I could have been clearer in my original post. Let me clarify by addressing your points.
1. People get into planes and cars every day routinely because we have been using these technologies for decades. Any new technology that comes along will be held to much higher expectations of safety and quality than currently available modes of transport were at their inception. Understanding the risks and learning how to mitigate them to an acceptable level will take some time but we should be able to get some of the benefits of high speed transport earlier.
2. I'm not suggesting making an unsafe for humans version or a version with incompatible design goals. At some point on the path towards having a system that is acceptably safe for humans, there must be a system that is safe enough to send non-human cargo so why not take advantage of that? Having a real functional system would allow larger scale testing that could provide very useful feedback when trying to work out all the kinks before human passengers get on board.
3. Good point. I've got no idea of the scale of any impact on road usage that this would have. It may or may not be a compelling goal by itself, but as a side effect of implementing technology that is a necessary stepping stone to transporting people, it could at least be a nice bonus. Negative PR from job losses comes along with many disruptive technologies and is generally an area where there's lots of room for improvement.
Shipping may be invisible to most but people do notice the ease of access to goods, in their availability and price. The eventual possibility of cheaper same-day shipping over large distances sounds okay to me. This doesn't have the same direct impact to an individual as being able to travel that far or fast themselves but its a start. I might be going out on a limb here but it seems to me that having faster means of moving goods around at large scale should make much more efficient distribution of resources possible. That's something we can all benefit from.
I'd be disappointed if Elon did shy away from the goal of a new form of human transportation but no one has said that's what SpaceX are doing when they send unmanned capsules up to the ISS.
Replacing a rail freight line with a system that can potentially move physical goods at the scale of road freight at twice the speed of air freight, and doing it cheaply, could be Hyperloop's minimum viable product.