I saw some amazing analysis from some of the folks in the group Geoff mentions. It was calm, objective and measured, and didn't attempt to draw conclusions that

I saw some amazing analysis from some of the folks in the group Geoff mentions. It was calm, objective and measured, and didn't attempt to draw conclusions that couldn't be backed by observable fact. Disappointingly, there was also a class of analysis (coming from a smaller part of the group) that was based conjecture or guesswork that fuelled a narrative of COVIDSafe that was wrong or inaccurate. I saw this bring out both the best side of the tech community (methodical, analytical, evidence-based, objective) and the worst (conspiracy-driven, inflammatory, chaotic) that was dejecting. I felt the former was committed to making conclusions from observations, whereas the latter made observations to fit the conclusions they wanted to reach. Doing so in the public domain alongside legitimate analysis simply gave credibility to unfounded and baseless conclusions. Geoff's analysis says: >These problems would not have existed had Australia chosen the Exposure Notification Framework (GAEN) path. however this would never have helped out contact tracing, which he says is his motivation: >Medical professionals (contact tracers) deserve technical tools that work. Forcing them (via political pressure) to use ineffective tools robs them of precious time which could be better spent contact tracing. This is disingenous and trivialises the situation. One of the value propositions behind the BlueTrace stack was to provide greater visibility to contact tracers, by giving them access to the data in cases of close contacts. The app development was in full swing well before Apple & Google had established their notification protocol. Using BlueTrace was a smart move at the time, as it leveraged a model that had already been rolled out to Singapore. Of course, there were/are obvious challenges with this model (e.g. the unreliability of iOS) - but at the time, there was a true need to get something out, and to their credit this was turned around exceptionally fast in the absence of anything else. Chucking it out and starting again simply because GAEN was released would have been a misfire, particularly where it could have been accommodated by a future release. The shift away from BlueTrace to something like GAEN would have represented a significant shift away from the fundamental promise of the system - that contact tracers get access to the contact data. Whether you agree with that decision or not, that was COVIDSafe's policy objective. There are/were many legitimate issues with COVIDSafe that made the product (in my opinion) non-viable: the combination of the definition of a close contact was "1.5m + 15 min exposure time" with the means of recording encounters made using it all the time nonsensical, but also seems inconsistent with how we know COVID spreads (well, especially the delta strain). Many fleeting or substantial encounters would be completely missed, because they didn't satisfy that requirement, even though we would be legitimately concerned about them now. Add to that fact that deriving distance from BT strength is something that's challenging to do right and is only so accurate if done perfectly. But harder than that - recording "15 mins" of exposure means what exactly? Continuous exposure at what interval? xposure at the 1st and 15th min? How many inbetween? GAEN may have solved much of that problem, but at the cost of data for contact tracers. In my mind, simple beats clever, which is exactly where the QR check-in apps have flourished.