Migration guides¶
The programme's practitioner track: two volumes on what it would take to run an organisation's collaboration stack on open source, and what it costs. They adapt six public studies (two commissioned by the Swiss Federal Chancellery in 2024, one by the German state of Schleswig-Holstein in 2022, one German and one French study of the Linux desktop, and the French field evidence) from their public-administration setting to organisations generally, and update the product assessments. The economics come from the rest of the programme: exit cost rather than licence cost is the object, supplier depth is a first-class selection criterion, and neither guide claims a discount. Product statuses are stated as of a given date, and entries the guides mark as moving need re-checking before a procurement decision.
Open Source Back-End Services¶
Guide 1 of 2: directory, file services, mail and calendar, document collaboration, and real-time communication. The framing decision comes first, and it is not binary: between staying with the incumbent and migrating off it sits a third position, building credible optionality (federated identity, open formats, a working reserve), which buys most of the benefit for a fraction of the cost and is the correct first phase of a migration in any case. The guide departs from its sources in treating supplier depth as a selection criterion alongside function and maturity, in splitting directory of record, authentication, and device policy into three procurement decisions, and in pricing exit rather than licences. It then covers what the source studies leave out, which is most of the engineering: the hosting decision, reference architectures and sizing, data-migration mechanics, coexistence patterns, backup and continuity, security baselines, operations, and the contract. The cost model derives the incumbent per-seat cost at which migration breaks even, so no assumption about your agreement is needed. Credible options now exist in every domain examined, and two positions improved materially after the sources were written: groupware with native Outlook compatibility, and integrated sovereign suites. No source in the corpus supports a claim of savings. Companion explainer: The middle option almost nobody picks.
The Open Source Workstation¶
Guide 2 of 2: the applications on the end-user machine, the operating system under them, and the machinery that manages both across an estate. It treats the two stages as separable. Open-source applications on Windows or macOS cost no licence money, install alongside the incumbents, need no fleet-management change, and are reversible; the Linux desktop is a multi-year programme whose difficulty sits in provisioning, policy, packaging, client identity, encryption, printing, and remote support rather than in the applications, which reached parity years ago. The finding worth acting on inverts the usual plan: the Swiss study's own prioritisation method, applied to its own recommendations, puts stateless single-purpose tools first and the office suite and mail client last, because those carry the format entanglement. The German and French sources agree that the operating system changes last. And the French field evidence, the only observational evidence in the corpus, finds the strongest resistance coming from the IT department rather than from users. Companion explainer: The office suite goes last.