Proposal: Usability Improvements for Juggluco with React Native, Starting with Freestyle Libre 2

[BoulderBuild]

I’m interested in contributing to Juggluco by proposing usability improvements to enhance the user experience, particularly for Android users. My goal is to make the app more intuitive and user-friendly, and I’d like to explore implementing these changes using React Native for the front-end, leveraging its component-based architecture and rich ecosystem of UI libraries. However, I have some questions about the feasibility of this approach, especially regarding the current stack and sensor compatibility.

However, I’m unsure if React Native is compatible with Juggluco’s current stack, particularly for handling Bluetooth communication with the Libre 2 sensor. The repository’s documentation mentions native Android code for sensor integration, so I’d like to discuss:

Is the current stack tightly coupled to sensor logic, or could React Native interface with existing native modules?
Are there known challenges with React Native’s Bluetooth libraries (e.g., react-native-ble) for Libre 2 sensors?
Would a hybrid approach (React Native for UI, native code for sensor communication) be feasible?

I’m happy to start with a proof-of-concept for the Libre 2 UI improvements in React Native and share it for feedback. If this aligns with the project’s goals, I can create a more detailed plan or contribute directly via pull requests.

Looking forward to your input!

[j-kaltes]

Let I now like the current design and approach.

EDIT:
Isn't React Native JavaScript? Can’t you make a web browser script that receives its data from the web server in Juggluco and makes that people can easily see that on their computer.
The webserver in Juggluco makes it possible to receive all types of data as numbers (http://127.0.0.1:17580/x/stream?header) or you can receive an image of the curve (http://127.0.0.1:17580/x/curve?start=2025-05-30T12:12&days=0.5&darkmode) or statistics (http://127.0.0.1:17580/x/report), but there isn’t much of an interface.

See: http://www.juggluco.nl/Juggluco/webserver.html

[BoulderBuild]

I would like to do it on a mobile device instead of a computer, I want to test different interfaces with different people from different ages to see what could suit them better. Most of people with diabetes type II are old. They treat a screen like an old fashioned button. So i would like to know if there is a way you could give me a tiny approach of how the interaction with .so files is? I saw the scan steps, how it recognizes each sensor and how it proceeds with Libre 2 in case 3 or Libre 3. I would like to see if I could replicate this and then be able to test with non tech people. I know your app can be easy once you figure it out. But there is people who won't be able to figure it out and most of them don't even have people around them to help them either, considering their age... So, I would like to see if i could try something and see results and perceptions from users. I know your app is good and does all the job.

[jrdeutsch]

I think you should give BoulderBuild a shot. The current UI isn't very intuitive for non-technical people.

…

On Mon, Jun 2, 2025 at 4:44 AM Jaap Korthals Altes ***@***.***> wrote: Let I now like the current design and approach. — Reply to this email directly, view it on GitHub <#304 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AD5XLC4TCWISBL6R3H5SYID3BQFHZAVCNFSM6AAAAAB6LXN6Y6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGMZUGA4TKMA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[j-kaltes]

It is very intuitive, but unusual.

[jrdeutsch]

In UI/UX, "unusual" also means it "lacks affordances" of common UI interaction design. It's quite easy to use once you figure it out, but figuring it out initially is a significant barrier to entry for some people. Ideally, there would be an "easy mode" UI for people who are less technical and just want to use an app that looks and operates like most of their other apps do. But, in the end, it's your app -- and a great app! -- and you get to decide who you want to use it!

…

On Mon, Jun 2, 2025 at 8:40 AM Jaap Korthals Altes ***@***.***> wrote: It is very intuitive, but unusual. — Reply to this email directly, view it on GitHub <#304 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AD5XLC6O25ZTN5MUXGQ4HL33BRA4HAVCNFSM6AAAAAB6LXN6Y6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGMZUGMZTANA> . You are receiving this because you commented.Message ID: ***@***.***>

[j-kaltes]

Touching the screen to open a menu, has nothing to do with having a technical nature. Maybe it is difficult for people who don't tryout things and don't read instructions, but that is only the first time.

Because of the many menu's it is easy to find something in Juggluco and that is difficult in other diabetes apps I tried.

[jrdeutsch]

From my POV (doing GUIs on the web for over 20 years), there are two things that make Juggluco difficult for the "normal person" who is used to using a Smart Phone: 1. Virtually all utility and productivity apps default to portrait mode. Forcing landscape mode makes Juggluco an outlier and that alone creates some confusion and lack of understanding of how to use the app. (fun fact: when I first tried it years ago, I immediately un-installed it thinking that the forced landscape meant it wasn't compatible with my phone!) 2. The best GUIs offer the user visual cues as to where to click to get certain options. Often in mobile app design, it's a hamburger menu, but there are many different ways to accomplish this. Tapping on different parts of the screen for different menus is novel and interesting, but so far away from mainsteam app functionality that it can certainly confuse people who aren't interested in learning a new UI schema. Hopefully this perspective is helpful. Again, I'm fine with the app as-is for me, personally, but I know many diabetics who won't try the app due to its unconventional GUI. Though I try hard to provide training when I can. :)

