A look at FORM, a 1980s software used for the hardest particle physics problems, as it risks obsolescence while being maintained almost entirely by a retiree
Maintenance of the software that's used for the hardest physics calculations rests almost entirely with a retiree.
Context & Ripple Effects
FORM sits in a lineage of scientific infrastructure built by a handful of people and kept alive by fewer: Paul Ginsparg's arXiv preprint repository showed how one physicist's tool can become load-bearing for an entire field, and FORM did the same for symbolic calculation in particle physics. The difference is that arXiv institutionalized while FORM's maintenance still rests almost entirely on a single retiree.
The story rhymes with the COBOL problem that has shadowed US government and banking systems, where dwindling expertise around decades-old code leaves critical infrastructure one retirement away from being unmaintainable — except FORM's user base is a research community with no revenue stream to fund a rescue.
First-order effects
- Particle physicists who depend on FORM for the field's hardest symbolic calculations face a bus factor of effectively one: if the retiree maintainer steps away, results that rely on it become hard or impossible to reproduce or extend.
- Research groups building on FORM have no vendor or institution contractually responsible for fixes, so any breakage lands directly on their own timelines.
Second-order effects
- The COBOL precedent suggests what comes next: as with banks and agencies that struggled to find programmers when old systems broke under pressure, physics collaborations will face a scramble to either train new FORM maintainers or port calculations to modern alternatives at real cost in time and verification.
- Funders and universities may be pushed to treat maintenance of unglamorous scientific software as fundable work, since the alternative — losing the tool — costs more than sustaining it.
Third-order effects
- If the pattern holds across fields, critical software outliving its creators becomes a structural condition rather than an anomaly, much as IBM mainframes have persisted for 50+ years as a business mainstay despite their age — forcing institutions to budget for code archaeology alongside new development.
- Scientific computing may bifurcate between fashionable new tools and irreplaceable old ones, with prestige flowing to the former while the latter quietly carries the field — a gap only deliberate stewardship closes.
The trend: Critical software in science and finance is outliving its original maintainers, and institutions are only beginning to treat sustaining aging code as infrastructure worth funding.