Every night, a CPAP machine records an extraordinary amount of information: every breath, every airflow limitation, every apnea and hypopnea, the exact pressure delivered, the mask leak rate, the precise timing of each event. But most people who use these machines never see almost any of it. The manufacturer's companion app — ResMed's myAir, for instance — typically shows a simplified "score" and a few summary numbers designed to reassure rather than inform.
The software that changed that for hundreds of thousands of patients was not built by a device maker, a hospital, or a startup. It was built by one man who had sleep apnea himself and simply wanted to see his own data.
One Developer, One Frustration
SleepyHead was written by Mark Watkins, an Australian software developer who went by the handle "Jedimark." Diagnosed with obstructive sleep apnea, Watkins was frustrated that the detailed data his machine collected every night was locked away behind proprietary clinical software that patients were never meant to touch. So he wrote his own.
There was a real technical obstacle in the way. The data on a CPAP machine's SD card is not stored in any open, documented format — it is a manufacturer-specific mix of copyrighted software and raw sensor data, protected by digital rights management. To read it, Watkins had to painstakingly reverse-engineer those proprietary formats, machine by machine, decoding what each byte on the card actually meant.
From roughly 2011 through 2018, Watkins developed and maintained the result as free, cross-platform, open-source software. It ran on Windows, macOS, and Linux, read the raw data card from a wide range of CPAP and bilevel machines, and rendered it in a way no consumer had been able to see before: breath-by-breath flow curves, detailed event flagging, pressure and leak graphs, and statistics a patient could actually use to understand their own therapy.
For a community of engaged patients, it was transformative. Someone struggling with residual apnea events or mysterious mask leaks could now look at the same class of data their sleep clinician saw — and often spot patterns their clinician had missed.
Why the Data Access Mattered
The significance of SleepyHead went beyond convenience. It sat at the center of a broader shift in medicine: the rise of the informed, data-literate patient.
Manufacturer apps are generally designed around adherence — proving to insurers and clinicians that a patient is using the machine enough hours per night to qualify for continued coverage. What they are not designed to do is help a patient optimize their own therapy. SleepyHead filled that gap. On patient forums like CPAPtalk and Apnea Board, experienced users would ask newcomers to post their SleepyHead charts, then walk them through interpreting a night's data and suggesting adjustments to discuss with their doctor.
That crowdsourced, data-driven self-advocacy would have been impossible without a tool that exposed the numbers in the first place.
The Legal Gray Zone
That tool, however, existed in a precarious legal position. Because reading the SD card meant circumventing the machine's DRM, SleepyHead brushed up against Section 1201 of the Digital Millennium Copyright Act, the US law that makes bypassing digital protection measures illegal — even when the underlying purpose is entirely legitimate.
In 2015, following advocacy from medical-device researchers and patient-access groups, the US Copyright Office granted a DMCA exemption covering certain access to medical-device data. But the relief was narrower than it first appeared. The Copyright Office generally grants exemptions for the use of circumvention, not for the tools that make it possible — meaning a patient might be permitted to use software like SleepyHead, while writing, distributing, or contributing to that software remained in a legal gray area. In 2018, Boing Boing described the situation bluntly, framing thousands of sleep apnea patients as dependent on "a lone Australian CPAP hacker" operating in that uncertain space.
None of this was theoretical for the person maintaining the code. A single volunteer was carrying both the technical burden of the project and its legal exposure.
The Break, and the Fork
Open-source projects live and die by the people who maintain them, and by 2018 the pressure on a single volunteer developer had grown. Following a disagreement with some of the people helping to develop the software, Watkins decided to step back, and the SleepyHead project effectively closed.
For any other program, that would have been the end. But because Watkins had released SleepyHead under an open-source license, the code could not simply disappear. In February 2019, a group of volunteers formally forked the final release, SleepyHead version 1.1.0, and renamed it OSCAR — the Open Source CPAP Analysis Reporter. The first release, OSCAR v1.0.0, was deliberately close to the SleepyHead version it descended from, preserving continuity for the users who depended on it.
This is exactly the safety net that open-source licensing is designed to provide: the departure of the original author ended his involvement, not the software's life.
A Community-Run Project
OSCAR today is maintained not by a company but by a rotating group of volunteers drawn largely from the CPAPtalk and Apnea Board communities — the same forums where SleepyHead had become essential. It remains released under the GNU General Public License version 3, and the project is emphatic that it is free and always will be.
Functionally, it has continued to expand well past where SleepyHead left off. OSCAR now reads data from a broad and growing list of machines — ResMed and Philips Respironics devices alongside newer entrants like BMC, Yuwell, and Löwenstein — and offers something proprietary software still generally does not: the ability to compare data across machine types and manufacturers in a single, consistent view, right down to breath-by-breath analysis.
What This Means for Patients
OSCAR is not a medical device, and it does not replace a sleep physician. Its own maintainers are careful to say that any therapy changes should be made in consultation with a clinician. But for a patient trying to understand why they still wake up exhausted, or whether a new mask is leaking, or how their pressure settings are actually performing night to night, it remains the most powerful window available into their own treatment.
Its history is a reminder that some of the most useful tools in patient care do not come from the medical-device industry at all. They come from patients themselves — and survive because the choice to make them open meant no single person's departure could take them away.
