If we can crowdsource simple lists, we can crowdsource the very tools and symbols that we use to communicate and interact.
My reasoning:
A graph is a versatile mathematical object that can be used to format and represent just about any type of data. Any individual graph can be specified in full using nothing more than two simple lists: one list for nodes and one list for edges. If we can figure a way to crowdsource a simple list in a genuinely decentralized fashion (and we'll need to discuss as a community what that means), then we can crowdsource a graph. And if we can crowdsource a graph, we can crowdsource just about any piece of data, including the symbols used in the representation of that data.
That’s my motivation for DCoSL: decentralized curation of simple lists.