I think it’s not that they necessarily want this to be what happens but more that they haven’t currently explicitly written code to show something different if there’s stale data. The number updates if there is a new CGM datapoint but otherwise does not change. When I was deciding between loop and iAPS I was advised that Loop is prettier but carb-centric and iAPS is uglier/hack-ier but less carb-centric, this is just a one of many instances of the app not being as polished
4 Likes
Cool, thanks for the distinction. I appreciate that because I am really not very aware of all the versions and differences between them!
2 Likes
@Eric ”They” boils down to one primary developer and a few others that can program and have influence, all of which have different priorities and innovate on their own dime and time, i.e. no funding stream and presumably with their own primary jobs. Given that, I can’t complain too much and probably wouldn’t have much effect. I, like you, would appreciate a known systematic method that prioritizes functions, functional reviews, and documentation. I hope and believe the “they” people working on Open-iAPS do value a more systematic approach, though they too are operating on their own dime and time. I hope they continue their efforts as long as the hardware, need and desire exists. Would that I had the skill set to assist…
1 Like
@TomH, yes I too appreciate the efforts they have made and what they have done. And I understand that it is done without any funding.
I am not bashing them or complaining.
I was posting my initial observation as a possible warning, in case people did not know the stale CGM number was not being updated.
And my second comment was just out of surprise that it performed this way, since the older (FreeAPS) version correctly makes a stale CGM number disappear.
But I know, I have not paid anything for it. So I have no room to complain.
If I can get G7 working on my version, I will be super happy.
1 Like
@Eric I didn’t mean to imply anything otherwise, and know you have the best regard for efforts to improve. Sorry if it was taken otherwise. I made the comments to ensure others reading them understood the situation. I wasn’t aware an older version of iAPS did the right thing; I can’t imagine making a change to be less informative about a disconnected/faulty device, but know a tweak in one area can have unintended/checked for consequences elsewhere due to variable usage. I used to write some C code to modify my WWIV BBS (The State Pen) back in the late 80s/early 90s; seems archaic by today’s standards.
Ref the G7, I seem to remember you already looked at the current Loop code to see if the G7 references there would help out. If not, that might be helpful to your efforts (I presume its marked and all in one place, but haven’t looked myself.)
2 Likes
Actually no, it is my older (and customized) version of FreeAPS that does the right thing. My apologies, I did not clarify that in my comments above.
I have only tried out that single version of iAPS, and I only got it to try and get an idea of how to implement the G7 stuff into mine. I have no idea what any of the other versions of iAPS do!
(I just edited my comment above for clarity. It gets confusing when there are many versions being discussed.)
1 Like
@Eric Thanks for the clarification on which program had the correct notification. You’re right, it’s easy to get confused with the various versions of the various programs (Loop, iAPS, AAPS, FreeAPS, FreeAPSX, and others…); I’ve done it more than once! With your info, I’ll propose the iAPS/Open-iAPS developers both address the situation to ensure people know when their devices aren’t connected.
2 Likes
There are obviously a lot of moving parts!
Obviously I’m missing a lot of the background on it all. But what I don’t understand is why FreeAPS went away. 
The FreeAPS fork has been frozen for a while. The original fork developed by Ivan Valkou.
2 Likes
@Eric Wish I could offer more, but I never used FreeAPS. My understanding is two Russians (one is the Valkou @ClaudnDaye refers to) forked an older version of Loop, altered it significantly, called it FreeAPS. Eventually, it was altered/forked again and became FreeAPSx. Then Putin invaded Ukraine, the US/other’s sanctions reportedly cut off obvious work with the two Russians, so others reforked FreeAPSx into what became iAPS. I’m sure there’s a lot more to the story…
You may already be aware of it, but incase not, the LnL site (FreeAPS Fork | Loop and Learn) has a page with a link to build what may be the last FreeAPS app that purports to have Dash pump support (seems to require a Mac with Terminal and presumably works). There appears to be quite a bit of information there.
3 Likes
Thanks @TomH.
