On string fields and superstring field theories

We offer some thoughts regarding the space of string fields. We suggest that this space should be identified as the odd component of a star-algebra and focus among other issues on the role of the mid-point. We argue that theories with mid-point insertions in the action, such as the modified cubic theory can be well behaved, even if this mid-point insertion has a non-trivial kernel. We then discuss the recent proposal by Berkovits and Siegel of a non-minimal superstring field theory. In this theory the action contains a mid-point insertion of a non-zero conformal weight. We show that, while this is a-priori a problem, it might be possible (in the NS sector) to make sense out of this theory by regularizing it. A cleaner resolution of the problem is to extend the non-minimal sector in a way that allows a zero-weight mid-point insertion with the desired properties. We also study the generalisation of the theory to the NS- sector and explain the problems with defining the Ramond sectors. We show that the non-minimal theory supports all the known solutions of the standard modified cubic superstring field theory, including the GSO+ vacuum solution. The properties of the solutions carry over to the non-minimal theory. In particular, the vacuum solution has the correct tension and cohomology.