| 1 |
Finish
SGML markup, put SGML document(s) and DTD on server |
| 2 |
Based
on SGML & it's DTD, OpenText builds an index, which it uses
to perform searches (more efficient than actually searching through
an SGML document for every search). Not really important HOW this
index is built. |
| 3 |
User
enters a search query (WWII
poster database) |
| 4 |
OpenText takes the
user's query, performs the search, returns a list of "hits"
and some other abbreviated data from the associated record.
Displays these hits in a Web browser as clickable links.
|
| 5 |
The user can click
on these links to retrieve a complete record. Very much like
Voyager, other database searches up to this point.
|
| 6 |
FOR MORE COMPLEX
DOCUMENTS (EAD, ETEXT), it's not enough to just deliver the
entire record, because it is probably very long and quite complex.
Let's look at how the University
of Michigan handles its EAD searches through OpenText.
|
| 7 |
Custom scripts must
be written to not only display abbreviated data in a search
"hit" page, but to then break down the document further
for web display and navigation. This is the complicated part.
|