| developed at Carnegie MellonŐs Center for Machine Translation in the School of Computer Science |
| Hardware Architecture: 7 workstations used as server engines, independently accessing the same index database using replicated data on local disks |
| Gathering software: Written in Perl; uses C and libwww to fetch URLs |
| SCOUT Indexer: full boolean retrieval with spelling correction and approximate phonetic matching |
| Search Capability: document title, heading, link and keyword, approximate matching and probabilistic retrieval |