You may ponder about the wisdom of certain design decisions of Crawl. 

In a nutshell: This game aims to be a tactical fantasy-themed dungeon
crawl. We strive for strategy being a concern, too, and for exquisite
gameplay and interface. However, don't expect plots or quests.
You may ponder about the wisdom of certain design decisions of Crawl.

if you are used to other roguelikes and want a bit of background on the 
differences. In a nutshell, prime mainstays of Crawl development are the
following, most of which are explained in more detail below. Note that 
almost of all of these date back to Linley's first versions.

if you are used to other roguelikes and want a bit of background on the
differences. Prime mainstays of Crawl development are the following,
most of which are explained in more detail below. Note that many of
these date back to Linley's first versions.

 o avoidance of grinding (a.k.a. scumming)
 o in particular, permanent levels, shops won't buy, and food clock
 o clarity, i.e. playability without access to spoilers
 o class/race differentiation using skills and aptitudes
 o killing no-brainers (well, we try), e.g. no permanent intrinsics
 o replayability (using branches, different styles, and gods)
 o out of depth monsters
 o gameplay supporting interface _and_ newbie support

Major design goals:
 o challenging and random gameplay, with skill making a real difference
 o meaningful decisions (no no-brainers)
 o avoidance of grinding (no scumming)
 o gameplay supporting painless interface and newbie support

One basic design principle is avoidance of grinding (also known as
scumming). These are activities that have low risk, take a lot of time,
and bring some reward. This is bad for a game's design because it 
encourages players to bore themselves. We try to avoid this!
This explains why shops don't buy: otherwise players would hoover the 
dungeon for items to sell. Another instance: there's no infinite 
commodity available: food, monster and item generation is generally not
enough to support infinite play. Not messing with lighting also falls 
into this category: there might be a benefit to mood when players have 
to carry candles/torches, but we don't see any gameplay benefit as yet. 
The deep tactical gameplay Crawl aims for necessitates permanent dungeon
levels. Many a time characters have to choose between descending or 
battling. While caution is a virtue in Crawl, as it is in many other 
roguelikes, there are strong forces driving characters deeper.
Another key feature is clarity: things ought to work in an intuitive 
way. Crawl definitely is winnable without spoiler access. Concerning 
important but hidden details (i.e. facts subject to spoilers) our policy
is this: the joy of discovering something spoily is nice, once. (And 
disappears before it can start if you feel you need to read spoilers - a
legitimate feeling.) The joy of dealing with ever-changing, unexpected, 
and challenging strategic and tactical situations that arise out of 
transparent rules, on the other hand, is nice again and again.

Minor design goals:
 o clarity (playability without need for spoilers)
 o internal consistency
 o replayability (using branches, races, playing styles, and gods)
 o proper use of out of depth monsters

The skills and aptitude system is one factor encouraging strategic play.
It also serves to clearly differentiate the many species, thus providing
replayability, in particular since the class/race combinations are by no
means homogeneous in difficulty. Note that a rough idea about aptitudes 
is definitely enough to win, yet players can optimise here, as well. It 
can be said that race differentiation is still not finished - we try to
make differentiation better by going beyond aptitudes alone. A weak spot
of the current skill system is 'victory dancing', where characters spend
the experience accumulated in a big battle with stupid actions (like 
casting Magic Dart at the wall) in order to increase specific skills. 
While this surely seems dubious, it also allows players to adapt their
characters anytime during play - e.g. to make a transition from a pure
melee fighter to a hybrid using enchantments. For this reason, changing
the experience system is no easy task.

   Balance:
The notions of balance, or being imbalanced, are extremely vague. Here
is our definition: Crawl is designed to be a challenging game, and is
also renowned for its randomness. However, this does not mean that wins
are an arbitrary matter of luck: skill of the players will have the
largest impact. So, yes, there may be situations where you are doomed -
no action could have saved your life. But then, from the midgame on,
most deaths are not of this type: by this stage, almost all casualties
can be traced back to actual mistakes; if not tactical ones, then of a 
strategical type, like wrong skilling (too broad or too narrow), unwise
use of resources, or wrong decisions about branch/god/gear.
The possibility of unavoidable deaths is a larger topic in computer
games: in the absence of a human moderator, games can either be soft
in the sense that optimal play ensures a win. Apart from puzzles,
though, this means that the game is solved from the outset -- this is
where the lack of a human gamemaster is obvious. Or they can be hard in
the sense explained above. We feel that the latter choice provides much
more fun in the long run.
(The next topic can also be filed under balance; see Replayability for
what balance does not mean to us.)

   Crusade against no-brainers:

Speaking about games in general, wherever there's a no-brainer, that 

Speaking about games in general, wherever there's a no-brainer, that

that's really not an interesting choice at all. And that's a horrible 
lost opportunity for fun. Examples for this are the resistances: there 
are very few permanent sources, most involve a choice (like rings or 
specific armour) or are only semi-permanent (like mutations). Another 

that's really not an interesting choice at all. And that's a horrible
lost opportunity for fun. Examples for this are the resistances: there
are very few permanent sources, most involve a choice (like rings or
specific armour) or are only semi-permanent (like mutations). Another

that most artifacts are randomly generated. Furthermore, even fixed 

that most artifacts are randomly generated. Furthermore, even fixed

items in general. Likewise, there are no sure-fire means of life saving 
(the closest equivalents are controlled blinks, and good religious 

items in general. Likewise, there are no sure-fire means of life saving
(the closest equivalents are controlled blinks, and good religious

Concerning replayability again, Crawl's dungeon layout was also
constructed with this in mind: even veteran players will find the Tomb or
the Hells exciting (which are construed such that life endangering 
situations can always pop up - this tries to avoid the walking tank 
phenomenon). Another strong point is the religous system, because Crawl
allows you to choose gods in the game, regardless of class or race (and
even to switch to other gods).

