This chapter introduces the usage of the World-Wide Web interface to the BIBDIA Online Public Access Catalogue, the WWW-OPAC. This system allows you retrieve information from the libraries online catalogue.
This is done by filling out one of the three available query-entry-forms and submitting the selected form for execution. The available query-entry-forms are
The query is being processed by matching the user-provided information with the contents of the internal search registers. The result is being presented in the first step as a list of short-titles, the so called "Short-Title List". In the next step, you can mark any number of titles from the short-title list and ask the system to return the "Full-Title Information".
To get the best results its essential to understand how the system works. First, think of the following hypothetical book:
Author : Andersen, Eva Title : Makramee als Kunst und Hobby Subject : Knüpfen Signature : 6,7 Catc ISBN/ISSN : 3-8068-4085-7 Publisher : Falken-Verlag Place of Publication : Niedernhausen, Ts. Year Published : 1980
When you want to search for this book in the catalogue you have to provide some input for the OPAC search engine. This is for example
When you look back to the definition of the hypothetical book you see that the information on the right side defines the values for the categories on the left (author, title ...). In order to make the values of these categories searchable, some conversion and filtering has to be applied. The result of this processing then becomes the contents of the search registers. Before we'll look at the concrete conversion and filtering, we'll explain the reason for it.
Example: When you look at a title like "Advanced Programming in the UNIX environment" you notice that the title consists of four words that are really worth indexing. These words are:
The other two words "in" and "the" are not worth searching, that means it doesn't make much sense to include these as single words in a search register. But we cannot simply remove these words from the title, because a book might exist that is called "Advanced Programming for the UNIX Environment". These two titles are different and this must be reflected in the search registers. The problem is solved in the following way:
We have two registers, one for the exact title and one for selected words from the title.
But removing stop-words from the contents of the word-registers is not the only conversion that is applied to the values of the categories. Normalization also includes
All these conversion are applied to increase the likelyhood that you find what you search for without having to know every little detail like wether the author is written with or without a tilde, accent, dash etc... Note that the same normalization is applied to your input when entering search values in a query-entry-forms.
The system distinguishes between search registers for single words and search registers for strings. In the WWW-OPAC strings are enclosed in < > signs. But not all categories support the search on either word or strings.
The following registers are used when only a single word is specified as the search value.
AW : Author TW : Title DW : Subject IS : ISBN/ISSN PU : Publisher PP : Place of Publication PY : Year Published
The following registers are used when a string (usually containing more than one word) is specified as the search value. Note that the search strings must be enclosed in < > signs.
AA : Author string TA : Title string DA : Subject string SI : Signature string
See the following examples:
AW=Andersen AW=Eva AA=<Andersen, Eva> TW=Makramee TA=<Makramee als Kunst und Hobby>
Of course you can use any wildcards when defining values for the word or string registers. See the following examples:
AW=Anders? AA=<Andersen?Eva> TW=Makrame# TA=<Makramee als Kunst?>
When defining a query, the search registers are associated with search values. These search values are either words are strings (containing more than one word). Strings must be enclosed in < > signs. There are registers that allow words and other registers that allow strings.
When defining search values for more than one search register or multiple values for the same search register you have to specify how the intermediate results are to be combined. You do this by using one of the following logical operators:
and : logical AND or : logical OR not : logical AND NOT
Instead of using {and, or, not} you can use the following operators.
* : logical AND + : logical OR % : logical AND NOT
However, we recommend to use the first and more verbose form.
Unless you use wildcards, the string is always searched exactly as you specified it. That means, when you enter "peter" as an authors name, you won't see "petera", "peterle" etc.. Therefore the system supports some wildcards:
# zero or one character ! exactly one character ? zero, one or more characters
So if you want to see all authors starting with "peter" just enter "peter?". Note the difference between the ! and the ?: If you enter "peter!", so won't see authors whose name is just peter. This is because "peter!" searchs for a string that is exactly 6 characters long and where the first 5 five characters are "peter".
Please note, that some limitation exist in terms of maximum occurences of a field type. For example, you cannot specify a query that defines more than 5 authors. The current limitations are as follows:
The following fields have no limitations, that mean you can specify any number of words or strings.
Home Page