# Eric Holloway: Algorithmic Specified Complexity

**URL:** <https://discourse.peacefulscience.org/t/eric-holloway-algorithmic-specified-complexity/1375>\
**Category:** Office Hours\
**Tags:** Design\
**Created:** [August 28, 2018, 6:07am UTC](https://discourse.peacefulscience.org/t/eric-holloway-algorithmic-specified-complexity/1375 "2018-08-28T06:07:29Z")\
**Posts on this page:** 1\
**Showing post:** 11

<div class="post-metadata">

**Author:** ![EricMH](https://sea2.discourse-cdn.com/flex016/user_avatar/discourse.peacefulscience.org/ericmh/32/679_2.png) [@EricMH](https://discourse.peacefulscience.org/u/EricMH)\
**Post date:** [September 1, 2018, 12:55am UTC](https://discourse.peacefulscience.org/t/eric-holloway-algorithmic-specified-complexity/1375/11 "2018-09-01T00:55:19Z")

</div>

> [@swamidass](#):
>
> Marks is correct to point out that intuitive concepts of information do not map to information theory:
> 
> > While humans operate using an intuitive concept of information, attempts to develop a theory of information have thus far fallen short of the intuitive concept.
> 
> This is the problem that he is trying to solve here, by providing a quantitate way of measuring the “information” content that is indicative of design.

Yes, it is a variant of mutual information. If P(X) = 2^{-K(X)}, and we use the maximally compressed form of C (denoted by C^\*), then ASC(X,P,C) = K(X) - K(X|C^\*) = I(X:C), where I(X:C) is algorithmic mutual information. Per our last discussion, algorithmic mutual information cannot be generated by random and computable processes any better than by flipping a coin. Hence my claim that naturalism cannot create information that fits our intuitive notion of information.

> [@swamidass](#):
>
> So the function I implemented is claimed to be a reliable way of detecting design. A process that produces high ASC should, therefore, be only possible if intelligence is being injected into the process for Marks to be correct.
> 
> Do you agree @EricMH?

If you properly formulate P based on all possible chance and necessity causes, then yes, X with high ASC would give high probability something other than chance and necessity created X.

---

_[View the full topic](https://discourse.peacefulscience.org/t/eric-holloway-algorithmic-specified-complexity/1375)._