Polarized light (Rival conventions)

[Galileo]

I'm really getting confused. I have to write an article about LCD's and I`m refreshing my memory about polarization.
The problem is with the definition of right-cricularly polarized light and left-circularly polarized light. I have two books one is 'Introduction to Optics'
by F. and L. Pedrotti and the other one is 'Optics of liquid crystal displays' by P. Yeh and C. Gu.
In the former book, when the light is approaching you and the E-vector is rotating counterclockwise, then the light is left-circularly polarized.
In the latter book it's the other way around.
I didn't know what to use, so I googled for it to see if there is a general
consensus about this and one of the two authors is simply mistaken.

Here are some sites which say counterclockwise rotation corresponds to left-circularly polarized:
http://electron9.phys.utk.edu/optics421/modules/m8/polarization.htm
http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/polclas.html

And here is a site which says otherwise:
http://www.enzim.hu/~szia/cddemo/edemo14.htm (nice applet!)

There are lots of other sites about polarization and mention that right/left-polarization exists, but do not tell which is which...

The (small) majority seems to be in favor of the convention that for left-circularly polarized light the E-vector should be rotating counterclockwise when looking at the wave against the direction of propagation.

Saddly, copies from the book my professor gave me which I have to used got that convention the other way around.
This is so frustrating! Why don't they call it clockwise or counterclockwise polarization? left and right is meaningless in this context. but even so, I don't care what they would call left or right as long as everyone uses the same convention! Did anyone of you ever encounter a similar case?

The difference in convention works through the whole calculation and theory of the Matrix method of polarization. The normalized Jones vector for left circularly polarized light in either case is:
\frac{1}{\sqrt{2}}{1 \choose i}
but the actual physical condition (state f polarization) is different.

What convention have you guys learned?

[gerben]

I think it should follow the “right-hand rule” (and “left hand rule”):
If you direct the thumb of your right hand in the direction of light propagation your fingers curl in the direction of E-vector rotation corresponding to right-circularly polarized light.
If you use your left hand the fingers curl in the direction of E-vector rotation corresponding to left-circularly polarized light.

Edit: I do not know this, I think this is how it should be, as it would make it seem reasonable to call them right-circularly and left-circularly polarized

[Galileo]

Gerben,

Yeah that's true. But the small majority got it opposite convention.
The right-left hand rule is indeed in favor for this convention, but the other
way around is also ok. If you want to turn your car left, you rotate the wheel counterclockwise, but again, it doesn't really matter what is used as long as everyone
uses the same.

[gerben]

Yes if everone uses the same convention there is no problem, but I meant that it would be easier to also use the same convention in different problems (even if you where the only existing person), so one should always use the right-left hand rule when talking about left or right rotations whatever the topic. I hope this is the case with these E-vectors (but I will see when someone really knows the answer...)