The "S" in SCID officially stands for "Structured" as in Structured Clinical Interview for DSM. But as far as I can tell for the last three editions (5, IV and III), non-official sources (research papers etc.) often refer to it as a semi-structured interview. e.g. , ,  / , or  etc. The first link  is actually APA's own publishing blurb for SCID-5. The last one  is Spitzer et al. (1992) paper on SCID for DSM-III... which has over 4,000 citations in Google Scholar.
So why is there this discrepancy between the name (as fully spelled out) and content of the SCID?
I find it even more baffling after reading the introductory section of Spitzer et al., which emphasizes that SCID was designed from the get-go as semi-structured in opposition to previous [completely] structured interviews that were being advanced by others.
The major contribution of DSM-III was its inclusion of specified diagnostic criteria for virtually all of the mental disorders. This made possible, at about the same time as the publication of DSM-III, the development of the National Institute of Mental Health (NIMH) Diagnostic Interview Schedule (DIS). The DIS was designed for use by lay interviewers with 1 week of intensive training. Although originally developed for use in the Epidemiologie Catchment Area Study in which the cost of employing clinically trained interviewers would have been prohibitive, it has been used by investigators in a large number of nonepidemiologic studies. In the presidential address at the 1983 annual meeting of the American Psychopathological Association, titled "Are Clinicians Still Necessary?" one of us (R.L.S.) discussed the potential limitations of the DIS in clinical research. Without questioning the advances embodied in the DIS, it was argued that the most valid diagnostic assessment still required the skills of a clinician with experience in evaluating a range of psychopathology.
Unlike a lay interviewer administering a completely structured interview, such as the DIS, an experienced clinician can tailor an interview by phrasing questions to fit the subject's understanding, asking additional questions that clarify differential diagnosis, challenging inconsistencies in the subject's account, and judging whether the subject's description of an experience conforms to the intent of a diagnostic criterion. This is particularly true in the diagnostic evaluation of subjects with psychotic symptoms who have little insight and present with a complicated story from which the interviewer must discern the symptom profile. For these reasons, two of us (R.L.S. and J.B.W.W.) started work on a clinical assessment procedure that would not only be linked to DSM-III but would incorporate several features not present in previous clinical diagnostic instruments.
So was the naming of the SCID a case of "embrace and extend" the concept of "structured" interview? I couldn't find the answer in Spitzer et al...