ecn-pa:scott (10/25/82)
Several months ago (I think it was the beginning of July), I posted a query about library applications on UNIX. I received several responses and have since learned about other applications. This article is a summary of the efforts that I'm aware of. There are three classes of library automation efforts. 1) Building something on top of a DBMS or other set of tools. 2) Relatively small text handling tools that can handle bibliographic records, but which don't handle things like circulation, acquisitions, etc. 3) Full blown library automation systems (the buzzword is "bibliographic control system"). People in the first category mentioned things built using INGRES, MRS, and refer. There were a handful of people in the second category. I haven't been able to get much information about most of these, and will submit a second article summarizing them if and when I do. As far as "real" library systems (as in "real men..."), there were basically three. 1) ILS (Integrated Library System) was written at the National Library of Medicine. It was written in MIIS/MUMPS. It is public domain and relatively complete. If there is sufficient interest, I'll post more on it separately. 2) BRS (Bibliographic Retrieval Systems?) developed a system for the Dartmouth College library. It does full text retrieval with reasonable response time. The user inter- face isn't one of their strong points. It has limited relational capability. Another company, Emanuel Data Systems, developed a hand-holding user interface to sit on top of BRS's system. They were at Usenix in Boston. It was written to run on an 11/70, but they are reported to be in the process of porting it to a Vax. It also runs on an Onyx machine and other UNIXish micros. 3) Bell's library system. Bell has an online catalog which is Dialog-like and a circulation system. I don't know a whole lot about it, but they are reported to be considering marketing it. Last, and I hope not least, Purdue is at the front end of a project designed to bring the Purdue Libraries out of the dark ages. If all goes well (and we have absolutely no reason to believe that it will) we will have a reasonable integrated library system in something on the order of five years. We also plan to spend some time evaluating some of the other systems just mentioned. Scott Deerwester Purdue University Libraries