When you put JS on top of it to intercept, you are creating two paths. That may be easier than a totally separate code base, or maybe not (I have seen compellin

When you put JS on top of it to intercept, you are creating two paths. That may be easier than a totally separate code base, or maybe not (I have seen compelling arguments either way), but it still boils down to a different code path in terms of potential for different bugs, different cross-browser issues, and more testing. In addition, developing that way (instead of with JS in mind) also has a cost which in the end means that you are sacrificing resources developing for those 2% instead of the other 98%. Regardless of whether that approach leads to an easier situation in terms of maintaining JS vs. non-JS, it certainly does not lead to an easier situation than ignoring non-JS altogether.