…

On Mon, Jun 2, 2025 at 9:00 AM Jaap Korthals Altes ***@***.***> wrote: Touching the screen to open a menu, has nothing to do with technicality. Maybe it is difficult for people who don't tryout things. Because of the many menu's it is easy to find something in Juggluco and that is difficult in other diabetes apps I tried. — Reply to this email directly, view it on GitHub <#304 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AD5XLC4ZDG4VBNVNLRACLSL3BRDGNAVCNFSM6AAAAAB6LXN6Y6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGMZUGM2TEMI> . You are receiving this because you commented.Message ID: ***@***.***>

[j-kaltes]

1. Displaying the glucose value on the short axis and the time in the long axis is the best way to display the glucose curve.
   Android apps will automatically switch to portrait mode when the app allows it and Android works in such away that user will end up using it in portrait mode.
   I just can’t stand the idea that users would get the impression of Juggluco was they would have when used in portrait mode, as I have seen people use other apps.
2. I was rather impressed by the interface of Koreader.

[ChrissHF]

I really like the app, as it works far better than the FreeStyle Libre App.
But a UI overhaul for portrait mode would be great. Do you think it would be that bad?
We don't have to get rid of a landscape view. Maybe it could switch to landscape when you click the graph, or a toggle in settings.

I made a quick mockup (in Paint, please; don't hate me) as I would imagine a good portrait mode could look like.

BTW: do you have a PayPal or something to support you? I really appreciate the work and improvements you put in.

[j-kaltes]

On a computer or tablet screen you could do something like that. You could remove some of the images from the report you can get out the the webserver of Juggluco, see: https://www.juggluco.nl/Juggluco/webserver.html
report.pdf

[ChrissHF]

Sure, but the app on the phone is what most people want to use, not a website on a computer.
Also, that wouldn't really help with putting in meals, insulin amounts, etc.

Not to be rude, but of course you find it intuitive to use, as the developer. You know where you put what setting and button.
But as a random user, it's quite tricky to find what you are looking for. I sometimes still have to search for the right button/setting in all menus after almost 3 years of use.

This app would be so helpful for a lot of people, but I think many, especially "non-techies", find it very hard to use and rather stick to a buggier but simpler app instead of trying to understand juggluco.

[j-kaltes]

I didn't meant to say, that you can't do it on a phone. You can easily do just turn on the web server in Juggluco and press left middle menu->Statistics->Save. I meant to say that displaying it that way on a phone doesn't make sence, because you can't see the curve very well that way.
On my computer screen this is displayed at the same size as my smartphone: https://www.juggluco.nl/Juggluco/portrait
Only people who don't really look at the glucose curve, can want something like that.

[jrdeutsch]

For what it's worth, I agree with ChrissHF: The current UI is not intuitive for the average, normal, non-technical user. A more traditional portrait UI would probably benefit more people. However, the developer actually may *not* want to support that many people?? Something to think about....

…

On Thu, Jun 19, 2025 at 9:52 AM ChrissHF ***@***.***> wrote: Sure, but the app on the phone is what most people want to use, not a website on a computer. Also, that wouldn't really help with putting in meals, insulin amounts, etc. Not to be rude, but of course you find it intuitive to use, as the developer. You know where you put what setting and button. But as a random user, it's quite tricky to find what you are looking for. I sometimes still have to search for the right button/setting in all menus after almost 3 years of use. This app would be so helpful for a lot of people, but I think many, especially "non-techies", find it very hard to use and rather stick to a buggier but simpler app instead of trying to understand juggluco. — Reply to this email directly, view it on GitHub <#304 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AD5XLCZ3A3SY7DN4T2PF6HT3EK6B7AVCNFSM6AAAAAB6LXN6Y6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGNJSGA3DGMY> . You are receiving this because you commented.Message ID: ***@***.***>

[ChrissHF]

Yes sure, I get that. It would be extra work to deal with, and that j-kaltes may not have or want to spend more time on this.

Unfortunately, I don't really have the skill to help with development, but BoulderBuild offered his help, and I'm wondering why j-kaltes seemes so anti portrait mode in his comments.

Currently, the only thing I can do is to bring in some ideas and feedback and maybe donate a few Euros,
That's why I asked for a PayPal, I'm sure lots of people would happily support this app with a few donations :)

[FuLiDu]

Yes sure, I get that. It would be extra work to deal with, and that j-kaltes may not have or want to spend more time on this.

Unfortunately, I don't really have the skill to help with development, but BoulderBuild offered his help, and I'm wondering why j-kaltes seemes so anti portrait mode in his comments.

Currently, the only thing I can do is to bring in some ideas and feedback and maybe donate a few Euros, That's why I asked for a PayPal, I'm sure lots of people would happily support this app with a few donations :)

I agree 100%.