OK, I've taken a lot of time reading up on these two subjects of codominance and incomplete dominance and there is a discernible difference between the two that are hardly interchangeable if you understand the concepts. Unfortunately, these two traits can be easily misunderstood which is why some people (and literature) sometimes equate them into being the same thing, they are not.
Caveat: This is all based on stuff I've read in the few books I own as well as online resources. I am not stating that what I'm posting here is definitively correct but it is how I understand these topics based on what I've read and interpreted. I'm sure there will be contradicting articles elsewhere and if anybody can add to this discussion and cite those sources, it would of great interest and help to the rest of us. 
Let us first examine the primary Mendelian traits, Dominant, Codominant and Recessive as well as the Incomplete Dominant trait which was not discovered by Mendel.
Dominance: A Dominant trait is one that will be expressed over another trait 100% of the time. In Dominant traits there are only 2 possible phenotypes that are expressed, the Dominant trait and the Recessive trait. For example, a Dominant specimen (example: typical wild caught colored specimen) when bred to a Recessive specimen (example: Albino mutation) will produce 100% Normal looking specimens every single time. If the Normal specimen is not het for any genes of any Recessive trait it is bred to then all of the offspring will breed true to be Normal colored 100% of the time. Naturally they will now carry the genes for any Recessive trait they have been bred into which now makes them Heterozygous (Het for short) but that trait will not be visible in any way.
Kevin stated this: "A trait that when crossed with another trait, produces 50% offspring looking identical to it and the other offspring looking like the other parent.
The biggest problem I have with this statement is that we can prove it's false in regards to breeding a pure Dominant to a pure Recessive with the F1 generation. However, this statement might work when breeding two different Dominant traits together or breeding a Dominant trait with a Codominant or Incomplete Dominant trait. This is now where we investigate Codominant and Incomplete Dominant traits.
Codominance: A Codominant trait is one where neither trait is dominant over the other but both are expressed visibly together. The most common example I have found for this is in regards to breeding a red petaled flower with a white petaled flower; the offspring flowers from this pairing will have both red and white petals. To clarify, neither color will be mixed with the other but both will be expressed independently. In Codominant traits there are 3 possible phenotypes that are expressed, the 1st parent's trait (example: red petals), the 2nd parent's trait (example: white petals) and the combined parent's trait (example: red and white petals). According to some literature, combining Codominant traits results in 100% combined trait offspring. In other words, if we breed our 100% red petal flower to a 100% white petal flower, 100% of the offspring (F1 generation) should have both red and white petals. When the F1 generation is then bred back to each other (inbred), the resulting offspring should then match the Punnett Square ratio for that of Het to Het pairings: 50% red & white petals, 25% red petals and 25% white petals.
Incomplete Dominance: An Incomplete Dominant trait is one where neither trait is dominant over the other but both are expressed combined as a third phenotype. Again, the most common example I have found for this is in regards to breeding a red petaled flower with a white petaled flower resulting in the offspring flowers to have pink colored petals, a result of mixing the color red with the color white. As with the Codominant trait, there are 3 possible phenotypes that are expressed, the 1st parent's trait (example: red petals), the 2nd parent's trait (example: white petals) and the combined parent's trait (example: pink petals). According to the same literature regarding Comdominance, combining Incomplete Dominant traits results in 100% combined trait offspring. In other words, if we breed our 100% red petal flower to a 100% white petal flower, 100% of the offspring (F1 generation) should have pink petals. When the F1 generation is then bred back to each other (inbred), the resulting offspring should then match the Punnett Square ratio for that of Het to Het pairings: 50% pink petals, 25% red petals and 25% white petals.
Recessive: A Recessive trait is one that is only expressed when both alleles for the same trait are present. In Recessive traits there are only 2 possible phenotypes that are expressed when mixed with a Dominant trait. When combined with other traits however, Recessive traits can still be Dominant, Codominant or Incomplete Dominant compared to those other traits.
My favorite example of this when we breed an Albino to an Axanthic. When the phenotypes are expressed, a visible Albino and a visible Axanthic, they are currently considered Dominant traits as they are currently visible. When we breed them together, the resulting offspring are now Normal colored because neither Albino or Axanthic carried the full set of genes to produce another of it's phenotype, so the Normal dominant genes came back into play. However, all the offspring are carrying the genes for both traits which now makes them 100% Het for both Albino and Axanthic that we call "Double Het Snows". When we inbreed the "Double Het Snows" together, we get a whole lot of different genetic pairings (as covered extensively here. The most interesting outcome of this though is that of a "Snow" which is the combined coloration of an Albino and Axanthic which is the exact definition of what Incomplete Dominance is, a combined trait that takes no precedence over the other one but that combines both traits together to produce a third phenotype. The paradox here is that these are Recessive traits exhibiting Incomplete Dominance definitions which is why we must understand that our definition of these traits should be defined on how they work when compared to other traits. In other words, what might be Dominant for trait Y against trait X might prove X to be Recessive against trait Z. Z (Dominant over Y & X) > Y (Dominant to X but Recessive to Z> > X (Recessive to both Z & Y). I am currently unaware if there is such a hierarchy of traits that might fall in this order, but I don't see why it's not possible.
