Google+ Badge

Friday, September 27, 2013

Finally I get to do a tablet only app

We have a dashboard style web page used by our clients. I recently converted the charting side of that from flot to HighCharts. HighCharts gives much nicer looking charts by a long shot and the price was reasonable. All of that work was in JavaScript which I must say I don't miss and am very happy to get back into the mobile side of things.

JavaScript is just barely acceptable on an iPad or Android Tablet. Yes, I tweaked various UI aspects to make them more touch friendly but if we want to be able to look a client in the eye and tell them we have a dashboard that works on their tablet we need to go true native mobile.

All the code I have done in the past has been universal phone and tablet. For this type of dashboard we are going to start tablet only. We really want to design to use the space available. Of course 1024x768 (older iPad and iPad mini current maximum) is not a ton of space but it is what a number of folks run on even older desktop PCs if they only have a 15" LCD.

At this point I am sketching out ideas and have started a framework on the Android side. I am going to start there as this seems a perfect fit for fragments within an activity. Previously I have used fragments to learn about them and knowing it is the proper way to approach the future. Now I will get to exploit so much more of their power. I have not had the need to swap in one fragment for another or to add new ones to an activity, I have only defined them in the XML and let them operate as is.

Since this will target 3.0 and above on large devices I don't even have to worry about the compatibility library. Nice to get to be able to fully use the newer API features.

I am using achartengine for the pie, line and bar charts on the Android side and core-plot on the iOS side of things. The current smaller app just displays a pie chart so I will need to learn my way around the others.

Not sure what approach I will take on the iOS side of things. There is nothing like fragments there so I will get to deal with UI views and manual placement of things. This is why I plan to get it up and running on Android first so I can knock out the initial design without fighting layout issues. I should be able to figure out a number of UI patterns such tab controls, menu options, widget manipulation buttons etc. and run them by beta testers before I begin the iOS conversion.