Log-log interpolation of hazard curves

Per @zkortum ...noticed that the response spectrum data given back by the service seem to be interpolating off the hazard curves in semi-log space, rather than log-log space. I talked with @nluco and he thought the static hazard curve service ought to be interpolating in log-log space. I wanted to check if there was a specific reason the web service was interpolating in semi-log space, or if this is an error that should be fixed?

Theoretically this is done in the same place for both dynamic and static hazard curve services.