An old telescope back in agreement
The observatory’s largest telescope had not moved for nine years. Its steel tube remained beneath the dome, but the control system depended on obsolete components and the mirror coating had degraded into a faint grey haze. When engineer Salma Idris was asked to return it to public use, supporters imagined a repair followed by a ceremonial first view. Salma described the task differently: the telescope was not one machine waiting to be restarted but an arrangement of optics, structure, software, building and practices that had ceased to evolve together.
37The original control computer could still be powered, although its hard drive failed unpredictably. Replacing it seemed obvious until the team discovered that several safety interlocks existed only in undocumented software routines. A modern controller could move the telescope more accurately while omitting protections nobody had realised were part of the old program.
38The team therefore recorded the electrical signals produced during every permitted movement and compared them with paper maintenance notes. Former technicians returned to explain alarms whose labels had outlived the faults that created them. One warning called “north obstruction” no longer referred to an obstruction; it had become a convenient way to slow the final degrees of rotation after a gearbox replacement in 1998.
39Meanwhile, optical specialists removed the mirror. Its surface looked uniformly poor, but tests showed that the central region remained usable and that most loss came from contamination near the edge. Recoating the entire mirror would provide the best image, yet transport introduced risk and cost. The public project could proceed with a partial cleaning if expectations were adjusted.
40That qualification changed the scientific programme. The telescope would not compete with modern research instruments, but it could reveal bright objects and, more importantly, expose the process of observation. The new control interface was designed to show alignment, tracking corrections and uncertainty rather than hiding them behind a single “go to target” button.
41The dome introduced a different problem. During the years of inactivity, its wheels had developed flat spots, so the opening could lag behind the telescope and block part of the sky. Sensors could automate coordination, but Salma insisted on a manual mode for training and failure. Operators needed to understand the geometry well enough to recover when automated assumptions became wrong.
42A month before reopening, the team conducted an overnight rehearsal. The telescope reached its first target, but condensation formed on a newly installed camera because the warm control room leaked air through a cable duct. The fault had nothing to do with astronomy and everything to do with the building modifications made to support visitors.
43The opening night was delayed by two weeks. When it finally happened, the first image was not displayed as evidence that the old telescope had been restored to its former state. The control room screens showed the corrections, cleaning limits and environmental readings that made the image possible. Visitors saw a functioning instrument, but also the continuing work required to keep its components in agreement. Documentation itself had to be recommissioned. The observatory possessed manuals, handwritten amendments, photographs of control panels and memories that contradicted one another about which procedures were mandatory. Salma created a dependency register linking each instruction to the component and risk it addressed, then required the team to demonstrate the relation physically where possible. An instruction with no remaining function was not deleted immediately; it was isolated until the absence of function had been tested under failure as well as normal operation. Public access introduced operational states absent from the telescope’s research history: queues changed floor vibration, explanatory lighting affected dark adaptation, and visitors required the dome to pause at positions that operators had previously crossed quickly. These were not external inconveniences added after technical restoration; they altered the conditions under which tracking, cooling and safety had to work. The commissioning tests therefore included people as part of the system, not merely as beneficiaries waiting outside a machine whose successful operation could be established in an empty building.
Paragraph Options
This did not mean visitors would operate the telescope without constraints. Manual mode used reduced speeds and required two people to confirm critical movements. Its purpose was to preserve comprehension and recoverability, not to celebrate unaided human control over automation.
The decision frustrated donors who expected the restoration to culminate in the sharpest possible photographs. Salma argued that image quality was one objective among several and that an instrument closed for an ideal recoating would teach less than a limited instrument whose compromises were documented and observable.
For that reason, recommissioning began with dependency mapping instead of component replacement. The team listed what each action assumed about another part of the observatory and which assumptions could no longer be tested because staff, documentation or hardware had disappeared.
This made the old warning labels historical evidence rather than instructions that could be transferred literally. The new system retained them in an annotated log, linking each label to the present condition, the earlier modification and the test used to decide whether the protection still mattered.
The rehearsal was redesigned around that vulnerability. Instead of repeating the successful sequence after sealing the duct, the team varied room temperature and visitor numbers to identify the conditions under which condensation returned. A building-service fault became part of the observing protocol.
The missing routines made simple replacement unsafe. The first phase of modernisation would therefore imitate the old controller closely enough to test its behaviour, while every divergence had to be justified through physical inspection rather than assumed to be an obsolete software quirk.
Several local schools had continued using smaller telescopes on the site. Their teachers were invited to advise on the public programme, although the restoration team resisted turning educational popularity into evidence that every technical compromise was acceptable.
The limitation became an educational advantage only because it was made explicit. Rather than promising access to distant objects as finished images, the programme treated every observation as a chain of adjustments whose reliability could be discussed.