I would like to understand a bit better the nature of *bad* triangulations of $S^5$, discussed in two Lectures of Jacob Lurie 

https://www.math.ias.edu/~lurie/937notes/937Lecture2.pdf

http://www-math.mit.edu/~lurie/937notes/937Lecture3.pdf

The example in the lectures is based on the fact that the double suspension of the Poincaré homology sphere is homeomorphic to $S^5$.

I would like to fiddle a bit with Definition 1 from lecture 3. Here is the definition. 

**Definition 1**. Let $K$ be a polyhedron and M a smooth manifold. We say that a map $f \colon K \to M$ is called piecewise differentiable (PD) if there exists a triangulation of $K$ such that the restriction of $f$ to each simplex is smooth. We will say that $f$ is a PD homeomorphism if $f$ is piecewise differentiable, a homeomorphism, and the restriction of $f$ to each simplex has injective differential at each point.

In the situation that I want to consider, I would like to impose the condition instead that $f$ restricts *analytically* to each simplex (stronger condition). But only ask that the restriction has injective differential *in the interior* of each simplex (weaker condition). So here is a question:


> **Question.** Suppose $X$ is a simplicial complex that is homeomorphic to $S^5$. Suppose that the homeomorphism $\varphi: X\to S^5$ can be
> realized so that its restriction to the interior of each simplex in
> $X$ is an analytic diffeomorphism onto its image. Is it true then that
> $X$ is $PL$ homeomorphic to the standard sphere? (This would mean that
> there is a map $\varphi': X\to S^5$ that is $PL$ on each simplex of
> $X$ for a standard $PL$ structure on $S^5$.)