You may ask, well, so what. Of course, assuming a constant rate of mutation and so the molecular clock was a stupid idea, and it was being reflective of NeoDarwinian simplistic thinking.
It was not, however, reflective of thinking in the Synthetic Model, which
has always been wary of the molecular clock hypothesis, and (once again) contrary to your assertions, scientists have been fully aware of all the problems in it.
As a result, the molecular clock hypothesis only has limited application for investigating the Synthetic Model, used mainly in the specialized field of molecular phylogenetics. In other words, it's flaws are well known, and so it's used cautiously as an adjunct to help with cladistics, and it's certainly not the basis of any of the
framework of the Synthetic Model.
Even the paper that your creationist website claims is evidence for how "NeoDarwinism" is falling apart as long-believed things are abandoned is false. See the sentence where it says "Here, we propose that these innovations represent welcome progress towards obtaining reliable dates from the molecular clock"?
It says
progress towards obtaining reliable dates because the paper's authors are well aware that the dates are unreliable, and the paper is about discoveries they've made which would make the molecular clock reliable enough to provide accurate dating.
The most promising
approaches for analysis, such as that of Drummond
et al. [21], allow for uncertainty in the dates attributed to
calibration points and do not impose unproven assumptions
about the pattern in clock-rate variation among
lineages. Some pressing questions remain, many of which
are not theoretical problems but require empirical investigation.
If the substitution rates turn out to be autocorrelated
after all, or if they are predictable from the biology of
the species, then it should prove possible to exploit this
knowledge.
For the moment, we need to determine the precision of
clock-based date estimates when realistic errors are specified
for the dates of calibration points and when appropriate
allowance is made for the existence of rate variation.
By combining information from many species, the recently
developed Bayesian methods enable the extent and pattern
of the clock-rate variation to be roughly characterised,
and allowed for. When the information is available, the
inclusion of additional calibration points to the analysis
should therefore produce more accurate clock-base dates,
but it remains to be seen whether they will be more precise
(e.g. have reduced standard errors).
But evos roundly and severely derided Denton as wrong.
Because he was, about a whole lot of things. Really
basic things, too.
The molecular clock has very little to do with why Denton is so roundly derided.
If evos had honestly and objectively considered Denton's arguments, they might have seen that their ideas on the molecular clock were premature, but typically evos maintain a dogmatic orthodox approach in order to silence their critics and so continued in their error when the ID theorist, in this matter, was correct.
As even a cursory reading of the cited paper shows, Denton's arguments have absolutely zip to do with theirs. So listening to Denton certainly wouldn't have helped them.
Tomoko Ohta's "Nearly Neutral Theory of Evolution" of 1993, cited in the paper mentioned at that creationist website, actually contradicts Denton's claim by confirming that amino acids show the very similar substitution rate across lineages with wildly divergent generation lengths (Denton, remember, said they should show equally wildly different substitution rates).
Instead, it's
DNA-sequence data that is affected most by generation size, and therefore the thing that shows so much variation among lineages (and, therefore, the reason why the molecular clock is so unreliable, contrary to Denton's claims about it).
Here's the relevant part of the paper, conveniently unquoted by your creationist website in their effort to whitewash and redeem Denton's ridiculous book:
In fact, Ohta [5] found that generation-time effects are
more apparent in DNA-sequence data than in comparisons
of amino-acid sequence. She explained this pattern with
her Nearly-Neutral extension of Kimura’s theory (Box 1),
which argues that substitution rates can be elevated in
small populations by the fixation of mildly deleterious
mutations, and that this effect, among others [6–8], can
compensate for longer generation times. Mutations are
more likely to be deleterious if they are amino-acid changing
(i.e. non-synonymous), hence the compensation
between population size and generation time might be
more effective for amino-acid sequences than for DNA.
The fundamental principles associated with theNearly
Neutral Theory mean that the rate of the molecular clock
is known to vary between evolutionary lineages, and that
it does so in a way that is not precisely predictable,
although broad trends might be anticipated.
Wonder why they left that out of your link, there.
So even on something where you'd think most evos (and there were some) would be more cautious than insisting the molecular clock was valid because it was part of a defense of evolutionism, they plunged ahead just as they did with Haeckel, recapitulation, evolution only through a change in allele frequency, depicting a fully terrestrial animal as a swimming whale, insisting years ago the fossil record detailed gradual change, claiming pseudogenes, etc, etc,.....
You have no idea what the molecular clock hypothesis is used for, when, and why, in evolutionary studies.