I like games that offer tight, well-rounded experiences, so I’m not really fond of complex maps or unexpected uses for everyday objects. Inventory items should be used for the obvious actions and the whole game state should fit in the player’s mind (the less hunting for discarded items or aimless wandering the better).
5 Likes
Ask/tell-style communication, and any specialist communication techniques with the parser. e.g. ‘JOHN, blah blah blah’ etc… for tedious personal dislike reasons repeatedly elaborated by me over the years on this forum.
I guess I don’t do instructions any more. I now use a tutorial that’s part of the start of the game, if a player wants it.
I feel like most of my avoidances can’t otherwise be generalised. Well, they’re specific objects, in context.
When I’m making a location, there are a lot of things I could add that fulfil some required functions or ideas that are in my head. And the first one or two I think of may just lend themselves to too much interaction (like @kamineko 's wrench). So I keep trying to think of alternatives until there’s one that’s Doing The Job, Artistically Satisfying, and… which won’t have the player trying it on everything. All the objects I mentally threw out on the way, or if unlucky, started to implement, those were/are the things I avoid putting in my games.
-Wade
9 Likes
I appreciate that this is a personal preference, but using an item for its obvious action doesn’t sound like much of a puzzle to me. My favourite example of a puzzle in which an object is used in an unexpected way is probably King Arthur’s Night Out, in which the magical sword Excalibur is used to retrieve something from under a bed. It’s a good use of lateral thinking and it made me laugh out loud.
6 Likes
Yes, I don’t mind weird inventory manipulation as a player. It’s just that I’m not inclined to it when planning a game myself.
4 Likes
But unless it’s a game in which it’s purposefully ridiculous (ahem… the components of the Goddess-defeating machine in LGOP), they have to have even the most minimal reason of some sort. Like, tea is used as part of a movement device in H2G2 because of its Brownian motion!
Actually… So far, the only person who can away with things like this is Steve Meretzky. I mean, winning a broom from a large Double Fanucci match?! How tight is that? 
1 Like
The example I gave was ridiculous, but there are plenty of circumstances in which an everyday object could be used for a purpose it wasn’t intended for. IRL I regularly use a biro or a house key to slice through parcel tape.
6 Likes
In my next game, I will have a puzzle where you have to cut a frayed rope with a key, quite like your thing!
2 Likes
Interesting topic. Although I am not (yet) an established author, I do have some notes on what I want (not) to do within my game.
Empty rooms: I also do not like ‘nonfunctional / corridor’ rooms but I find I still want to use them to avoid disambiguation questions. E.g. a courtyard with four exits (doors on the N, E, and W ends, and a gate on the S end) I stretch horizontally into three rooms so each room will have one door (and the central room also has the gate.)
Diagonal directions: I do use them for e.g. implementing an octagonal room, but then again I do not hide ‘visible’ exits (I use Exit Lister by Eric Eve.)
Up and Down directions: Although we are planar creatures (widescreen in theatres and the like, we are very much horizontally oriented), I do like to add a bit of verticality. But the compass (at least the default one) is heavily biased (8 horizontal vs 2 vertical directions.)
Undo prevention: No way. However a player should still be free to make a complete a… of themselves but I will try to limit that to the very beginning of the game, so when the player realises the impact of their st.p.d.ty they can still restart and reconsider without losing a lot of progress.
Mazes: I do love mazes (and is a parser game not a huge maze in a way?) but only if there is some system or theme to it which can be discovered by the player for fun and profit. Even more fun if I can create a maze which is not a maze in the players view 
.
Default descriptions: I like to add some colour to the game, and descriptions help a lot. I also do not like this ‘dry Englishman’ voice which comes with Inform7 standard rules. So I try to go the extra step and make the ‘game voice’ match the setting of the actual game. Actually the ‘game voice’ I currently implement as an NPC following the PC. Whether this will work out, I guess I will know once some poor folks had a look at my endeavours.
Large inventories: A large number of objects X a large number of verbs… ouch. I try to minimize object usage to keep this under control (e.g. no need to implement a fully functional kitchen if it is not important to the story.)
Conversation: This is the one ‘feature’ I am most hesistant to implement. I try to avoid the problem by going for ‘action driven dialogue’ (e.g. the player does something triggering a conversation thread for an NPC), and refuse more direct communication until I find a way which works for me (possibly the ‘talk to NPC’ method which more or less sidesteps the topic problem.)
Environment size: My environment contains more than it strictly needs to get to the ‘end’. But the additional stuff is then related to a ‘side quest’ which is optional (to satisfy the completionist and curious folks.) I might also add some ‘Did you try …’ section to the achievements record at the end of the game to hint at things people have missed.
Menus: I do not like menus within a parser game, to me they feel like ‘breaking the flow’ of the game. I do want to add a ‘tutorial’ section to more or less teach players the commands they can use (like the tutorial extension by Emily Short.)
Unusual use of common objects: I wish I had more experience in this. Thinking ‘out of the box’ and encouraging players to do so. Maybe for a future game once I get more experience under my belt.
7 Likes
Hunger daemons have been mentioned already, but exhaustible light sources are somewhat related (I have ragequit at least one game upon finding out it featured them). I would need a very good reason to put either of these in a new game.
