I hadn't been aware of a "spectacular failure," which is to be expected as I don't keep up with it very well. Could you spare a link or a very top-level view?
I'm hardly a molecular biologist, r9etb. The allegation of "spectacular failure" is not mine — it's something I've been coming across in my reading lately, which is to say in Robert Rosen and others. That characterization evidently is based on three things: (1) the incredible hype that went into the HGP; (2) the astounding resources (in scores of millions of dollars) that were poured into it; (3) its general failure to find what it was looking for. Though as metmom has already pointed out, it wasn't a complete bust; I'm sure some valuable things were learned that can help the effort in the long run.
On my reading, Robert Rosen suggests that the entire conceptual model of the HGP was the wrong way to attack the problem. For the genome is a complex system, and there is as yet no "good physics" for complex systems. In short, HGP was attempting to address the genome as a simple "mechanical" system, one that would conform to the standard of scientific "objectivity" that satisfies Church's Thesis — which is that all "effective" material systems in nature are "computable" — that is, theoretically reduce to the syntactical level capable of execution by a Turing Machine. The thesis says that only in this way can we be sure that our results are truly "objective."
Of course, the thought comes to mind that imposing Church's Thesis on science means that a good deal of reality — indeed, possibly even the most of it — cannot be addressed by science at all. (One can impose the "Church test," but material systems that fail to satisfy it do not then just conveniently disappear from reality.)
Complex systems — which Rosen says are the generic or "typical" case in nature, and simple systems the nongeneric — are simply not computable; they do not reduce to algorithms. They have the beastly habit of spawning what Bertrand Russell called impredicativities, or "vicious circles" involving self-reference and semantic meaning which cannot be reduced to syntax/algorithms/computability. Worse, they seem to invoke the Aristotelian "final cause" — which has been utterly banished from science for the past three hundred and more years because final cause in some sense seems to put the "cause" of something in the present in the future, and this the Newtonian formalism absolutely forbids. Though biologists working on what are generically called "anticipatory systems" evidently are trying to put it back into the scientific mix.
At the end of the day, perhaps we will discover that Robert Rosen (a very great mathematician and biophysicist) is right: Biology is not a "special case" of physics at all. Rather, the reverse is true: Physics is a special case of biology. Biology is "generic"; physics is nongeneric, or "special." My suspicion is that, if science can just get the hang of that notion, maybe then science could find an effective way to get a handle on complex material systems, such as the biological genome.
Paradigm shift, anyone?