(edit: perhaps I should have posted this elsewhere, but it seems like the answer must be pretty simple)

Hi, I've been wondering about a situation I sometimes encounter, probably an overplay - black (4) in this diagram. However it might be fine of course. Could someone point out what they have found to be a good solid (and hopefully simple) response? Anyway this is the situation:

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . 4 a . |
$$ . . . . 2 1 . . |
$$ . . . . B 3 . . |
$$ . . . . . b . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

and it seems that white should then play a or b. On SL it's discussed a little bit here:

http://senseis.xmp.net/?33PointInvasionQuery1 Bill Spight suggested 'b' is generally better. But I did find those pages a bit confusing and the sequences don't go out as far as I'd like. The Joseki at the bottom of this page http://senseis.xmp.net/?33PointInvasionQuery1 only seems to work when there's a White stone outside for black to worry about.

blocking with 5 here leaves white vulnerable to black 6

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . 4 5 . |
$$ . . . . 2 1 . . |
$$ . . . . B 3 . . |
$$ . . . . . 6 . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

Which seems to get complicated.

and it also resembles the hane and inside hane variation given here.

http://senseis.xmp.net/?44Point33InvasionJoseki

It seems obvious that if White plays 1 black can cut at R15.

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . B . . |
$$ . . . . B W . . |
$$ . . . . B W . . |
$$ . . . . 1 . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

but say I choose option 'b' - avoiding getting sealed into the corner, is it fine to crawl along? Black can (I think) play 'a' 'b' or 'c' in this diagram. (though if 'a', white at 'b' is probably too good - depending on the surrounding situation).

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . B . . |
$$ . . . . B W a . |
$$ . . . . B W . . |
$$ . . . . b 1 . . |
$$ . . . . c . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

.. and how should White play now, balancing white pushing out along the end, pushing up off the third line and trying to avoid a black descent to S18.. Perhaps its just a messy situation that needs reading out?

In one of my current DGS games (not my most thoughtful game) I played

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . B 5 . |
$$ . . . . B W . . |
$$ . . . . B W . . |
$$ . . . . B W . . |
$$ . . . . 1 2 4 . |
$$ . . . . . 3 a . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

Which looks bad for me (White), I should probably have played 1 at 'a' or 2.. Looking at those three stones I guess there's a couple of proverbs somewhere telling me I should have extended to 2.

_________________
I am John. John-I-Am.

First, let's remember that the 3-3 invasion is not a balanced joseki. The defender gets a better result from it than the invader does.

Let's first examine the simplest case - just blocking black's last move.

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . X 1 . |
$$ . . . . X O . . |
$$ . . . . X O . . |
$$ . . . . . 2 . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

You said that this gets complicated. How?

Click Here To Show Diagram Code

[go]$$Wcm3
$$ ----------------
$$ . . . . . 7 . . |
$$ . . . 6 5 X O . |
$$ . . 8 . X O . . |
$$ . . . . X O . . |
$$ . . . . . X 1 . |
$$ . . . . . 2 3 . |
$$ . . . . . 4 . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

This is very similar to the typical 3-3 invasion, except white has better shape in the corner now than he did before. Among other things, the sequence below doesn't work for black anymore:

Click Here To Show Diagram Code

[go]$$Bc
$$ ----------------
$$ . . . . . . . . |
$$ . . . X O O 8 5 |
$$ . . X . X O . 6 |
$$ . . . . X O 9 4 |
$$ . . . . . X O . |
$$ . . . . . X O 3 |
$$ . . . . . X 1 7 |
$$ . 2 . . . . . . |
$$ . . . . . . . . |[/go]

Click Here To Show Diagram Code

[go]$$Wcm10
$$ ----------------
$$ . . . . . . . 3 |
$$ . . . X O O O X |
$$ . . X . X O 1 O |
$$ . . . . X O . O |
$$ . . . . . X O 2 |
$$ . . . . . X O X |
$$ . . . . . X X X |
$$ . 2 . . . . . . |
$$ . . . . . . . . |[/go]

As well, black can no longer kill with when he has the marked descent, like he can with the default shape:

Click Here To Show Diagram Code

[go]$$Bc
$$ ----------------
$$ . . . . . . . . |
$$ . . . X O O . . |
$$ . . X . X O . . |
$$ . . . . X O . . |
$$ . . . . . X O . |
$$ . . . . . X O . |
$$ . . . . . X X B |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

So black's weird hane in question is clearly garbage if all goes normally.

