Horizon 2000 Plus: extend the plan without freezing its designs

Horizon 2000 Plus was developed in 1994–95 to continue European space science beyond the original Horizon 2000 commitments. Its contemporary description called it a rolling program: a way to preserve continuity while responding to scientific, technical and institutional changes. The plus sign did not mean every earlier plan would simply be repeated with an unchanged spacecraft.

The recommendations included a Mercury investigation, very precise stellar astrometry and low-frequency gravitational-wave observation. They also retained competitive medium-mission opportunities and identified technologies needing further study. These were scientific directions and development recommendations. Their appearance in a planning document should not be read as proof that every proposed instrument configuration later flew unchanged.

Gaia and BepiColombo illustrate the extension’s legacy. ESA’s BepiColombo history records a Mercury proposal in 1993, inclusion in the extended cornerstone planning in 1994 and a package approving both BepiColombo and Gaia in October 2000. BepiColombo then underwent additional approvals and redesign during the Cosmic Vision era. Its history crosses administrative boundaries without becoming an unrelated mission.

The Gaia launch image records a later milestone: liftoff from French Guiana on December 19, 2013. It connects the long preparation of an astrometry mission with an actual launch. It does not show the star catalog or demonstrate the precision of stellar positions; those results depend on subsequent observing, calibration and data analysis.

A proposal can define the measurement it seeks while engineers continue comparing ways to achieve it. The early Plus recommendations described advanced astrometric and gravitational-wave concepts. Later design choices need to be established from the specific mission’s development record, rather than assumed from a short summary of the original recommendation.

For a fictional mission history, build a sequence of dated decisions: initial recommendation, selection, redesign, adoption, launch and results. Record which source supports each stage. This approach prevents an early concept from being confused with delivered hardware and avoids assigning an entire mission to one framework solely because a database offers only one program field.