Don't know what to make of this

http://www.massdevice.com/news/diabetes-certain-jj-insulin-pumps-destined-fail-over-software-issue

I think it is one way of getting the older pumps replaced with newer models.
I did get a letter from
Animas telling me the same thing but I had already upgraded to a ping.

Thanks. I am a recent Ping user and didn't know that Animas pumps were not the same as the Ping.

The ping is the newest one. The 2020 and IR models are the older pumps.
If you have a Ping you should not have to worry.

Hi Robert. Really? I didn't know this. When I see articles like that, I tend to think it's a rumor; it just didn't register in my head that an insulin pump would be made to one day not function. we have to be so careful what we read these days. I think it's really messed up that Animas would knowingly do that, especially since so many don't have insurance. I'm sure if they really wanted to, they can offer an "Firmware update".

-Yogi

I got a letter from Animas in mid December. I think everyone affected by this (ie models before the Ping) should have heard from Animas by now. AS there is another 2 years before the pump stops working I am not too worried.

Contrary to what others have speculated I believe this was a genuine software error and not a design feature. I have written enough software to know that these sorts of errors can incur, a base date is picked & then dates are stored as so many days after the base date, eventually this number gets too big to fit in the area allocated to store it. It is an error that shouldn't happen but it does.

But why have code with a base date at all? Write the code and leave out sudden death subroutines. They serve no legitimate purpose!

FDA does not think "patients" are competent to do firmware or software updates.

The reason for a base date is that the date is converted to the number of days since the base date, & this is the number that is stored in the computer (or pump in this case) memory, & this number is used for all computations involving dates, like IOB calculations, the time is the number of seconds since midnight. Unfortunately the area in memory where these numbers are stored is finite so if the number of days gets too large it wont store properly. So there is no sudden death subroutine just a miscalculation by the original code designer on how long the pump should function.