   Anti-grinding:
Another basic design principle is avoidance of grinding (also known as
scumming). These are activities that have low risk, take a lot of time,
and bring some reward. This is bad for a game's design because it
encourages players to bore themselves. Even worse, it may be optimal to
do so. We try to avoid this!
This explains why shops don't buy: otherwise players would hoover the
dungeon for items to sell. Another instance: there's no infinite
commodity available: food, monster and item generation is generally not
enough to support infinite play. Not messing with lighting also falls
into this category: there might be a benefit to mood when players have
to carry candles/torches, but we don't see any gameplay benefit as yet.
The deep tactical gameplay Crawl aims for necessitates permanent dungeon
levels. Many a time characters have to choose between descending or
battling. While caution is a virtue in Crawl, as it is in many other
roguelikes, there are strong forces driving characters deeper.
   Interface:
The interface is radically designed to make gameplay easy -- this sounds
trivial, but we mean it. All tedious, but necessary, chores should be
automated. Examples are long-distance travel, exploration, and taking 
notes. Also, we try to cater for different preferences: both ASCII and
tiles are supported; as are vi-keys and numpad. Documentation is plenty,
context-specific and always available in-game. Finally, we ease getting
started via tutorials.

Likewise, there are many fundamentally different playing styles to
discover (melee oriented fighter, stabber, etc.). There have been even
deliberate design choices that allow meta-styles: For example, Mummies
do not need to eat and so are principally suited for an infinite play. 
On the precise opposite end, players who prefer to be rewarded for 
accepting a more severe "food clock" can play Trolls with their massive 
strength and regeneration from wounds, or Centaurs, with their amazing 
missile skills and the speed to take advantage of them. Draconians, on 
the other hand, reveal their adult form (including aptitudes, and 
sometimes resistances) only at level 7. These are a deviation from the 
usual rule that, after choosing a race, the complete future of that 
character lies in the hands of the player. Nearly every race in the 
game "offers" a style of its own like the above! Sadly, however, at 
present some do so far less than others.

   Clarity:
Things ought to work in an intuitive way. Crawl definitely is winnable
without spoiler access. Concerning important but hidden details (i.e.
facts subject to spoilers) our policy is this: the joy of discovering
something spoily is nice, once. (And disappears before it can start if
you feel you need to read spoilers - a legitimate feeling.) The joy of
dealing with ever-changing, unexpected, and challenging strategic and
tactical situations that arise out of transparent rules, on the other
hand, is nice again and again.
In concrete terms, we either spell out a gameplay mechanic explicitly
or leave it to min-maxers if we feel that the naive approach is good
enough.

From time to time a discussion about Crawl's unfair OOD (out of depth) 
monsters turns up, like a dragon on the second dungeon level. These are 
not bugs! They serve as motivation, first of all: in many cases, such a
situation can be survived somehow and the mental bond to the character 
will then surely grow. OOD monsters also help to keep more experienced 
players on their toes. The same can be said of uniques. Also, frequent
and early trips to the Abyss are not deficits: there's more than one way
out and possibly doing so should be exciting for all characters.

   Consistency:
While there is no plot to speak of, the game should still be set in
a consistent Crawl universe. For example, names of artefacts should fit
the mood, vaults should be sensibly placed, and monsters should somehow
fit as well. Essentially, this is about player immersion. As such, it's 
good to have in mind, but consistency is always secondary to gameplay.
A typical example is player vs monster behaviour: while we try to make 
these identical (or similar), there are good reasons for keeping them 
distinct in certain cases.

Finally, the interface of Crawl is designed to be understood at a glance
and to support gameplay as far as possible. In particular, it should 
make tedious activities (like making notes of important stashed items, 
or doing long travels) less tedious. This is how interlevel travel, 
autoexplore, the stash tracker, the dump file, inscriptions, and macros
got into the game.

   Replayability:
This is actually quite important, but in some sense just a corollary
to the major design goals. Besides these, there are several other 
points helping to make playing Crawl fun over and over again:
* Many species, which actually play differently: This is partly due to
the skills and aptitude system. Similarly important are the built-in
starting boni/handicaps of races; these often have great impact on
play. To us, balance does not mean that all combinations of class and
race play equally well! Some are much more challenging than others,
and this is fine with us. Each race has at least some classes playing
rather well, though.
* Dungeon layout: Even veteran players will find the Tomb or the Hells
exciting (which are construed such that life endangering situations can
always pop up). These and other branches may or may not fit a given
character's buildup. By the way, we strongly believe that games are 
pointless if you can reach the invincible state.
* Religion: This addresses new players, as getting to the Temple and
choosing a god becomes the first major task of most games. But religion
is also a point in favour of replayability for experienced players,
since the choice of god can matter as much as race does.
* Playing styles: Related to species, class, god are fundamentally
different playing styles like melee oriented fighter, stabber, etc.
Deciding on whether (and when!) to make a transition of style can make
or break games. 

Still, we make a lot of effort to care for new players (who naturally
have no access to most of these utilities). The tutorials are one step 
in that direction, and, in general, all information should be clearly 
available and documented in-game. The interface is designed with being
radically player-friendly and customisable.

   Out of the depths:
From time to time a discussion about Crawl's unfair OOD (out of depth)
monsters turns up, like a dragon on the second dungeon level. These are
not bugs! Actually, they are part of the randomness design goal. In
this case, they also serve as additional motivation: in many situations,
the OOD monster can be survived somehow and the mental bond with the 
character will then surely grow. OOD monsters also help to keep players 
on their toes by making shallow, or cleared, levels still notr trivial.
In a similar vein, early trips to the Abyss are not deficits: there's
more than one way out and successfully escaping is exciting for anyone.