Touche! However, at least I CAN mess around with xdrip... if this same situation occurred, I'd just fix it, build, and put it on my phone. The problem would have been solved within 24-48 hours
All kidding aside, one interesting thing I'm finding with xdrip is it seems to be more accurate tham Dexcom's vaunted 505 firmware, and requires less calibration to stay accurate. How can that be?
I mean, the developers working on various pieces and parts of the We're Not Waiting movement are simply hacks like me -- not PhD physiologists with major degrees in biochemistry or something.
Yet, I get better results with xdrip (i.e. more consistent with my Freestyle meter) than I do with the Dexcom receiver. Further, I only calibrate once a day now, as that's consistently been more than good enough for the several weeks I've been using it. So, while we wait for the next generation and the awesome ability to only calibrate every 24 hours, seems the technology is already there with the G5, at least for me, with the right software interpreting the data from the sensor.
IMO, if Dexcom had a modicum of business intelligence they'd be on the developers of xdrip like white on rice, hiring them outright, or getting into some sort of formal contract arrangement to bring the technology into their product.
Too often pride gets in the way of innovation. Truly smart companies don't let this sort of thing get in the way.
While Dexcom seems to be ignoring this effort (as far as I can tell), Apple was very smart and hired the creator of Loop (one of the open-source APS projects).