I looked at that most recent FreeAPS version. It does not have G7. That’s pretty much all I am looking for at this point.
But that page does give a good bit of information, so thanks for that!
2 Likes
What assurance is there a fork hasn’t been altered to an unsafe state?
If you are ever looking to cross over to the dark side of the force and try Android APS I can help getting that set up. Pretty sure G7 is incorporated although I’m still running G6.
I also am looking to maintain this setup as there are no additional bells or whistles that would help my blood sugar management.
2 Likes
@jim26 You’re right, there is little, other than anecdotal reporting of issues, to assure un-altered code! This is one of the reasons why the safety of iAPS and the primary developer’s use of partially hidden code was questioned and why some began the Open-iAPS effort. All non-programmer users of iAPS, Loop, AAPS, etc., rely on “reasonable” testing to assure functionality, but there’s still a significant degree of trust the code is nothing more or different from that represented. Of course, the same can be said of commercial products even with FDA/CE or other clearance or certification. The code that gets submitted gets checked, but there’s little guarantee the code compiled/built is the same as that reviewed…unless there’s some level of review I’m not aware of. When it comes to software/firmware, all products depend on reliable honest forthright developers/manufacturers. Even then, there are likely backdoors, trapdoors, and Easter Eggs that get added prior to or after the review process. We’ll only “know” when and if they get revealed.
3 Likes
Thanks Tom,
Yep, that would be concerning especially when you essentially have anonymous developers.
1 Like
I should have added “No one thinks there was anything ‘nefarious’ going on, just a bit odd…” The developer seems to have done a lot for the community, but communication isn’t a strong suit and it’s important given the impact.
2 Likes
Yep, life happens and may not even have the time or desire to make needed changes.
FYI, the folks putting Open-iAPS together apparently have decided to change the name to “Trio” and release a version on Github for beta testing last night. I think the name change is good to avoid confusion of the two products (iAPS and Open-iAPS/Trio). For those interested, you can read a bit more here: iAPS (Soon to be Trio) | Facebook. I plan to monitor Trio for now, vice switching in the middle of trying iAPS.
On my personal efforts trying iAPS (v2.3.3), I’m still playing with a couple of settings to refine/simplify my efforts. I have Dynamic CR turned on and it seems to help my control. I found turning on DynamicISF would send me low consistently and have turned it back off; I need to track down the why and a fix or just leave it off. I’ve also learned using the iAPS ability to dose for Fat/Protein with various associated settings helps a great deal in avoiding post meal spikes in the morning and most lunch/dinner meals, but not all (I haven’t figured out a common cause). The devs of iAPS instituted a version of the Warsaw FPU dosing that allows for an initial delay period, maximum duration, interval dosing control, and a settable override factor. For those not aware, the Warsaw method is known historically for driving people low, hence the various adjustments. While I’m still refining my settings, I was able to eat an entire Costco gluten free pizza (12-14") with my own addition of pepperoni, onion, and a little mozzarella…but the FPU settings resulted in in no drastic or incremental significant rise! I can’t say the same for an impromptu small piece of my grandson’s b-day ice cream cake. I didn’t have the particulars for dosing for it and estimated; it seemed well controlled at first, but after going to bed I jumped to between 225-265 and stayed there much of the night…iAPS wasn’t a match for it and it’s the first time I’ve had a pod run out of insulin early (iAPS kept giving me significant SMB’s much of the night). The adventure continues…
4 Likes
This is all super interesting to me. I just need my eyes to stop pouring outta my head so I can read some more. I am seeking an app that I can say don;t give me even a little bit of insulin below XX amount. I want to control it. This is super! Thank you!! Certainly want to investigate more. 
1 Like
This looks great with the Start Up Guide HERE
Problem is you still need an Apple Program License. Until January 2024 I had that License & was running Loop Vs 3.2.3 til Apple kicked me ro the curb.
Instead of going Off Topic on your great thread I am going to make a RANT thread where I will also ask some questions. Thanks again @ TomH
1 Like