katlog.orgdevelopment journal Demo catalogueDemo
ID003
Ideas2 min read

Moving a library off Aleph: a step-by-step scenario

Everything we learned during the experiment adds up to a scenario others can repeat. Not an installation manual, but a sequence of steps and the questions a library should answer in advance.

Moving to a new system is daunting because of its scale. But broken into steps, most of it can be tested with a small experiment before any decisions are made.

  1. Get the dataFor a real migration: a full export using Aleph’s staff tools, with the system administrator’s permission. For a first trial, the standard export of the public catalogue is enough. Compare the number of records in the file with the number in the catalogue straight away.
  2. Study the dataWhich fields are filled, which collection codes occur, what the status codes mean, how years and subject headings are recorded. There are always surprises — as in the holdings lines.
  3. Build a prototype on your own dataA single search page on real records turns a conversation about “what we want” into a conversation about concrete screens.
  4. Assess data qualityHow many records lack a year, a call number, a language; how many duplicate ISBNs. Decide what to fix before the move and what after.
  5. Map the fieldsWhich field of the old system goes where in the new one. What to do with fields that don’t exist — which is what custom fields in Katlog are for.
  6. Set the rulesReader-facing collection names, loan rules, loan periods, the number of renewals. These are the library’s decisions, not the software’s.
  7. Trial import and reconciliationLoad, compare the numbers, and go through a few dozen records by eye — especially rare and complex ones.
  8. PilotA few weeks of parallel running: the old system stays primary, the new one is tested on real tasks.
  9. Switch-overA final export, a repeat import that preserves corrections — and only then the new system for readers.

Questions to answer in advance

  • Who in the library will be responsible for the data after the move?
  • Which reader and loan data actually needs to be migrated — and is it allowed?
  • Where will the system run, and who will update it?
  • What to do with MARC records and authority data, if there are any?
  • Is a connection to a regional union catalogue or the national library needed?
Being honest about the limitsKatlog is an experiment on open data. It shows that this path is realistic, but it doesn’t replace the legal, organisational and technical preparation a library needs for a real migration.