I’m interested in starting to write IF. However, before starting out on a grand, epic project, it seems wiser to create something small and simple. After all, I wouldn’t want to lose motivation halfway, or make something complex without first learning the proper way to structure and plan it out. Do most other people start with writing IF this way? Is there any other advice you have along this line?
I think that’s a good approach to starting out. You might think up a simple story that involves just two or three rooms (or locations), a small number of objects (three or four), and only one or two puzzles. Focus on making sure everthing has descriptions, everything works together (i.e., there are responses for all the likely ways each object might be used with each piece of scenery and with each other, etc).
Although trying to write too large a game can be an educational experience in itself, as long as you know when to quit.
Starting small sounds like a good idea, given that writing a game takes much longer than it would at first seem. It gets a bit faster when you’ve familiarized yourself with the authoring system and don’t have to fight as much with the coding so practicing with a smaller game speeds up the development of a possible longer game later.
I’d like to go against the common opinion here: I think the best thing to do is to write a game you are interested in making, and if that is a large game, so be it. Now obviously, you should perhaps not start with something the size of Blue Lacuna–I’m sure there are smaller games that are still interesting. But writing a small practice game… where will you find the motivation to keep going, if it’s not the motivation of writing a game you actually want to share with people?
Personally, I started with The Baron, which is not small.
There is nothing wrong with starting with a large game, as long as you can program it in small chunks. The first scene of a larger game could be the equivalent of a small game, after all.
If you do write a small practice game first, make it really small. Do it just to familiarise yourself with the language, not to get to know all the subtleties, and make sure you don’t spend too much time on it.
I guess there’s no wrong way as long as you’re motivated and stick with it. I just know that I often regret starting out in Hugo with Trading Punches. I overestimated the extent to which my programming experience would translate into award-winning IF, and I really put too much trust in my prior experiences writing home-brew IF (the most recent of which had been five years prior). It’s an interesting story with epic potential, but I don’t think I was ready enough. The implementation didn’t match the ambition. I hope to do better when I (some day) continue with the next game in that series, but I think I should have done more learning before tackling even an IFComp-sized game like that.
But I guess it was still good as a learning experience. I’d just rather that my learning experience hadn’t been that particular game. Something much smaller might have been better. 
When I started out about a year ago, I decided to write a four-room practice “game” on making a Margarita, with salt in a cabinet in one room and a glass in another, etc.
I thought that sounded so simple as to be a waste of time, but I just wanted to get my feet wet with Inform. But I was amazed at the amount of work it took, and the number of things that I thought would be child’s play that I ended up having to look up in the documentation. Plus there’s those little tiny syntax things to learn, where if you word something slightly different you get an error or an unexpected result.
So it was a very valuable experience. I’d really recommend making a simple game like that.
If you start out with a “real” game, then I’d definitely make it in a stepwise fashion. In other words, put in a few broad things and get them working together, then move on to the details of acquiring or using the broad things.
I have to agree with ShaeSays 
the scope of my first game is fairly simple, but a bit broad. I wrote up a basic outline on a plane flight, then started programming it. It is quite amazing how much you think “ooh that needs a bit more detail” or “to help the player get where I want them to be, I need to just…” and suddenly it’s a whole load of creative thinking required, which you then need to implement.
I’ve only got two basic locations so far - a pub and a market, but there are 14 “rooms” and 5 characters! And there aren’t any market traders yet! By the time the player gets to the Palace (where the ‘quest’ is given to them) it’s going to be a fair bit of code 
Just doing the small amount I’ve done so far has taught me so much 
I don’t know that it’s necessary to start with a “small” game, but I think it should be an uncomplicated one. You probably don’t want to do anything innovative with the IF format on your first attempt, and you may also want to stay away from difficult programming and extremely complicated plot devices. As the wise posters above, who have experience in successfully completing IF, have said, the important thing is that you’re motivated to finish it.
I find that when I try to do something very simple, it often ends up becoming rather ambitious as I think and fantasize about it.
I certainly wish you success! 
It’s very satisfying to have written an IF game. I’m glad I ended up finishing “Dreary Lands” to enter it into the 2005 IF Competition, even though it was horribly buggy and amateur.