some worries protecting his e5-pawn. It might be useful to compare this opening variation to the mainline steinitz french, which goes 1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Nfd7 5. f4 c5, and now, sooner or later white will take on c5 or black will take on d4 (although sometimes black prefers to push the pawn to c4 instead), when we arrive at the same pawn structure. In the Steinitz french however, White has a much better set-up in the center here: the d4-square is well under control, the e5-pawn is well protected and white has very easy development with Nf3, Be3, Qd2, Be2, castling short or long, etc. Meanwhile black's position is rather cramped, though he can work up his own play (mainly on the queenside with ..a6 and ..b5).

Going back to your variation. A set-up with f4 and Be3 is already pretty much impossible. This means you have a hard time keeping the blockading squares (d4 and e5) under control. This means black has an easy game.

5. c4?! really makes no sense to me at all. I mean, what would an exchange on d5 achieve? You want to free black's c8-bishop and at the same time give him a passed pawn? Most importantly, it does not help your development at all.

In your first diagram you give 5. c4 Nc6 6. Nc3 Nxe5 (note that black can force white to play a dubious gambit with 6. ..d4 7. Ne4 Nxe5! 8. Nxc5 Qa5+) 7. cxd5 exd5 8. Nxd5 "with an interesting game to follow".

Well, combining the facts that black is ahead in development (1 extra piece developed + he is to move) and the center is wide open, I'd say black is already significantly better (maybe even already winning). I doubt you'll find that interesting for white.

Let's say, for example, that black continues 8. ..Nf6. The idea is very simple: the Nd5 will be chased away (or exchanged) after which black can play Qb6 (winning time by attacking f2), castle, maybe play Bg4 in the meantime and place his rooks on d8 and e8. And if white plays Bb5+, black can even just play ..Kf8! so that after he moves the c8-bishop out, he can immediately follow up with Re8.

Here are some concrete variations after 8. ..Nf6:
9. Nxf6+ Qxf6 10. Nf3 (10. Bb5+ Kf8! will be followed by developing the c8-bishop and Re8) O-O (Nxf3+ is good too) 11. Be2 Nxf3+ 12. Bxf3 Re8+ and white is one step too late: his king is caught in the middle.

9. Bb5+ Kf8! 10. Nc3 Qb6 11. Qc2 a6! and white's bishop must go back to f1 to avoid losing material (eg. 12. Ba4? Bxf2+! 13. Qxf2 Nd3+)
Note that there are lots of transpositions with the moves of Nc3, Nxf6+ and Bb5+, because black's ideas are all the same.

Maybe there are some improvements earlier on, eg. 8. Bb5+ or 8. Qxd5, but even then I'd say black is definitely better. The main things is that white is opening up the game while he is behind in development. So I suspect 5. c4?! is the real culprit.

@OP

The main problem is that none of your lines are forced. Black can do all sorts of other, better moves than the ones you suggest.

There is no point analysing lines which do not feature the best play from your opponent or are not forced.

I think Black should just play 4. ... Nc6 hitting the e-pawn. If White tries to prop it up (say with f4) then Bxc5 which then protects the square d4, should he need to push a pawn to that square.

T.D.Harding (1979) said that 4. dxc5 goes back to Steinitz and calls 4...Bxc5 "dubious." The line goes 4. dxc5 Nc6 5. Nf3 Bxc5 6. Bd3 Nge7 (though f6 is a major alternative) 7. Bf4! Qb6 8. 0-0 Ng6! 9. Bg3 Qxb2 10. Nbd2 where both 10...0-0 and 10...Ngxe5 lead to unclear play.