The trouble is that VCS pain points are far from universal. Everyone has their own experience, for better or worse, and will have a view on what's important based on their personal experience (or lack of experience).

In other words, arguing for a model (mainline or trunk) without first setting a context is pointless.

For instance, consider the needs of a very large team versus a very small team. Consider a team that releases often versus a team that releases infrequently. Consider a team releasing multiple projects according to a roadmap versus a team releasing a single project on an ad hoc basis.

The variations are many, and so too are the strategies for managing source code.