@koraks
As you might know the major component of gelatin is partly hydrolyzed collagen.
Collagen is an unusual protein. First, it is a fibrous protein. Most proteins have a globular structure. However, the native structure of collagen is long and narrow, sort of like a sphagetti. Second, collagen has a very stable structure. It is difficult to denature.
The partial hydrolysis which converts collagen into gelatin results in shortening the fibers but leaving its structure more-or-less intact. This results in a molecule that is water soluble at higher temperatures but turns into a gel at cooler temperatures, as I am sure you are well aware!
The formation of the gel occurs because the fragments of collagen stick to one another. These interactions are relatively weak (and thus reversible). In chemical terms these interactions are non-covalent.
The hardening of gelatin involves stabilizing the gel by introducing stronger (covalent) bonds between the collagen fragments... i.e. cross-linking them together.
There are many ways to harden gelatin. The gelatin in traditional silver-gelatin emulsions is hardened using formaldehyde or glutaraldehyde. This reaction is not dependent on light.
Dichromate is the traditional way to harden gelatin in a light dependent reaction.
DAS will also introduce cross-links between collagen fragments in a light dependent manner.
However, the nature of the cross-links made by the two compounds are very different (both in their chemical details and in their frequency). Thus the behavior of hardened layers made by different methods are likely to be similar (especially in terms of trapping pigment) but not exactly the same.
Lastly, it might be worth noting that the chemistry of gum printing is, at one level, similar to that described above even though gum is a carbohydrate polymer and not a protein (an amino acid polymer). In other words gum arabic is a long thin molecule that is cross-linked to make it insoluble while trapping pigment particles to form an image.