Loop questions

Continuing the discussion from Dexcom accuracy:

Hi Jacob - I split this topic off from the previous thread so I don’t stray too far off topic there. Here are two threads I wrote about Loop, one when I started in November 2016 and one at the one-year mark in November 2017.

I have been happy with using Loop. I would not willingly go without it. I also low-carb but probably in the 50-75 grams/day, not as strict as the Berstein 6-12-12 routine.

For exercise I will set Loop at a higher target, 95-105 mg/dL (5.3-5.8) about 60 minutes before I start the exercise. My usual Loop target is 83-84 mg/dL (4.6). My preferred exercise routine is walking, so I likely would need to add carbs and/or set a higher BG target for a more aggressive cardio workout.

I use this USB 6-port charger and charge my iPhone, RileyLink, and Dexcom receiver as I sleep.

Since the RileyLink uses wireless tech to communicate, I just lay it on the bed nearby and it communicates quite well. The weak link is between the pump and the RL so I try to keep those two near each other. I use a 43" pump infusion set which makes this relatively easy.


Still haven’t left the ‘want to, but thinking about the logistics’ stage…

1 Like

In reading through the current version of the Loop documents, I noticed two different Loop versions being explained. They are 1.5.9 and 1.9.
Can anyone say for sure that these are not the same version?
I think it would be a good idea to have dates on the different sections of documentation, and notes to indicate master vs development versions with the dates. That way, descriptions of development versions would not be confused with current master versions.

According to this Loopkit GitHub site, there is no master version 1.5.9 but there is a 1.5.6. Version 1.9 is a separate version. The latest version is 1.9.3, released 11 days ago. This links leads to the official list of master releases along with some descriptive narrative.

I’ve now discovered some conflicting info on the LoopDocs:

According to this screen-shot, there was an official master version of 1.5.9. Now I’m not so sure of my answer. Maybe @Dragan1 could help clear this up.

Development versions are not on this list. Anyone can modify the source code to make custom changes but their modifications do not necessarily make it into the master version. You can follow comments about some of these development versions at the Facebook Looped group. This is a closed group, but joining is easy.

There is also a more technical Loop community that discusses things like this. It’s a Gitter chat room.

Thank you, Terry.
My conclusion is that the version 1.5.9 reference under “Apple Computer” is an error.
Do you know if the building of v1.9.3 is the same as for v1.9? I didn’t see v1.9.3 mentioned anywhere in the documentation.
I expect to try it soon.

I don’t know for sure. I’m all the way back at V1.5.2.

The LoopDocs does point to the GitHub site for more info on version releases. LoopDocs are written/maintained by Katie DiSimone. About a week ago, it was announced that she is accepting employment with the non-profit, TidePool. TidePool intends to develop a version of Loop that will be FDA approved and could work with current under-warranty pumps. That might account for the LoopDoc update delay. Or perhaps LoopDocs do not update for minor version updates.

Thanks for all your explanation. I haven’t had much luck finding information on github itself. I also tend to avoid Facebook from a security standpoint. But it looks like I’ll need to change my behavior.
I hope that Katie DiSimone’s move to TidePool might mean that they are nearing submittal to FDA, and need to get their documentation in order.
If TidePool can get a version out that can work with the currently locked-out pumps, they could have a big winner! I would certainly jump for it.

This effort was just initiated last week. I’ve seen no estimates as to how long this will take but I’d definitely not put this in the near-term. It will be interesting to see the current pump manufacturers’ decision on this opportunity for them. I know the Dana pumps made by South Korean company SOOIL have already released pumps with open source communication protocols. They’re sold in Europe but not yet in the US.

It’s possible that pump manufacturers could concentrate on their hardware and leave the software selection up to a consumer who could then select from a list of different algorithms to install on a new pump. At least that’s the concept I’m gleaning from all the coverage in this last week.

Have you looked at this Loopkit Gitter chat site yet?

At the time you asked, I had never seen that chat site. It has reawakened my interest in learning the Swift language. That site can allow me to follow the debug process for new features of dev versions of Loop.
I don’t think that FDA has an attitude of leaving any choices to the user. Weren’t they the reason that Medtronic shut down rf access to their pump software?
Probably, any pump sold in the USA would have to be non-communicative.

The recent new designation that the FDA awarded to the Dexcom G6 paved the regulatory path for this new modularity. I forget the name the FDA gave it; it was iXXX something. The FDA will still control all parts of the process.

The FDA designation is iCGM for “integrated CGM.” This should streamline the process for all entities that follow in Dexcom’s steps.

That sounds promising.
Thank you for raising my spirits, and possibly lowering my a1c.

1 Like

@Terry4, @DA1958, the 1.5.9 version on the Loop docs Apple Computer page is a typo, should have been 1.5.6. Furthermore, that particular page is just slightly behind the current Loop release, which is 1.9.3. The best place to look at the most up-to-date summary of the current Loop version and version history, together with a list of new features included in each release, is the Loop release page on GitHub. As @Terry4 mentioned, Loop gitter chat room is the best place for technical questions and issues. BTW, we do need more Swift programmers!