Topic: Upcoming app for the GW1000 - and some advice from you

[Da9L]

Hi everyone!

Here in Denmark we have been adviced to stay at home as much as possible due to the corona virus, so I thought I might as well begin on building the app for the GW1000 i've been wanting to create ever since i first got it, and its coming along nicely i think. I plan on releasing it soon and might need some beta testers. Anyone interested? Since Ecowitt haven't released the API yet, i have had to get the data to the app the other way around, and setup the Customized  Weather service in the app.

This is how it looks at the moment:
 [ You are not allowed to view attachments ]

You will be able to set what sensors you want to see graphs from and if you want to see live graphs from one or two sensors at a time, and for how long the timespan of said graphs should be stretched over in the view, and also create custom alerts for when sensors reach a certain value etc. More on that in another post when i have tackled the below.

Because i do have a question for you in here, since I'm a little lost at what to do, and wanted to hear from people actually using this daily what you want. You can furthermure see a graph of each sensor, by tapping on its value.

 [ You are not allowed to view attachments ]

BUT due to memory limitations, i cannot store each update from the GW1000 as that would pretty quickly slow down the app and cause bad performance. Therefore my initial plan was to only store one update from the GW1000 every 10 minutes for history graphs. But this leaves a problem: Say that you want to see how cold it has been during a night, and the temperature has been down to 25F during the night. However this happened in between two of the saved updates from the GW1000 and in those two saved updates, the temperature was only down to 26F. This would result in you not seing the actual lowest point of the night.

How do you think i should go about doing this? I could just make an average of all the updates in between the 10 minutes and store that as one, if that makes sense, but this would still not show the actual lowest.

And also this could also be the other way around. If you want to see the highest temperature one day.

Hope to get some feedback from you!

[Rover1822]

Very nice , impressive

If you are looking to pass this out, there are so may ways to do so, frankly not worth your effort. Or sustained maintenance effort, but it looks awesome.

 Go for what you want, because in the end that is your most important client. And you obviously have the talent.

As far as data storage, you really need to look at database solutions for whatever language and platform you are dealing with. if you are not meeting your goals, you have to examine everything , platform, hardware etc. You seem to have reached a plateau where you are looking for direction, gosh, there are so many possibilities at the stage you are at.

What is your ultimate goal?




 

[Da9L]

Very nice , impressive

If you are looking to pass this out, there are so may ways to do so, frankly not worth your effort. Or sustained maintenance effort, but it looks awesome.

 Go for what you want, because in the end that is your most important client. And you obviously have the talent.

As far as data storage, you really need to look at database solutions for whatever language and platform you are dealing with. if you are not meeting your goals, you have to examine everything , platform, hardware etc. You seem to have reached a plateau where you are looking for direction, gosh, there are so many possibilities at the stage you are at.

What is your ultimate goal?

Thanks for the kind words! Honestly, my initial goal was actually just to build an app, as i haven't done that before and wanted to try it out. I thought that the GW1000 was a perfect opportunity to build something for. I had a tablet i wanted to use as a permanent display for the GW1000 weather data. The GW1000live app was what i used to begin with, and its really good actually, but that still didn't satisfy my need for creating my own app, so here we are. It more or less took shape as the ideas came along while i was coding it. First i only wanted a graph. Then i wanted two graphs, Then i wanted to be able to select what sensor the graphs should represent. Then i wanted alerts of the data etc.

So i guess the ultimate goal would be for the app to act as a permanent display for the people, like me, having a tablet they don't use anymore, and want to make useful again.

I have no plans on requiring any accounts for cloud storage or anything like that. Right now all the history is stored in the app and memory, which also means that if the app closes, the history is lost, which is of course not ideal. So for that i would need some more permanent storage and, as you suggest, a database would be the right way to go. I will have to look into what kind one normally uses in an app. I also think that might make everything easier, as i would be able to store all the data and not have to worry about performance in the same way as i do now.