Lenses for puzzles

I have been reading Jesse Schell's excellent book "The Art of Game Design: A Book of Lenses". It is possibly the best game design book I've read so far. Recently I got to the chapter 12, which is about puzzles. Since After Now Archeology is also a kind of puzzle game (or riddle), I found the chapter really interesting and relevant to my work.

In the chapter, Schell gives 10 principles for puzzle design:

1. Make the goal easily understood
2. Make it easy to get started
3. Give a sense of progress
4. Give a sense of solvability
5. Increase difficulty gradually
6. Parallelism lets the player rest
7. Pyramid structure extends interest
8. Hints extend interest
9. Give the answer!
10. Perceptual shifts are a double edged sword

When looking my design through these lenses, it is clear I have done something right - and something wrong. The first test version of the first puzzle exposed some clear violations of these principles. And I have to say, I violated the rules purposely.

1. The goal was not easily understood. As Schell also writes, sometimes figuring out what to do is a part of the puzzle. But maybe I should skip that part and give the player some more hints about what to do. There seems to be enough challenge without the extra mile - which I could save for the most hardcore players.
2. Make it easy to get started. This relates to the principles 5, 6 and 7. There are just a couple of things the player can do, she just needs to start exploring the navigable space. After a while, the puzzle starts to become clearer. Since it has quite many different parts, the player can do many things at the same time. But I promise, I'll violate rule number 2 in many places. After all many of the puzzles or riddles in the game are purposely hard to get started.
3. Give a sense of progress. This is definitely something I need to focus more in the future. Many of my current designs don't tell the player if she is on a right track. Many of them will stay that way.
4. Give a sense of solvability. All puzzles are solvable. But as the first test revealed, small graphical glitches or minor flaws in presentation, instructions etc can make the puzzle feel unsolvable. This is something I need to keep in mind, since I'd like to have the player to use her brain power on the problem itself, not all the unimportant things around it. This relates to good UI design and transparency. 
5. Increase difficulty gradually. I have tried to organise the puzzles in the game so they are in increasing difficulty order. Also different parts within a puzzle should be in increasing difficulty. This is hard, since the game is based on the knowledge outside the game, and different players have very different knowledge basis. For example, if the puzzle consists of lingual part and mathematical part, there is no way to ensure that all the players feel the lingual part easier than the mathematical part. 
6. Parallelism lets the player rest. This is why I decided to go with three separate paths. If you get stuck on one path, you can continue on another path. In the very beginning the player can choose from three different puzzles. But after solving one of them, she has 4 choices and after solving one of them she has at least 5 possible puzzles to continue with. And you don't need to solve all the puzzles to solve the last one, although it might help.
7. Pyramid structure extends interest. Well, the whole structure of the game is kind of pyramid. There is the final puzzle, for which you need to solve at least 9 of the other puzzles and there are some extra puzzles too. Also many of the individual puzzles in the game consist of several parts, leading to the final answer. Building the puzzles around three different main themes can bring in players interested in different areas, leading them to explore the other areas too.
8. Hints extend interest. This is one of the cornerstones of the design. Personally I love difficult puzzles. But I hate to get stuck in them. Small hints every now and then could help me to continue playing. I have been thinking to include a "easy mode" button that automatically gives the player all the hints. 
9. Give the answer! Well, Schell says I should give the answer to the player after she has struggled with the puzzle for a long time. All the answers will be in the internet anyhow. I will not. The progress in After Now Archeology is based on solving the puzzles and if I give the answers, there is no game at all. The internet will give the answers, but that is something I can do nothing about. It is up to the player if she wants to spoil her experience by going to some forums to look at the answers. On the other hand, quite many of the similiar games have player created forums for discussion, and these forums are very nicely self-regulated. People do not want to spoil another players' game, hints are usually quite subtle and I have not accidentally stumbled on complete answers on these forums.
10. Perceptual Shifts are a double-edged sword. This is something I know and am willing to exploit. Thinking outside the box, adding extra dimension to the thought etc etc. 

Well, they say you learn game design by designing games, not reading books. But I say you get some excellent tools for the game design by reading books.