Ancient Dragon
Retired & Loving It
30,040 posts since Aug 2005
Reputation Points: 5,662
Solved Threads: 2,341
and you may as well consider some time management.
if you have a deadline, a lot of your stressy reactions may come of you worrying whether or not you'll be able to meet that deadline.
draw lines for yourself: what is the most important? what is the most difficult? what is the easiest way to 'show' your employer you are making progress?
it's a bad idea trying to do it all at the same time. start with the back-end, with some mock classes to test your classes and methods. sure, it may look like you're not making progress for someone who isn't really paying attention, but when you start with your GUI, it' ll look as if you've finished everything in a matter of days.
AncientDragon's advise to use a version control system is a golden tip, I would take it if I were you :)
also, make sure you give your packages logical names, that may help you navigate faster/easier between classes. I can assure you, if you enter in a project where a package name is something like 'brkv', it'll take a while before you know where to find everything.
stultuske
Posting Sensei
3,108 posts since Jan 2007
Reputation Points: 1,114
Solved Threads: 432
We used Visual Studio where I worked which did a great job of organizing large projects that contained several related executable programs, libraries and dlls (all called a "solution"). You can set up dependencies so that when you ask VS to build the entire solution (meaning compile all libraries, dlls, and executables in the solution) the IDE will know when to bild what and in what order they need to be compiled.
Ancient Dragon
Retired & Loving It
30,040 posts since Aug 2005
Reputation Points: 5,662
Solved Threads: 2,341
As I already mentioned that I am working alone on this project, so I don't think that using subversion is going to give any help unless atleast one more person is involved in it.Although I am still using svn for UI of it in case anyone else also get involved in it.
wrong thinking...
what if you decide to change something quite radical, in theory the invention of 'wheel 2.0', but in practice, turns out not to work at all. do you want to look over every line of code that you (may) have altered rolling back, or do you just want to 'click> replace by previous version'? it'll save you a lot of time, and it ensures certain versions to which you can return.I don't have a very specific deadline but want it to get done as soon as possible that is the reason why I am working on all things simultaneously just to feel like I am getting some progress. Ok so working with back end is the first thing that should be done.
without your back-end, your front end doesn't work, it's a pretty box with nothing in it.
the most of your functionality, so also the most of your difficult work is in the back end. doing all together doesn't mean you proceed faster, it just looks that way, it just means you are never completely focused on a single task.
stultuske
Posting Sensei
3,108 posts since Jan 2007
Reputation Points: 1,114
Solved Threads: 432
I will put it in few bullets:planning
coding
testing
version control
evaluating
peter_budo
Code tags enforcer
15,432 posts since Dec 2004
Reputation Points: 2,806
Solved Threads: 901