Of course this leads us to the Anacondas and what definition they should fall under. Obviously Anacondas are different than per se an Albino or Axanthic as Anacondas are a pattern based phenotype whereas Albinos and Axanthics are color based phenotypes. Technically there should be no crossover between the two but this is not entirely the case either as we will discuss shortly.
So, what is an Ananconda, is it a Codominant trait or an Incomplete Dominant trait? I guess this is highly subjective depending on how you want to view them. Let's recap a little, a Codominant trait is one where both traits are expressed together but independently to produce a third phenotype whereas an Incomplete Dominant is a trait where both traits are expressed as a combination between the two to produce a third phenotype.
We know that the "Super" form of an Anaconda is a patternless hognose (Superconda) and that the Anaconda is basically the Het form of a Superconda when bred to a Normal. With this evidence, we know that there are 3 phenotypes for the combination of these genes: Normal, Anaconda and Superconda. Therefore, the Anaconda gene cannot be Dominant nor Recessive as there are only 2 possible phenotype outcomes for those traits. So the Anaconda gene must be either a Codominant or Incomplete Dominant trait.
The typical look of a Normal (colored) Anaconda is a reduced dorsal pattern along with a dominantly solid inner black ventral with an unbroken white or beige outside ventral line on both sides. So, how do we define the reduced dorsal pattern, does it qualify as a Codominant or an Incomplete Dominant trait? Both traits exhibit a combination of two dominant traits, in this case, a patterned animal mixed with a patternless animal. If an Incomplete Dominant trait is the combined mixture of two traits (red flower x white flower = pink flower) then I would have to assume that for an Anaconda to be an Incomplete Dominant trait that the patterns would not be reduced but severely faded, semi-transparent as it were. They typically are not. I presume that the Anaconda trait is a Codominant (red flower x white flower = red & white flower) one because Anacondas typically are solid albeit reduced patterned animals. The reduction in pattern is merely the dominant part of the patternless trait showing itself while the remaining solid pattern is showing itself as well. Of course then we have the anomalous low key Anacondas which I think really cements that this is not an Incomplete Dominant trait as there is no mixture of the two genes apparent in the phenotype of the dorsal pattern. The only way to evidence whether or not an animal is a low key Anaconda or not is by it's ventral patterning. This I think is also part of the Codominant trait showing itself rather than an Incomplete Dominant trait.
Having talked about the defining ventrals in a Normal Anaconda does raise some interesting questions when you venture into breeding Recessive traits into the Anaconda line such as Albinos, Axanthics and Pink Pastels as Brent has done. I had presumed that the ventral patterns on Anaconda Albinos, Axanthics and Pink Pastels would have had a majorally solid colored center with the white unbroken outside ventral lines. Brent has informed me that this is not the case with his animals, that they are checkered bellied as are typical "normal" Albinos, Axanthics and Pink Pastels. How does this happen and why? If the solid and white lined ventrals are a dominant feature in Normal colored Anacondas, why does it disappear in colored mutation animals? Does it have anything to do with the color mutations being Recessive or is the checkered belly in these Recessive trait colored animals actually a Dominant trait when linked with the Recessive traits? There is a whole different topic of how multiple genes can be linked to each other in a way that they cannot split apart easily or at all. Perhaps this is what is happening with the colored Anacondas versus the normal Anacondas, that the gene linkage for the colored's checkered ventral is stronger (Dominant?) than that of the normal Anaconda solid ventral. It's definitely an interesting occurence and hopefully one that we will one day understand after enough of these animals are produced and examined.
I apologize for such a lengthy and in depth post. I'm sure I made some errors along the way but I spent the better part of my day not only writing this, but researching as well. Here are some of the links that I read about this as well as a couple of the books I referenced:
Web sites
Incomplete & Codominance
Incomplete Dominance and Codominance
Understanding Dominant and Recessive Traits
Dominance (genetics)
Incomplete Dominance - A Genetics Definition
Exceptions to Simple Inheritance
Books
Biology: Concepts & Connections
Genetics For Herpers
Phew.... 
-----
Troy Rexroth
Rextiles
