Nov 3, 2011 at 1:48 PM
As suggested by Sergei:
As there can be different approaches to the planning, I would suggest to write a list of all code parts which need to be rewriting, as well as the new features to be implemented.
The list can hold:
- a brief desription;
- estimated time on implementation;
- necessary preconditions (tasks to be implemented first).
Then some kind of formalized procedure can be used to choose priorities.
For example a voting with setting a value from 1 to 5. After we can choose several tasks with the largest number of points for the next development cycle.
For example, from my point of view very important things to make the project more
attractive to the developers and users are:
1. Writing documentation for MapWinGIS (including intellisense in .NET).
2. Localization of the GUI (symbology, projections, new GIS tools).
3. Writing demo application showing the usage of the new code (C#).
4. Merging legend control and symbology plugin in a single assembly. Adding new layer properties dialog for image and grid layers.
5. Making new MapWinGIS installer with additional NET assemblies (Tiles, Legend) and documentation.
6. Find some good solution for repository structure (move more code to Codeplex).
7. Fixing bugs that will be found during the next period of usage.
To make the further changes easier, the following can be added:
8. Moving some code to MapWinGis from MW4: scalebar, grid manipulations, etc.
9. Cleaning MW4.exe and MapWinGIS from obsolete code.
If we add tiles support to this list (through GMap), it can be a good TODO list for the next version I suppose.
Nov 3, 2011 at 1:53 PM
I agree with Sergei.
I think his proposal looks good.
We can already start with copying the ocx-code to mapwingis.codeplex.com and start cleaning it up.
After all old code is removed we can start with the documentation of the ocx.
Sergei already started with an approach to create XML files that will be used by intellisense.
We should continue with that.