CH004
Chronicle2 min read
What remains unsolved
A development journal would be dishonest without this article. Katlog works, but it has known weaknesses, untested features and open questions. Here they all are.
It is easy to talk about successes. Harder to talk about what hasn’t worked yet. We will keep this list up to date.
Known weaknesses
- Demo data. Some data corrections — removed false “holds”, collection codes, languages — were made in development only; on the demo site this data is still the old one. The catalogue will be loaded afresh later.
- Small interface issues. Some cards in the librarian workspace are cramped on a phone, and after a record is deleted the search result may briefly show the old state.
- No email notifications. Readers learn that a book is ready only when they log in.
- The import doesn’t accept MARC files. Only JSON, RIS and CSV. So far Katlog only parses MARC when looking up descriptions by ISBN.
Untested
- Working with real readers. All scenarios have been tested on test accounts.
- Large volumes. The import is designed for tens of thousands of records but has been tested with six thousand.
- The feature descriptions in this journal are based on the code and reports, not only on hands-on use. Some details may differ in practice.
Open questions
- Where and how to store reader data if the system is ever used in earnest: a question of law and contracts, not just technology.
- How to roll out updates to copies of the system in different libraries.
- How to connect the catalogue to regional and national databases beyond description lookup.
Why we publish thisBecause for colleagues thinking about their own project, a list of problems is more useful than a list of features. It shows where it will be hard.