A common characteristic of these neighboring information universes is that they are (at least partly) amenable to integration with mainstream chemical information systems. Another characteristic is that, in general, such integration is currently poor or non-existent.
This suggests that a productive avenue of research would be in development of methods which provide this integration.