Is it possible to construct a category $\mathcal{C}$ with $\mathrm{Ob}\,\mathcal{C}=\mathbb{R}$ and for every diagram of the from $$a_0\leftarrow a_1\leftarrow\cdots a_n\leftarrow\cdots$$ the inverse limit exists if and only if the corresponding sequence $(a_n)$ converges and in this case the limits coincide?

One should probably give some restrictions so that there are enough morphisms. To throw away stupid examples like $\mathrm{Mor}(x,y)=\varnothing$ for $x\neq y$ and $\mathrm{Mor}(x,x)=\{\mathrm{Id}_x\}$ let's assume that for every $x,y$ at least one of the sets $\mathrm{Mor}(x,y)$ and $\mathrm{Mor}(x,y)$ is nonempty(maybe someone will suggest more adequate restrictions)

Probably this question doesn't make any sense, though taking category structure from the structure of ordered set makes $\varprojlim a_i=\sup a_i$ which is close to the desired.

Any results for other metric spaces are also interesting.


First example. Let $\mathrm{Mor}(x,y)$ be empty for $x<y$, and consists of one element if $x\ge y$. So, if you can write such a diagram, $(a_n)$ is non-decreasing sequence. Therefore this diagram has the inverse limit iff $(a_n)$ bounded above, and inverse limit equals to supremum.

Second example. Let $\mathrm{Mor}(x,y)$ be the set of all paths connecting $x$ with $y$ (we need to consider this paths up to reparametrization, for composition law will be associative). This example works for all converging sequences, when $a_n\to a_{n-1}$ is "straight path", as it easy to check.

(it looks like a fundamental $\infty$-groupoid of a topological space, but i'm not shure that $\infty$-constructions will work as you want)

  • $\begingroup$ actually, it works for all (locally) path-connected spaces $\endgroup$ Aug 26 '15 at 15:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.