8 Likes
I like writing conversations. Which way to present them is another story. I used to think ask/tell with automatic hints was a good way to go about it without breaking the interaction flow. And this is indeed, in some ways, an aesthetically pleasing experience. Unfortunately, on second thought, it is either a clumsy replacement for a proper menu, or (due to expectation setting) an underclued puzzle to ask/tell a given NPC about something not included in the topic hints. Consequently I avoid this style. I went with menus before. Maybe I’ll be reducing this to a non-choice implementation of TALK TO in the future.
6 Likes
Interesting topic with some interesting responses. So, let’s cut to the chase. The things I actively avoid in my games (remembering that I write parser-based games):
- Long, verbose pre-game rambles littered with press-any-key, multiple screen clears, irrelevant guff, quote boxes and other unnecessary clutter. Just get me to the game.
- Menus, multiple choice or conversation trees in any way shape or form.
- Sudden death. (I will occasionally kill off the player if they do something stupid, but they can always undo.)
- Guess-the-verb, as best as possible. (Testers are really helpful here. I won’t necessarily provide alternative ways of solving the same puzzle, but I’ll provide lots of synonyms and alternative ways of expressing what you want to do, e.g. give meat to dog vs feed dog with meat.)
- Making the game unwinnable. (I try to avoid this as best as possible. If the game can get into an unwinnable state, I’ll try to warn the player beforehand so they can save the game and if it gets into an unwinnable state, I’ll let them know straight away, but as subtly as possible.)
- Long flowery room descriptions that add no value to the game or require you to press Enter every time you enter a room before you can even get to the prompt. This just slows the game down.
- Scenery that can’t be examined. (See the previous point.)
- Objects that don’t belong in the location. (In the retro era, objects were randomly scattered over the map like a mad woman’s breakfast. Whoever heard of a horseshoe in a living room or a flower pot in a stable? Place objects where you would expect to find them in the real world, as best as possible.)
The things I don’t use very often, but don’t actively avoid:
- Mazes. (I actually like mapping and a logical maze is a challenge to my mapping skills. I will never, ever use an illogical maze, unless I’m porting it from another game.)
- Empty rooms. (Things like corridors and hallways are a necessary evil for a realistic map, so I don’t avoid them, but I’ll try to make them interesting, perhaps with hints, scenery or random ambient activity.)
- Red herrings. (If it makes sense for an object to be in a particular location, then I’ll include it, even though it is not essential to solve any puzzles, e.g. who’s ever heard of a bedroom without a bed or a kitchen without some sort of cooking appliances? It’s up to the player to work out what’s relevant and what’s not. That’s part of the challenge.)
I’m sure there are other things, but these are the ones that spring to mind.
15 Likes
The “necessary evil” of empty rooms is one that I’m currently trying to accustom myself to, because I’m building puzzles around coordinating your movements with those of NPCs (like in a classic stealth game), and that means there are a lot of rooms that need to exist for routing reasons but don’t serve any narrative purpose of their own.
Plus the game is onboard a ship, which means I only have cardinal directions to work with, so I’ve had to add some extra rooms just to get enough exits in some places…
6 Likes
Well, this thread is less about ‘what are the best design decisions’ and more about ‘what are your personal quirks?’
Steve Meretzky seemed obsessed with empty, useless rooms and red herrings, and he’s one of the best selling IF authors of all time. Emily Short’s Savoir-Faire has a whole maze of rooms that are mostly just doors; Andrew Plotkin’s Delightful Wallpaper is mostly empty rooms in the first half; Adam Cadre’s Narcolepsy has quite a few pointless locations.
I don’t like making empty rooms, but with the right setup it can be either unnoticeable or great.
14 Likes
True. I should make it more clear this is a personal hangup, not a universal rule—in Scroll Thief I tried to make every single room necessary for a puzzle in some way, which is why the core map is so dense. Similarly in Enigma every room has one or more puzzles all its own. Now that I find myself unable to do that I’m trying to figure out how to break the habit.
5 Likes
I dislike anything that gets between me and the ending. I’m a busy man and don’t have time for games.
I think that’s a different case, since there is as much emphasis on current action as there is on geography. I wouldn’t avoid something like that. Unfortunately, I don’t think I could code it!
3 Likes
That makes me wonder, has anyone ever done a game with more vertical directions? Up-north, up-west and so on? I could see it in a zero-G environment, although it would be very disorienting (which would have to be the point, I guess).
6 Likes
I think 10 directions (eight horizontal, two vertical) is already approaching the limits of people’s memory; expanding that to 26 (three “rings” of 8 plus straight up and straight down) is probably well across it!
6 Likes
My first released game, Ether, is in a zero-G environment and uses commands like NEU for northeast-up.
10 Likes
(adds Ether to my to-play list)
5 Likes
There was a thread asking about games with more vertical maps somewhat recently. Ether was mentioned there, and I pointed out Andrew Schultz’s The Cube in the Cavern, where you can go in all eight direction on each face of a cube, so it does all the 2-axis combinations… I can’t remember if other games came up that do it, but probably?
7 Likes