Whenever I get this amazing idea, whether it be a mod, an UMS map or otherwise, I can never stay motivated to finish it.
The reason I make 95% Melee maps is because I can never finish an UMS or a mod. I either get really lazy or I just dont feel like doing it.
My question: How do you stay motivated to finish a map or a mod?

Playtest the early version with your friends. If they like it they'll motivate you, if not it's maybe time for a new concept.

There is the problem, there is no early version.
I dont complete anything.
Out of my 25 mods in my mod folder I created, I only completed 1. The rest are unfinished.
Not to mention the hundreds of unfinished maps in this folder I have in my maps folder.
All of this was over the course of a couple years.

...
And you all thought IskatuMesk was bad at finishing mods. Pfft

I have the same exact problem, I've completed one very simple mod and a few really crappy maps throughout my years of mapping. I have a fighting map in my map folder right now that only needs a few more triggers and I've left it like that since last year. I'd love to know the answer too, considering I get motivated up until the end.

It has to be something you really want done, and you have to publicize it enough that people will nag you about it. I have people in my clan that regularly ask how my map is coming along, for example. But the best thing is for it to be planned, for you to care about it, and to stop yourself playing games or procrastinating instead of mapping.

Schedule regular time intervals to work on it. This works best if, during those time intervals, you have nothing else to do.

I've been doing that with a website revamp I've been doing for a site dedicated to glitches... However, the plan's starting to fall apart, since I am now also working on a website revamp for my school. This is more because the new project is in a very different stage right now (development of the client-side JS for a cross-browser AJAX-enabled contentEditable-based content management system).

So works best if you have nothing else to do, and only one project to fill your time with.

Then just show off whatever you have. For mods this should be even easier as you'll probably have some changed graphics to show.
For maps get the basic concept done which should only take a few hours. If that's already too much then at least talk about your idea.

All of this really did help, I made some major progress. Thanks O.o

Another quick question:
I get distracted. For instance, I am in the middle of making a map and "Oooh, Starcraft!" or "Oooh, FireFox!" because I see the icon on my desktop or the window in the taskbar.
Now how is THIS avoided.

It seems by v6 I'll have written a tutorial on everything:
http://www.staredit.net/topic/8124/

It's a lot easier just messing with other people's maps

Delete all programs except starcraft and scmdraft and any modding tools. Then give your router to a family member and tell them not to give it back until your map/mod is done.

I've learned to accept that some ideas I have will never be completed. I would say I make roughly 5 maps halfway before I make one fully. (Part of this is because I tend to have strange ideas for maps, which tend to run into roadblocks that I don't have the patience to figure out.) The trick is to simply sit down and do it, no matter how aggravating it might be. In fact, the more you do something, the easier it is to do, and you actually might start enjoying the mundane things.

Plan, plan, plan. If you don't plan, you won't remember/won't know what to work on next. Ways to plan can range from...

• Having a list of things to do in Notepad (I prefer Notepad since it's handy-dandy and easily accessible to you while mapping) or any other text storage programs, having check boxes (I use boxes like [ ] and fill it in like this, [X], when the thing is completed.) or removing the item from the list once you complete it in the map. This is what I do and it helps keep me on track.
  
• Write down ideas on paper. When you write something down you remember it better.
  
• Use your most important tool, your brain, to store information. Not recommended as it's very easy to forget small details.
  

General things to do:

• Keep information within reach. Death counts, switches, and other things should be kept track of.
  
• Don't test too often until you get to the stage that testing is fun and the map is somewhat playable.
  
• If you have a bug and you can't get it right by the fifth or so time, get help. Nothing's more stressful than not being able to fix a bug.
  
• Pick a map that YOU like. If you don't like it, you won't want to work on it. Don't mind what other people say about it - it's yours.
  
• Don't make a production thread (At least, not until the map is at beta stage or higher).
  
• Keep triggers well organized. I'm not saying use comments on every single one, which you shouldn't, but organize triggers that work together with each other.
  
• Party hard.
  

I plan mine in Notepad++ and I set the language to XML so it's even more neat.