What if black plays differently?

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . X O . |
$$ . . . . X O 8 . |
$$ . . . . X O 6 . |
$$ . . . . 4 X 1 . |
$$ . . . . . 3 2 7 |
$$ . . . . . . 5 . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

This is almost the exact same as another joseki, except for the unfavorable for white marked exchange:

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . B W . |
$$ . . . . X O X . |
$$ . . . . X O X . |
$$ . . . . X X O . |
$$ . . . . . O . O |
$$ . . . . . . O . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

However, sometimes the diagram above, aside from the marked exchange, is better for white than the more normal way for black to play. This is true when black has a considerable investment in the outside area (hence why white entered the corner in the first place), and this joseki is inconsistent with maintaining that investment.

So now that we've considered the normal ways to play, what else is there?
Well, as you've said, the hane on the other side does look a little dubious white, and I think black will come out better from it.

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . X O . . |
$$ . . . . X O . . |
$$ . . . . 1 2 . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

There are other responses too. What about the simple extension? (Also possible: the one-space extension at 'b'.

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . X O . . |
$$ . . . . X O . . |
$$ . . . . . 1 . . |
$$ . . . . . b . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

Black can't play as below, as this leaves the bad aji of the subsequent cut:

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . 6 5 8 |
$$ . . . 9 7 X 3 4 |
$$ . . . . X O 2 . |
$$ . . . . X O . . |
$$ . . . . . 1 . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

Not that you'd have to play the cut right away, but it's always there haunting black, so it's no good. Instead, the proper move is here:

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . . . X 2 . |
$$ . . . . X O . . |
$$ . . . . X O . . |
$$ . . . . . 1 . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

However, you must keep in mind that this allows white sente to ruin black's outside potential.

Finally, what about pushing along for black like you mention? Well, play that out on a real board and then tell me why black would ever choose that fate as opposed to the normal variation.

So either white can live in the corner just as happily as before, or he can come out and ruin black's area. If black does play the hane underneath, I think he must have considerable outside strength in order to try to kill the white group. Thus, black must play as below. Anything else would be wrong.

Click Here To Show Diagram Code

[go]$$Bc B3 is the only move.
$$ ----------------
$$ . . . . . . . . |
$$ . . . . 3 1 2 . |
$$ . . . . X O . . |
$$ . . . . X O . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X X X X . |
$$ . . . . . . . . |
$$ . . . . , . . . |[/go]

However, in most cases, white has many answers to black's move, all of which seem to turn out just fine for him. If you just focus on the simple moves, you can very often find a satisfiable result when your opponent plays a questionable non-joseki move.

_________________
We don't know who we are; we don't know where we are.
Each of us woke up one moment and here we were in the darkness.
We're nameless things with no memory; no knowledge of what went before,
No understanding of what is now, no knowledge of what will be.

I think your is too slack in response to :

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . 9 . . |
$$ . . . 8 7 X O . |
$$ . . 0 . X O 3 . |
$$ . . . . X O 2 5 |
$$ . . . . . X 1 . |
$$ . . . . . 6 4 . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

This is not satisfactory for W. In fact, this is almost the same as a line of failure for W when B plays the hamete :

Click Here To Show Diagram Code

[go]$$Wc
$$ ----------------
$$ . . . . . . . . |
$$ . . . 8 7 9 . . |
$$ . . 0 . X O 3 . |
$$ . . . . X O 2 5 |
$$ . . . . . X 1 . |
$$ . . . . . 6 4 . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

You're right, I totally missed that. Interesting.

_________________
We don't know who we are; we don't know where we are.
Each of us woke up one moment and here we were in the darkness.
We're nameless things with no memory; no knowledge of what went before,
No understanding of what is now, no knowledge of what will be.

Ok, thanks - that's very helpful (I've even printed you replies out for reference). It seems like there's a fair amount of depth in the 3-3 invasion (I've just finished reading 38 basic josekis properly) - well depth for me anyway.

_________________
I am John. John-I-Am.

It's not just you, there are a lot of very interesting variations that can lead to different results. I love studying corner shapes in depth when I get the time to.