Grinding

Have this ever happened to you? You've finished all the side quests available, and now continuing the main quest is all that remains. There is only one problem, in order to continue the main quest, you need to venture in to this new area, where there are new kinds of monsters, and they are all too powerful for you. There is only one thing that remains to be done, go back to the areas you've already fully explored and kill more of those creatures, you've already killed by the thousands until you level up a few more times.

Was that fun?

Chances are that you say no, or at least admit to it not being the highlight of the game. The main problem is that the game is preventing you from actually playing the game, and forces you to do something achingly close to manual labor (and that's certainly not why I play games).
Grinding is used to denote these situations where the player spends his or her time, doing repetitive tasks to increase his or her character(s) abilities. Many games (particularly those with RPG elements in them) have this as an integral part of their gameplay. I will only talk about grinding in single player games, Grinding in multi player games, particularly massive multi player games, is quite a different issue.

I can think of four rationals for implementing grinding in your game.

• Content reuse. If the player spends a lot of time grinding on the same level with the same monsters, you get more play time out of the effort you put in on content creation.
  
• If your game requires grinding, you know it isn't too easy. If someone wants more of a challenge, he or she can just grind less. And if you grind more, the game becomes easier.
  
• Some weird, misplaced, idea that the player must be made to work a little in order to appreciate his or her success.
  
• Game X, Y and Z had it.
  

I'm not advocating not to reuse content, but there is a huge difference killing a thousand skeletons if they are guarding the gates to the castle of the evil warlock, compared to walk round and round the marsh in order to kill the thousand skeletons you need in to level up so that you can enter the warlock's castle. God of War and Diablo II are good examples of games that requires no grinding even though you kill the same kind of monsters over and over again. The difference is that there is a purpose other than just trying to level up, there is always a quest, you always progress the story.

Grinding is not a good substitute for proper balancing and play testing.

If you've created a good game, players will appreciate it just fine, enjoying the unfolding story, overcoming the challenges, and not spend countless of hours doing mundane tasks over and over again in order to be able to get to the good bits. Creating the main story, and its assets like environments and characters are usually the most expensive part. If you plan this phase well, adding a couple of side quests using mostly existing assets should be fairly easy. For instance, take an unused corner of your map, put up an odd formation of rocks, some monsters (you do have both monsters rocks, right?), and add an item that must be retrieved. Once you've got the assets, adding a quest like this should be fairly simple. You can also create mini-bosses by boosting a common monster, scaling its size, or add an extra effect, or maybe tint its textures in a different color. Diablo II is again a good example of this.

Imitating other games to a certain extent is good. It allows you to use conventions know to your players and also reuse proven deigns. However, before you reuse any feature, you must critically study it, understand its strengths and weaknesses, and how these apply to your game. And remember, all features aren't good, and some conventions are outdated. Grinding is on of them.