puSSH | The cross platform , multi-functional screenshot utility | Image Editing library

This, in-of-itself, is not a easy thing to do and recognising when you should comes a lot from experience (oh, I remember last time I did this, it was horrible to manage and maintain ), but, there is a lot of available patterns which are used everyday to make software development easier and solve everyday common problems.

One thing you need to keep in mind is who has the responsibility to do what. For example, it might not be possible to seperate the action listeners of the buttons, as they need to perform operations which are internal to the UI itself.

However, you could make it easier through the use of Anonymous Classes or even the Actions API, which allows to isolate the functionality of the button

Another thing I would look at is isolating all the seperate containers (panels) to their own classes. This isolates the functionality and reduces the amount of complexity as it forces you to think about how each sub container is going to work and what it becomes responsible for as well as reducing unnecessary access from external influences.

A more complex solution would have the UI rely on a "model", which separated from the UI. The UI would then take actions from the user and update the model, which would intern generate notifications that the UI would use to update itself to reflect the changes, see Model-View-Controller for more details

So, what's the answer?

• Break your data apart from the UI. Use one or more container/model classes to represent the data. Make use of the observer pattern to allow the model to generate events when changes occur so interested parties can take appropriate action
• Break down your UI into "usable" components, where each component is isolated and responsible for managing a single portion of the UI (and how that happens is it's own business)
• I would also recommend making use of dependency injection to share objects between the separated elements, this will allow you to isolate and test individual areas of your code, bonus, which also leads into...
• Test Driven Development (TDD). Know what and how you want to test your code before you try and design it

You need to read about INotifyPropertyChanged interface, it's an interface your ViewModels and Models needs to implement if you want your Binding to be updated when the ViewModel/Model properties changes. Once you have a basic understanding of how INotifyPropertyChanged work, then you'll have to decide if you add Lat and Lon properties directly to your ViewModel or if you make a second ViewModel for your RootObject and bind to it.

Also, a good practice is to set the DataContext to the root element of the page (ie: Window), the DataContext is inherited by children if not specified.

Unfortunately, without INotifyPropertyChanged, you won't be able to make your Binding show any updated value to Lat and Lon.

Note that your RootObject a is being destroyed at the end of ExecuteMyMethod at the moment, you'll probably want to change your logic here to keep a reference to the a.Coord value somehow.

The issue is presumably that you have got that first code snippet at the class level. In that case, it will be executed before the constructor, in which case your controls haven't been created yet. What you need to do is declare the array variable at the class level: