Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Single use command considered harmful

6 views
Skip to first unread message

Radomir 'The Sheep' Dopieralski

unread,
Sep 28, 2008, 5:41:14 AM9/28/08
to
Roguelike games have traditionally different commands for using
items in different ways. For example, you have separate command
to quaff a potion and another one to eat a ration of food. This
distinction may seem superfluous, as the commands will not work
with any other kind of item: you cannot eat a potion, you can't
quaff a ration of food. This seems unnecessary: single, unified
use command seems simpler and easier to learn. There seem to be
no advantage of the traditional approach. I will show that this
is not true, and that there is indeed something valuable in the
traditional roguelike user interface.

There are two common arguments in favor of the multiple command
set of traditional roguelikes. One of them is about how this is
a way of filtering the inventory list: only showing potions, or
only showing scrolls. There is no need for additional filtering
and sorting commands in the inventory. The other argument tells
about the complexity that is added by allowing multiple uses of
a single item, the exploration game introduced by not revealing
these uses right away. Those are both valid points, but I think
there is much more to it.

In her paper "Object Concepts and Action", published with other
similar papers in book "Grounding Cognition", Anna M. Borghi is
presenting her theory, together with some experimental evidence
about how we represent artifacts in our minds: as references to
actions that can be performed with those objects. These actions
are usually very simple movements -- and they are remembered as
simulations of those movements. Assuming this theory is true, I
can see another important function of so many separate commands
in roguelike games: they work to better differentiate the items
in our minds by providing different affordances tied to motions
of our fingers on the keyboard.

Without these different commands items just feel like all those
objects you collect in a platform game -- they give you points,
sometimes they give you an additional life or some special move
or attack, but most of the time you don't need to look at them,
you just collect them all. Items in roguelike games are totally
different, even if there is some error made by the designer and
you have a potion and a scroll that do the same thing, they are
still different in the player's mind. This is an error, because
you shouldn't suggest things that are not true -- you break the
intuitive understanding of the game world this way.

Understanding the understanding of roguelike games may lead not
only to better description of the traditional mechanisms -- but
also to using these mechanisms better and improving both design
and interface of the games, leading to better user interface.

Or I might be completely, totally wrong :)
--
Radomir `The Sheep' Dopieralski <http://sheep.art.pl>
"Whenever you find yourself on the side of the majority,
it's time to pause and reflect." -- Mark Twain

Krice

unread,
Sep 28, 2008, 5:55:10 AM9/28/08
to
On 28 syys, 12:41, Radomir 'The Sheep' Dopieralski <n...@sheep.art.pl>
wrote:

> Or I might be completely, totally wrong :)

Yes you are. If for example there is a potion of healing
you can use in healing purpose there is no difference how
you do that, with (q)uaff-command or with (u)se. Or clicking
the potion in graphical user interface. Or whatever.

dpeg

unread,
Sep 28, 2008, 6:42:24 AM9/28/08
to
Radomir 'The Sheep' Dopieralski wrote:

> Roguelike games have traditionally different commands for using

> items in different ways. ..... There are two common arguments

> in favor of the multiple command set of traditional roguelikes.

[namely (1) inventory filtering and (2) complexity from having
more than one use for items]

> ..... I can see another important function of so many separate

> commands in roguelike games: they work to better differentiate
> the items in our minds by providing different affordances tied
> to motions of our fingers on the keyboard.

This is a good point, I call it (3).

> Understanding the understanding of roguelike games may lead not
> only to better description of the traditional mechanisms -- but
> also to using these mechanisms better and improving both design
> and interface of the games, leading to better user interface.

All true. However, I feel that there is a good way to have a 'use'
command: instead of saying 'use' first (essentially bringing up
the whole inventory) and choosing an item later, it is better to
first choose an item and then have 'use' be the standard action
(so noun-verb rather than verb-noun). Note that in traditional
roguelikes this requires an additional keypress ('i'nventory, the
item's letter slot, 'u'se). But if you have a (graphical) RL where
the whole inventory is accessible all the time, this point is moot.

Point (3), i.e. different item categories should be mentally
separated, would be done by sorting the on-sceen inventory by type
already. (This would also address (1).) And multi-use actions would
be done by having a simple action (say left mouse click) do the
standard 'u'se, with another action (say right click) bringing up
the full list of available actions; this would take care of (2).

Altogether, this may be as intuitive as the standard system, and even
a bit faster, given the needed tools (on-screen inventory, mouse).

> Or I might be completely, totally wrong :)

You aren't. Don't provide troll bait :)
Great formatting on your post, by the way.

David

Ido Yehieli

unread,
Sep 28, 2008, 7:15:57 AM9/28/08
to
On Sep 28, 12:42 pm, dpeg <pl...@zio.mathematik.hu-berlin.de> wrote:
>
> All true. However, I feel that there is a good way to have a 'use'
> command: instead of saying 'use' first (essentially bringing up
> the whole inventory) and choosing an item later, it is better to
> first choose an item and then have 'use' be the standard action
> (so noun-verb rather than verb-noun). Note that in traditional
> roguelikes  this requires an additional keypress ('i'nventory, the
> item's letter slot, 'u'se).

2 comments:

1. How is it different than just having choosing the item use it ('i'
followed by item's letter slot)?
2. If you agree that it is the same, what is the difference between
'i'->item letter and 'u'->item letter?

I agree with you main point though, sorting/categorizing the inventory
can create sufficient differentiation between different item types in
the player's IMO.

-Ido

Radomir 'The Sheep' Dopieralski

unread,
Sep 28, 2008, 7:26:48 AM9/28/08
to
At Sun, 28 Sep 2008 12:42:24 +0200,
dpeg wrote:

> Radomir 'The Sheep' Dopieralski wrote:

[...]


> [namely (1) inventory filtering and (2) complexity from having
> more than one use for items]
>
>> ..... I can see another important function of so many separate
>> commands in roguelike games: they work to better differentiate
>> the items in our minds by providing different affordances tied
>> to motions of our fingers on the keyboard.
>
> This is a good point, I call it (3).
>
>> Understanding the understanding of roguelike games may lead not
>> only to better description of the traditional mechanisms -- but
>> also to using these mechanisms better and improving both design
>> and interface of the games, leading to better user interface.
>
> All true. However, I feel that there is a good way to have a 'use'
> command: instead of saying 'use' first (essentially bringing up
> the whole inventory) and choosing an item later, it is better to
> first choose an item and then have 'use' be the standard action
> (so noun-verb rather than verb-noun).

This approach has two fatal side effects -- for one, it introduces
an additional mode in which all the commands work differently than
in the other two modes known from traditional roguelike interfaces
(the "map mode" and the "inventory mode"). By doing that, you also
make the inventory screen deeper, increasing the distance from the
game world. Once you go more than 3 or 4 steps away from the main
mode, players start building a mental map of the menus, instead of
mapping the world presented in the game. This number varies from
person to person and from situation to situation (it is tied to the
magical number seven, plus minus three), but it's best to keep it
as low as possible.

The other side effect is that the verb is not tied to a single,
particular physical motion -- depending on the kind of the item
and order of the actions in the menu, the same set of keystrokes
or moves of the hand holding the mouse can lead to completely
different effects. My theory is that tying the motion directly
with the verb makes it much closer to our minds, lets us immerse
much deeper in the actual world of the game.

> Note that in traditional
> roguelikes this requires an additional keypress ('i'nventory, the
> item's letter slot, 'u'se). But if you have a (graphical) RL where
> the whole inventory is accessible all the time, this point is moot.

I tried to make a roguelike in which whole inventory is accessible at
all times: it's called Z-Day. Maybe it's just bad implementation of
the idea, but I'm not happy with the way it worked. I managed to make
the interface modeless, but something important went missing in the
process. It might have been the verb-motion relation.

> Point (3), i.e. different item categories should be mentally
> separated, would be done by sorting the on-sceen inventory by type
> already. (This would also address (1).) And multi-use actions would
> be done by having a simple action (say left mouse click) do the
> standard 'u'se, with another action (say right click) bringing up
> the full list of available actions; this would take care of (2).

The problem is that clicking with your mouse on different parts of
the screen is not as different, in terms of physical motions and
corresponding sensorimotor perceptions, as pressing different keys.
I suppose that you could make different parts of the inventory work
differently enough to actually make a difference (Diablo's belt may
be one exmaple of such a mechanism), and it might be worth pursuing
in graphical roguelikes.

In similar vein, you could make different weapons actually feel
different by requiring activating them differently. For example,
with Wii-like controller, you could make the players actually swing,
thrust or block.

> Altogether, this may be as intuitive as the standard system, and even
> a bit faster, given the needed tools (on-screen inventory, mouse).

The narrower is the information channel, the less efficient it is and
the harder it is to learn it by habituation. Keyboard accomodates a
much wider array of motions and tactile feeback than a mouse, which
usually only relies on visual feedback. This makes mouse-based interfaces
shallow, in the sense that novice user can operate it almost as
efficiently as experienced user, there is little habituation.
But habituation is usually viewed as something good in user interface
design, something that lets you forget about the interface and concentrate
on the (virtual) objects you are actually interfacing with.

Gelatinous Mutant Coconut

unread,
Sep 28, 2008, 10:01:49 AM9/28/08
to
Why not have both?

I think that a generic 'use item' command can be useful at the very
least as a key command tutorial, as a way to make the learning curve
less intense upfront. Consider something like this:

@user@ u
#game# Use what item?
#game# a - A Treasure Map
#game# b - Ye Flask
#game# c - The One Ring
@user@ b
#game# Do what with Ye Flask?
#game# d - 'd'rop
#game# p - 'p'our
#game# q - 'q'uaff (default)
#game# t - 't'hrow
#game# u - default action ('q'uaff)

The player now not only knows what they can do with Ye Flask, they
also now know the shortcut keys to bring up lists of items that can be
used with the same action.

Additionally, the 'u'se command DOES map to a very important action:
"I am in a desperate situation; let me check my inventory for
something I could use as a Hail Mary, and then use it". Without a
generic 'u'se command, this requires that you check your 'i'nventory,
recognize the proper action necessary to access the item you want,
close out of the inventory screen, and then perform the action.

dpeg

unread,
Sep 28, 2008, 10:11:39 AM9/28/08
to
Ido Yehieli wrote:

> On Sep 28, 12:42 pm, dpeg <pl...@zio.mathematik.hu-berlin.de> wrote:
>>
>> All true. However, I feel that there is a good way to have a 'use'
>> command: instead of saying 'use' first (essentially bringing up
>> the whole inventory) and choosing an item later, it is better to
>> first choose an item and then have 'use' be the standard action
>> (so noun-verb rather than verb-noun). Note that in traditional
>> roguelikes  this requires an additional keypress ('i'nventory, the
>> item's letter slot, 'u'se).
>
> 2 comments:
>
> 1. How is it different than just having choosing the item use it ('i'
> followed by item's letter slot)?

I implicitly assumed that you want a dedicated command key (like 'u') in
order to also have other commands available. If this is not needed, then
you are completely right in that there is no additional key stroke.

> 2. If you agree that it is the same, what is the difference between
> 'i'->item letter and 'u'->item letter?

There is none. (In Crawl, we have a slight confusion between these modes,
and that's bad. No matter which is one superior, mixing them up is worse :)

David

Radomir 'The Sheep' Dopieralski

unread,
Sep 28, 2008, 10:22:53 AM9/28/08
to
At Sun, 28 Sep 2008 07:01:49 -0700 (PDT),
Gelatinous Mutant Coconut wrote:

> Why not have both?

Because having more than one way to do it prevents habituation and makes
it harder to learn, probably... but read on...

Isn't that basically equivalent to 'i'? With maybe additional
functionality of displaying detailed info and list of possible actions
when an item is selected? I think this is how Angband's 'i' works, no?

But you have touched an important thing I have a problem with: not all
item classes in roguelikes actually map into meaningful actions in the
game. Weapons, armor, wands, jewelry, food -- they are all fairly good and
well defined. Problem appears with potions, scrolls and maybe staves,
spellbooks and other magical items. Of course we can say that this is
actually a result of bad design, not taking the item classes into account.

For example, if I want to identify an item, I look for an identify scroll.
If I don't have the scroll, I might try to guess the kind of item by using
it or corrying it around for some time, or identyfying it at shop -- but
the primary means of identify is the scroll, and hence the association
between identyfying and reading is relatively strong.

In situations where you have scrolls of healing (acting as potions) and
potions of fire breath (acting as wands), this association is much weaker,
and maybe you really should just have a single command for 'invoke magic
effect' then. Situation is further complicated when you add staffs and
spells.

Obviously, in the situation you describe you might not be even sure
whether you want to heal (or change your status differently with potion),
attack at distance (with wand) or apply some other trick (with scroll),
so you would have to check all three options. But I think that with strong
distinctions between the three maintained through careful design, you
quickly learn to prefer one of these options depedning on the actual
situation (low hp, surrounded, strong monster ahead, etc.).

Derek Ray

unread,
Sep 28, 2008, 11:20:48 AM9/28/08
to
On 2008-09-28, Gelatinous Mutant Coconut <GelatinousM...@gmail.com> wrote:
> #game# Use what item?
> #game# a - A Treasure Map
> #game# b - Ye Flask

This strikes me as misleading. Everyone knows ye can't get ye flask.

--
Derek

Game info and change log: http://sporkhack.com
Beta Server: telnet://sporkhack.com
IRC: irc.freenode.net, #sporkhack

Simon Richard Clarkstone

unread,
Sep 28, 2008, 11:58:02 AM9/28/08
to
Radomir 'The Sheep' Dopieralski wrote:
> Roguelike games have traditionally different commands for using
[...]

> also to using these mechanisms better and improving both design
> and interface of the games, leading to better user interface.
>
> Or I might be completely, totally wrong :)

Um, how did you get all the lines of your post the same length? You are
using a fixed-width font, so all the lines must have the same number of
characters, but you aren't inserting extra spaces to make them the same
length.

--
Simon Richard Clarkstone:
s.r.cl?rkst?n?@dunelm.org.uk / s?m?n_cl?rkst?n?@yahoo.co.uk
| My half-daughter went to the GMH riots |
| But all I got was this stupid ・-shirt. |

Christian Knudsen

unread,
Sep 28, 2008, 12:34:19 PM9/28/08
to
On 28 Sep., 17:58, Simon Richard Clarkstone

<s.r.clarkst...@dunelm.org.uk> wrote:
> Um, how did you get all the lines of your post the same length?  You are
> using a fixed-width font, so all the lines must have the same number of
> characters, but you aren't inserting extra spaces to make them the same
> length.


I think it's pretty much just by chance... which is actually kinda
amazing...

Keith H Duggar

unread,
Sep 28, 2008, 3:52:09 PM9/28/08
to
On Sep 28, 12:34 pm, Christian Knudsen <webmas...@asciisector.net>
wrote:

It is not by chance. It is by careful selection of vocabularly,
grammar, etc, even with occasional compromises to such. We even
discussed this art form once or twice in the forum. The Sheep's
latest post is a supreme example of the craft. Well done Sheep!
Now I must return to reading rather than gazing at the post ;-)

KHD

Gelatinous Mutant Coconut

unread,
Sep 28, 2008, 3:55:01 PM9/28/08
to
> This strikes me as misleading.  Everyone knows ye can't get ye flask.

Of course ye can't 'g'et ye flask; it's already in ye 'i'nventory! Ye
would have to 'd'rop ye flask first!

Gelatinous Mutant Coconut

unread,
Sep 28, 2008, 4:00:02 PM9/28/08
to
> Isn't that basically equivalent to 'i'? With maybe additional
> functionality of displaying detailed info and list of possible actions
> when an item is selected?

I was thinking that you could 'u'se items that aren't in your
'i'nventory as well, like levers, chairs or nearby doors. Checking
your 'i'nventory would be for just that; checking what items you are
actually carrying.

Billy Bissette

unread,
Sep 28, 2008, 5:22:51 PM9/28/08
to
Christian Knudsen <webm...@asciisector.net> wrote in news:6bf31674-e72b-
4a23-8a86-c...@f36g2000hsa.googlegroups.com:

I've run into a few people who did it intentionally, either for the
challenge or just because they thought it looked neat.

Raymond Martineau

unread,
Sep 28, 2008, 9:59:33 PM9/28/08
to
On Sun, 28 Sep 2008 09:41:14 +0000 (UTC), Radomir 'The Sheep'
Dopieralski <ne...@sheep.art.pl> wrote:

>Without these different commands items just feel like all those
>objects you collect in a platform game -- they give you points,
>sometimes they give you an additional life or some special move
>or attack, but most of the time you don't need to look at them,
>you just collect them all. Items in roguelike games are totally
>different, even if there is some error made by the designer and
>you have a potion and a scroll that do the same thing, they are
>still different in the player's mind. This is an error, because
>you shouldn't suggest things that are not true -- you break the
>intuitive understanding of the game world this way.

I support a single 'u'se command. As you know, there isn't much
difference between 'e'ating and 'q'uaffing, and extremely limited
difference between 'z'apping and 'i'nvoking both wands or staves.
Roguelikes mostly don't have an exclusive 'e' and 'z' - you won't
need to eat a wand, and or zap a charge from food.

In the character''s mind, scrolls and potions are different - you eat
one, and read the other. If the character want's to use a potion,
he's not going to accidently try to read it (unless hit with
confusion). In the player's mind, it's unlikely he wants to do the
said combination, and there's no reason to require memorizing the
keyboard layout for the many flavours of use.

The best example is Nethack seperating 'W'earing armour, and 'P'utting
on accessories. There isn't much technical difference between the
two, and merging them together isn't likely to confuse or annoy
new/veteren players.

Billy Bissette

unread,
Sep 28, 2008, 10:50:58 PM9/28/08
to
Raymond Martineau <bk...@ncf.ca> wrote in
news:tfc0e4t7ah6ro3n31...@4ax.com:

> On Sun, 28 Sep 2008 09:41:14 +0000 (UTC), Radomir 'The Sheep'
> Dopieralski <ne...@sheep.art.pl> wrote:
>
>>Without these different commands items just feel like all those
>>objects you collect in a platform game -- they give you points,
>>sometimes they give you an additional life or some special move
>>or attack, but most of the time you don't need to look at them,
>>you just collect them all. Items in roguelike games are totally
>>different, even if there is some error made by the designer and
>>you have a potion and a scroll that do the same thing, they are
>>still different in the player's mind. This is an error, because
>>you shouldn't suggest things that are not true -- you break the
>>intuitive understanding of the game world this way.

The intuitive understanding of the game world is that when I want
my character to use an item, I want him to use an item. I don't
care whether he quaffs it, reads its, or waves it. I want it used,
I order it used, and if possible under the circumstances it gets
used.

That I might use the same key to use a potion as I would to read
a scroll doesn't make the potion less of a potion or the scroll less
of a scroll. I'm not going to go "It's pitch dark. Why can I still
use this potion but I cannot use this scroll?"

> I support a single 'u'se command. As you know, there isn't much
> difference between 'e'ating and 'q'uaffing, and extremely limited
> difference between 'z'apping and 'i'nvoking both wands or staves.
> Roguelikes mostly don't have an exclusive 'e' and 'z' - you won't
> need to eat a wand, and or zap a charge from food.
>
> In the character''s mind, scrolls and potions are different - you eat
> one, and read the other. If the character want's to use a potion,
> he's not going to accidently try to read it (unless hit with
> confusion). In the player's mind, it's unlikely he wants to do the
> said combination, and there's no reason to require memorizing the
> keyboard layout for the many flavours of use.
>
> The best example is Nethack seperating 'W'earing armour, and 'P'utting
> on accessories. There isn't much technical difference between the
> two, and merging them together isn't likely to confuse or annoy
> new/veteren players.

From what I recall, it took a while before the functionally
equivalent 'p'ray and 'm'agic were finally made interchangable in
Angband. You still had one function mapped to two keys, but at least
the game stopped complaining if you hit 'p' while playing a magic
user or 'm' while playing a priest.

(Both keys were kept because people who played prayer-based
characters were used to using "p" to cast while players of magic
users were used to using "m". As there was no decision over which
key to remove, nor even any desire to actually remove a key at all,
both were kept.)

Numeron

unread,
Sep 29, 2008, 2:25:48 AM9/29/08
to
Raymond Martineau wrote:
> In the character''s mind, scrolls and potions are different - you eat
> one, and read the other.  If the character want's to use a potion,
> he's not going to accidently try to read it (unless hit with
> confusion).  In the player's mind, it's unlikely he wants to do the
> said combination, and there's no reason to require memorizing the
> keyboard layout for the many flavours of use.
>
> The best example is Nethack seperating 'W'earing armour, and 'P'utting
> on accessories.  There isn't much technical difference between the
> two, and merging them together isn't likely to confuse or annoy
> new/veteren players.

In Nethack, i believe its possible to cast stone to flesh on a granite
wand to make a meat stick. I dont have it here on my work computer so
cant confirm whether you can still zap it, but there are occasions
where a single item can recieve the "default" use function of a
different item. You then have to provide the extra keypress to
differentiate that function which means you always need it for a
consistant UI.

That is, its silly being able to 'r'ead a t-shirt (as in nethack) and
not be able to 'r'ead a scroll. Likewise being polymorphed into a
metallivor, and having to 'e'at armor but 'u'se food. Its
inconsistant.

Being able to use a command for an item which is the default for a
different type of item is something probably quite frequent in RLs
once the item base grows and items become more complex. You then need
the extra key apart from use. To keep the consistancy you need that
seperate keypress always available for the command anyway. Use just
becomes one more button where the user not only has to memorise its
keybinding but memorise what function it is for every item, in turn
meaning 'U'se *increases* the complexity.

-Numeron

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 2:42:58 AM9/29/08
to
At Sun, 28 Sep 2008 21:50:58 -0500,
Billy Bissette wrote:

> The intuitive understanding of the game world is that when I want
> my character to use an item, I want him to use an item. I don't
> care whether he quaffs it, reads its, or waves it. I want it used,
> I order it used, and if possible under the circumstances it gets
> used.

Why do you want to use a potion, and not quaff it? Why would you use
a scroll, and not read it? Why would you use food instead of eating it?
Didn't your mom teach you to not play with food?

I believe your "intuition" in this case comes from experience with many
primitive, in terms of item interaction, games -- like platformers,
first person perspective shooters and, sadly, even adventure games.
But this concept is unnatural and you have specifically trained to make
it seem intuitive to you.

> That I might use the same key to use a potion as I would to read
> a scroll doesn't make the potion less of a potion or the scroll less
> of a scroll. I'm not going to go "It's pitch dark. Why can I still
> use this potion but I cannot use this scroll?"

You are not? What's stopping you? There is no mechanism that would hint
you that you cannot use a scroll in the dark, other than actually trying
it, is there? In similar manner, why should mummy not be able to "use" a
potion?

> From what I recall, it took a while before the functionally
> equivalent 'p'ray and 'm'agic were finally made interchangable in
> Angband. You still had one function mapped to two keys, but at least
> the game stopped complaining if you hit 'p' while playing a magic
> user or 'm' while playing a priest.

The next step is making "a magic user" character class, and getting
rid of "priests" and "mages". Of course Angband's spells and prayers
weren't different from each other in the first place -- so it seems
good that two mechanisms that have no practical or semantic difference
have been merged.

> (Both keys were kept because people who played prayer-based
> characters were used to using "p" to cast while players of magic
> users were used to using "m". As there was no decision over which
> key to remove, nor even any desire to actually remove a key at all,
> both were kept.)

That's the power of habituation. I still type "mad" when I want to
switch on the light in my room ;)

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 3:21:24 AM9/29/08
to
At Sun, 28 Sep 2008 21:59:33 -0400,
Raymond Martineau wrote:

> On Sun, 28 Sep 2008 09:41:14 +0000 (UTC), Radomir 'The Sheep'
> Dopieralski <ne...@sheep.art.pl> wrote:
>
>>Without these different commands items just feel like all those
>>objects you collect in a platform game -- they give you points,
>>sometimes they give you an additional life or some special move
>>or attack, but most of the time you don't need to look at them,
>>you just collect them all. Items in roguelike games are totally
>>different, even if there is some error made by the designer and
>>you have a potion and a scroll that do the same thing, they are
>>still different in the player's mind. This is an error, because
>>you shouldn't suggest things that are not true -- you break the
>>intuitive understanding of the game world this way.
>
> I support a single 'u'se command. As you know, there isn't much
> difference between 'e'ating and 'q'uaffing, and extremely limited
> difference between 'z'apping and 'i'nvoking both wands or staves.
> Roguelikes mostly don't have an exclusive 'e' and 'z' - you won't
> need to eat a wand, and or zap a charge from food.

As far as I know, there is considerable difference between eating
and quaffing. You quaff potions in order to heal or gain certain
temporary status effect. You eat food in order to satisfy hunger.
You eat monster corpses in order to gain intrinsics characteristic
for these monsters. In similar way, you zap a wand in order to
shoot a ranged magical attack at the enemy, and you invoke a staff
in order to cast a spell in certain area around you.
While in some games you might want to eat a wand or zap food, it's
not the point I have been making in the original post. The point
was that the effect you achieve is different in ways that affect
perception (targeting, timing, location, sense of self) and so they
should be invoked by motions that also feel differently, as to
improve the immersion and shorten the feedback loop between the
game and the player.

> In the character''s mind, scrolls and potions are different - you eat
> one, and read the other. If the character want's to use a potion,
> he's not going to accidently try to read it (unless hit with
> confusion). In the player's mind, it's unlikely he wants to do the
> said combination, and there's no reason to require memorizing the
> keyboard layout for the many flavours of use.

Game characters don't have minds, they don't perform cognitive
processing an have no way to "blend with a tool". They also have
no desires or wants, although the players are free to imagine otherwise.
The players, on the other hand, are very much affected by their bodies,
and the game interface may not only affect the efficiency and immersion,
but also the mental models used when thinking about the game. I'm
trying to argue that it's better to encourage mental models that are
more diverse and specific to actual in-game situations, rather than
very abstract and disconnected ones.

> The best example is Nethack seperating 'W'earing armour, and 'P'utting
> on accessories. There isn't much technical difference between the
> two, and merging them together isn't likely to confuse or annoy
> new/veteren players.

There is a huge technical difference between the two, and merging them
together is likely to confuse and annoy both new and veteran players.
Unlike you, instead of just stating this, I will also provide the
reasoning that leads to it, although I might be repeating what I already
wrote in the original post. This is OK, because I think that an example
might make it easier for you to understand.

Armor in Nethack has, except for a couple of unfortunate later additions,
no status effects -- only protection and resistance, and an occasional
sticky curse, of course. Accordingly, you try to wear a piece of armor
only when you want to affect your protection (and with the additions,
your invisibility, haggling, strength, intelligence, speed or ESP). Also,
when you are trying on a new piece of armor, you only expect to have
those features affected. Accessories have much wider selection of possible
effects, and usually don't affect the armor class -- you are much more
careful when trying them on and usually only change them when you want
some specific, special effect.

I think that the example of Nethack shows how the initial good design
of both user interface and game itself can go awry when people don't
actually understand why certain solutions work so well. Adding a cloak
of invisibility was, for example, driven solely by reference to legends,
not by idea on how to improve the playing experience -- form that point
of view it should have stayed a ring.

On the other hand, an occasional, rare exception to the rules makes the
game interesting and sparks our interest, so maybe it was added on
purpose after all? Only the developer can tell.

dpeg

unread,
Sep 29, 2008, 3:37:21 AM9/29/08
to
Raymond Martineau wrote:

> Radomir 'The Sheep' Dopieralski wrote:
>

[Three reasons why to distinguish use commands:
(1). interface filtering
(2). allowing several actions for some items
(3). mental differentiation from actions]

> I support a single 'u'se command. As you know, there isn't much
> difference between 'e'ating and 'q'uaffing, and extremely limited
> difference between 'z'apping and 'i'nvoking both wands or staves.
> Roguelikes mostly don't have an exclusive 'e' and 'z' - you won't
> need to eat a wand, and or zap a charge from food.

This addresses (2) only. In many roguelikes, wands are very attractive to
carry around. I am confident that any interface advantage gained from
reducing the number of commands (if there is an advantage) is lost if you
need to press another key within the commands more often. Taking 25 lines
as the default, combining some item types for 'u'sage, you may easily
surpass the screen threshold.

[In the case of Nethack, you could even polymorph into something (rock
mole?) that can eat wands.]

> In the character''s mind, scrolls and potions are different - you eat
> one, and read the other. If the character want's to use a potion,
> he's not going to accidently try to read it (unless hit with
> confusion). In the player's mind, it's unlikely he wants to do the
> said combination, and there's no reason to require memorizing the
> keyboard layout for the many flavours of use.

Radomir wrote a nice long article on this, and you ignored all this content
and reasoning. If you don't accept that different actions (i.e. command
keys, i.e. verbs) make the interface both faster and more comprehensible,
then why not simply conflating the item types? Instead of "potion of foo"
and "scroll of bar" you could use "item (scroll of bar)", "item (potion of
foo)". Question is not rhetorical -- single use command should at least
come with no item types.

> The best example is Nethack seperating 'W'earing armour, and 'P'utting
> on accessories. There isn't much technical difference between the
> two, and merging them together isn't likely to confuse or annoy
> new/veteren players.

Players usually carry around many rings and amulets. For filtering purposes
alone it is useful to distinguish these two.

David

Josh Sako

unread,
Sep 29, 2008, 3:51:30 AM9/29/08
to

> Why do you want to use a potion, and not quaff it? Why would you use
> a scroll, and not read it? Why would you use food instead of eating it?
> Didn't your mom teach you to not play with food?

What you are essentially arguing for here is semantic distinction.
That is, you want every item to have a semantic action equivalent to
its function. When people argue for a single key, they want function
and that's all; there is absolutely no difference, functionally,
between waving a wand, quaffing a potion, and eating food. It's all
about semantics.

Frankly, I'm far more with the functionalists here. It's a far simpler
means of accomplishing the same tasks and simplicity is always better.
Always[1].

[1] Before someone brings up the straw-man about "well, why not just
have a game where the player does nothing but watch, that's the
simplest!", obviously I am referring to accomplishing a specific task
(utilizing an item in a default case). A more general statement should
be: everything should be as simple as possible, but no simpler.

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 4:16:32 AM9/29/08
to
At Mon, 29 Sep 2008 00:51:30 -0700 (PDT),
Josh Sako wrote:

>
>> Why do you want to use a potion, and not quaff it? Why would you use
>> a scroll, and not read it? Why would you use food instead of eating it?
>> Didn't your mom teach you to not play with food?
>
> What you are essentially arguing for here is semantic distinction.
> That is, you want every item to have a semantic action equivalent to
> its function. When people argue for a single key, they want function
> and that's all; there is absolutely no difference, functionally,
> between waving a wand, quaffing a potion, and eating food. It's all
> about semantics.

It's about grouping functions in different categories of items, just
like we categorize the objects around us. This process is natural and
is part of the process of understanding the world.

It is important to make this semantic distinction consistent with the
semantics of the game world, though.

> Frankly, I'm far more with the functionalists here. It's a far simpler
> means of accomplishing the same tasks and simplicity is always better.

Flat filesystem is much simpler than one with directories. Why do you
use directories in your filesystem?

You are arguing that having just one category of items is simpler than
having them neatly divided into different categories by how they are
actually used in the game. This might be true for a developer: it may
be simpler to implement it this way. I doubt it's better for the player,
at least as long as the items are really different: there is no scroll
of healing and armor of invisibility. Of course, once you start mix
the effects taht belong to different categories, the categories are no
help anymore and they get in the way more than they help.

It so happens that "simple" doesn't always mean the same thing. Our
bodies are optimized to operate in a pretty complicated world, and
thus what is "simple" for them and what is "hard" doesn't match the
same notions transfered to abstract, idealized world. The interface
we are talking about works directly with our nervous system at very
low level, and thus has to be complicated the same way as the
"human API" is complicated. It doesn't work at the abstract, high
level of "pure thought" -- instead you deal with dirty, bloody meat.

Darren Grey

unread,
Sep 29, 2008, 10:13:38 AM9/29/08
to
On Sep 29, 8:21 am, Radomir 'The Sheep' Dopieralski

<n...@sheep.art.pl> wrote:
> At Sun, 28 Sep 2008 21:59:33 -0400,
>
> Raymond Martineau wrote:
> > On Sun, 28 Sep 2008 09:41:14 +0000 (UTC), Radomir 'The Sheep'
> > Dopieralski <n...@sheep.art.pl> wrote:
>
> >>Without these different commands items just feel like all those
> >>objects you collect in a platform game -- they give you points,
> >>sometimes they give you an additional life or some special move
> >>or attack, but most of the time you don't need to look at them,
> >>you just collect them all. Items in roguelike games are totally
> >>different, even if there is some error made by the designer and
> >>you have a potion and a scroll that do the same thing, they are
> >>still different in the player's mind. This is an error, because
> >>you shouldn't suggest things that are not true -- you break the
> >>intuitive understanding of the game world this way.
>
> > I support a single 'u'se command. As you know, there isn't much
> > difference between 'e'ating and 'q'uaffing, and extremely limited
> > difference between 'z'apping and 'i'nvoking both wands or staves.
> > Roguelikes mostly don't have an exclusive 'e' and 'z' - you won't
> > need to eat a wand, and or zap a charge from food.
>
> As far as I know, there is considerable difference between eating
> and quaffing. You quaff potions in order to heal or gain certain
> temporary status effect. You eat food in order to satisfy hunger.
> You eat monster corpses in order to gain intrinsics characteristic
> for these monsters. In similar way, you zap a wand in order to
> shoot a ranged magical attack at the enemy, and you invoke a staff
> in order to cast a spell in certain area around you.

Well, let's be clear here, it's quite common in roguelikes for people
to read a scroll or drink a potion for the purposes of satiation.
Similarly wands and rings and armour and weapons can all have a
variety of effects. In an environment with magical enchantments on
many things you literally can have any item capable of anything, up to
the whim of the developer (not always good, but it happens anyway).
Of course this in a way makes multiple usage commands a good thing,
because the player has to think out of the box on how to apply his
items correctly for the circumstance. The player will feel rewarded
and proud of himself if, running out of food, he remembers he can down
a few potions to keep him from starving for a while longer. In this
way the multiple item commands become a puzzle game of their own, and
the additional immersion that you describe becomes heightened when the
player is being inventive to solve certain scenarios. He has to
overcome his usual command associations, much like the character has
to do something drastic or different to survive.

> The players, on the other hand, are very much affected by their bodies,
> and the game interface may not only affect the efficiency and immersion,
> but also the mental models used when thinking about the game. I'm
> trying to argue that it's better to encourage mental models that are
> more diverse and specific to actual in-game situations, rather than
> very abstract and disconnected ones.

Well, to be frank, this is all just conjecture. And you really can't
speak for all players - everyone thinks differently to some degree,
and for many a single use command seems more immersive than having to
remember lots of different artifically thought up keypresses. If a
player has to sit and think of what key he's meant to press for some
simple action then the immersion is wrecked. With experience in the
game he/she will obviously get used to it and it will become natural,
but that can take a while. This is made worse when you consider that
many roguelikes use different key commands for different things - drop
in one is drink in another, etc.

By far the biggest problem I have in getting other people (even
experienced gamers) to play roguelikes is the insane number of
keypresses you need to know from the very start. You need to be
dedicated to the game from the get-go to ever be able to get into it
and get a taste of it. The sort of system that Gelatinous Mutant
Coconut described at least allows players to learn the commands
through a simpler interface, even if it's not ideal for experienced
play.

> > The best example is Nethack seperating 'W'earing armour, and 'P'utting
> > on accessories. There isn't much technical difference between the
> > two, and merging them together isn't likely to confuse or annoy
> > new/veteren players.
>

> There is a huge technical difference between the two, and merging them


> together is likely to confuse and annoy both new and veteran players.
> Unlike you, instead of just stating this, I will also provide the
> reasoning that leads to it, although I might be repeating what I already
> wrote in the original post. This is OK, because I think that an example
> might make it easier for you to understand.

Man, I can't believe you're actually defending that... For most
players wearing or equipping items is a global action that can produce
different effects from a variety of items. The vast majority of games
(mainstream and roguelike) have a single command or interface for
equipping things. To separate it into numerous commands becomes an
annoyance - and an illogical one at that. The simple fact is that
many people choose not to play Nethack because of its rather horrible
inventory interface - shouldn't those real complaints about the game
count more than conjectured justifications?

> I think that the example of Nethack shows how the initial good design
> of both user interface and game itself can go awry when people don't
> actually understand why certain solutions work so well. Adding a cloak
> of invisibility was, for example, driven solely by reference to legends,
> not by idea on how to improve the playing experience -- form that point
> of view it should have stayed a ring.

Surely the ring of invisibility is just a reference to legends/fiction
too? There's ONE RING in particular I'm thinking of ;) Logically
any item can be imbued with magical properties - rings in general are
more for convenience I imagine. Unless you invent some mythos of only
certain concentrated metals being able to hold magic, etc. A cloak of
invisiblity wrapping around a character is actually more intuitive to
think of than a ring that gives some invisible aura. Some items
associations don't always work well, but I think it's wrong to say
that certains items *must* behave in certain ways. I'm not a big fan
of Nethack's kitchen sink philosophy, but I still like to see creative
and imaginative item ideas in games.

--
Darren Grey

David Damerell

unread,
Sep 29, 2008, 11:23:31 AM9/29/08
to
Quoting Billy Bissette <bai...@coastalnet.com>:
>of a scroll. I'm not going to go "It's pitch dark. Why can I still
>use this potion but I cannot use this scroll?"

An advantage of specific commands is that they can tell the player their
action is impossible sooner. "It's too dark to read anything."
--
David Damerell <dame...@chiark.greenend.org.uk> Kill the tomato!
Today is First Brieday, September.

David Damerell

unread,
Sep 29, 2008, 11:20:52 AM9/29/08
to
Quoting Josh Sako <jgi...@gmail.com>:
>Frankly, I'm far more with the functionalists here. It's a far simpler
>means of accomplishing the same tasks and simplicity is always better.

What is simplest depends largely on how big the inventory is. If you have
two scrolls and three potions and "use" gives you a list of five, fine. If
you have fifty-three, maybe you'd like them subdivided.

Christophe

unread,
Sep 29, 2008, 11:51:38 AM9/29/08
to
dpeg a écrit :

> Radomir wrote a nice long article on this, and you ignored all this content
> and reasoning. If you don't accept that different actions (i.e. command
> keys, i.e. verbs) make the interface both faster and more comprehensible,
> then why not simply conflating the item types? Instead of "potion of foo"
> and "scroll of bar" you could use "item (scroll of bar)", "item (potion of
> foo)". Question is not rhetorical -- single use command should at least
> come with no item types.

Well, because potions can break because of cold, scrolls can burn
because of fire. Also, it has been mentioned that scrolls cannot be used
without a light source. Try to explain to the player why he can lose his
"item of healing" to an ice elemental attack but not his "item of
teleportation", while the first one can be used inthe dak and not the
second. And this is even before you decide to add a "Potion of Phase
Door" in your game.

You don't need a strong semantics on the interface to make the player
aware of the strong difference between the items themselves.

In Angband, we have:
- scrolls: need light to use, vulnerable to fire, not tied to magic
device skill (100% succes rate when used)
- potions: vulnerable to cold, effects self only, not tied to magic
device skill, bad idea to use when you are too full (tied to eating system)
- wands: vulnerable to lighting, includes only effects that need aiming,
limited number of charges
- staffs: vulnerable to fire, doesn't include any effect that needs
aiming, heavy, limited number of charges
- rods: invulnerable, includes any existing effect type, only 1 charge
but it regenerates on it's own after a number of turns

Frankly, those elements are much more important for the player than the
shortcut key used to use them.

>> The best example is Nethack seperating 'W'earing armour, and 'P'utting
>> on accessories. There isn't much technical difference between the
>> two, and merging them together isn't likely to confuse or annoy
>> new/veteren players.
>
> Players usually carry around many rings and amulets. For filtering purposes
> alone it is useful to distinguish these two.

Unless they carry around many rings and amulets AND armors, it doesn't
filter much when you use P ;)

Gelatinous Mutant Coconut

unread,
Sep 29, 2008, 11:54:12 AM9/29/08
to
On Sep 29, 8:20 am, David Damerell <damer...@chiark.greenend.org.uk>
wrote:

> Quoting  Josh Sako  <jgi...@gmail.com>:
>
> What is simplest depends largely on how big the inventory is. If you have
> two scrolls and three potions and "use" gives you a list of five, fine. If
> you have fifty-three, maybe you'd like them subdivided.

I think this actually comes down to game design, then; you can easily
get away with just a "use" command if the player is only going to have
a few items available for manipulation at a time, but if their
characters are becoming roving magical item warehouses, you really
should have fine-grain selection commands. (Or maybe implement some
kind of limit on what your players can carry; depends what kind of
gameplay you want.)

A single use command seems necessary for games designed for systems
with a key shortage, like most consoles. (It's been a while since I
played POWDER, and I only tried the PC version, but I believe the GBA
version used a single use command out of necessity?)

Pointless

unread,
Sep 29, 2008, 12:05:09 PM9/29/08
to
On Sep 29, 4:16 am, Radomir 'The Sheep' Dopieralski
<n...@sheep.art.pl> wrote:
>...............................................................Our

> bodies are optimized to operate in a pretty complicated world, and
> thus what is "simple" for them and what is "hard" doesn't match the
> same notions transfered to abstract, idealized world. The interface
> we are talking about works directly with our nervous system at very
> low level, and thus has to be complicated the same way as the
> "human API" is complicated. It doesn't work at the abstract, high
> level of "pure thought" -- instead you deal with dirty, bloody meat.
>
> --
> Radomir `The Sheep' Dopieralski <http://sheep.art.pl>
>   "Whenever you find yourself on the side of the majority,
>    it's time to pause and reflect." -- Mark Twain

New idea for a Roguelike, SamadhiRL? User interface can be driven by
pure thought. If you think "bad thoughts", they are produced in the
game world. First one to reach enlightenment wins!

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 12:18:42 PM9/29/08
to
At Mon, 29 Sep 2008 07:13:38 -0700 (PDT),
Darren Grey wrote:

>> The players, on the other hand, are very much affected by their bodies,
>> and the game interface may not only affect the efficiency and immersion,
>> but also the mental models used when thinking about the game. I'm
>> trying to argue that it's better to encourage mental models that are
>> more diverse and specific to actual in-game situations, rather than
>> very abstract and disconnected ones.

> Well, to be frank, this is all just conjecture. And you really can't
> speak for all players - everyone thinks differently to some degree,
> and for many a single use command seems more immersive than having to
> remember lots of different artifically thought up keypresses.

While I may agree that everyone may think differently to some degree --
I'm even pretty sure that the same person will think differently at
different times -- this is not something dependent on thinking, at least
not on the level that differs considerably between individuals.
Things like habituation, reflex building, interpretation of the
surroundings, to which I refer, are all pretty much build into our
bodies and our neural systems and cannot be easily changed, unless you
were rised in 6-dimensional quasi-space on Alpha Centuauri, or you
have gummibears instead of neurons.

Sure, the theories I rely on are still only theories: mostly because
there are many ways to explain what is happening. But what is happening
is actually pretty much obvious, and there is no disagreement on the
actual results of these mechanisms.

Lastly, you don't really remember the keypresses -- just like you don't
really remember where a particular key is on the keyboard. Your hands
remember (or more accurately, your hands and spinal cord).

> If a
> player has to sit and think of what key he's meant to press for some
> simple action then the immersion is wrecked. With experience in the
> game he/she will obviously get used to it and it will become natural,
> but that can take a while.

It takes much less time than you are making it sound. On the magnitude
of three or four games, not hundrends of hours of training. We do it
all the time with each and every device we take into our hands.
*Especially* with games. Even more, this habituation is actually part
of the fun we get from most reflex-based games.

> This is made worse when you consider that
> many roguelikes use different key commands for different things - drop
> in one is drink in another, etc.

Not really. People who have learned to drive a car have no problem to
go back to riding a bike -- except for maybe some high-level thinking.
Similarly, you can learn both ASDF and Dvorak keyborad layout and use
both of them, switching when necessary -- one does not interfere with
the other as long as there is enough context to know which one to use.

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 12:22:47 PM9/29/08
to
At Mon, 29 Sep 2008 09:05:09 -0700 (PDT),
Pointless wrote:

> On Sep 29, 4:16 am, Radomir 'The Sheep' Dopieralski
> <n...@sheep.art.pl> wrote:
>>...............................................................Our
>> bodies are optimized to operate in a pretty complicated world, and
>> thus what is "simple" for them and what is "hard" doesn't match the
>> same notions transfered to abstract, idealized world. The interface
>> we are talking about works directly with our nervous system at very
>> low level, and thus has to be complicated the same way as the
>> "human API" is complicated. It doesn't work at the abstract, high
>> level of "pure thought" -- instead you deal with dirty, bloody meat.
>

> New idea for a Roguelike, SamadhiRL? User interface can be driven by
> pure thought. If you think "bad thoughts", they are produced in the
> game world. First one to reach enlightenment wins!

Let's split the work. I can make the game if you provide the hardware
for the user interface.

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 12:29:44 PM9/29/08
to
At Mon, 29 Sep 2008 17:51:38 +0200,
Christophe wrote:

> You don't need a strong semantics on the interface to make the player
> aware of the strong difference between the items themselves.

[...]

> Frankly, those elements are much more important for the player than the
> shortcut key used to use them.

The point I'm trying to make is that using physically different movements
for them emphasizes that they are different and actually helps us to
categorize them as separate classes, by providing "groudning" for the
pretty abstract rules of the game. It's not a secret that physical tokens
aid in many mental processes. Operators of nuclear plants often put
different knobs on the otherwise identical levers, just to make them less
likeley to mix up in a fatal accident.

Radomir 'The Sheep' Dopieralski

unread,
Sep 29, 2008, 4:09:32 PM9/29/08
to
At Mon, 29 Sep 2008 14:57:39 -0500,
Paul Donnelly wrote:

> Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> writes:
>
>> Not really. People who have learned to drive a car have no problem to
>> go back to riding a bike -- except for maybe some high-level thinking.
>> Similarly, you can learn both ASDF and Dvorak keyborad layout and use
>> both of them, switching when necessary -- one does not interfere with
>> the other as long as there is enough context to know which one to use.

> Muscle memory does not work that way. If some cars were operated like
> a bike, but their controls appeared identical to a regular car's, a
> lot of people would die. Several times I have pressed ^w in Firefox,
> trying to delete a word, only to lose what I'm looking at. I do the
> same for many other hotkeys. We might technically be able to compare
> context to decide which to use, but at the unconscious level where the
> real work gets done, that's not going to happen reliably.

Obviously the context needs also to be pretty low-level, physical,
in order to affect reflexes. In case of bicycle/car it's the position
of the body, the shape of controls, the open/close cabin, etc. -- all
those are physical hints for your unconsciousness. I'm not sure what
does it for the ASDF/Dvorak distinction -- I have to look at it more
closely, I guess.

I can agree though that inventing a completely novel interface for
well-known and trained situations is going to lead to ambiguity. How
fortunate that the solutions I've been advocating are kind of traditional
in this genre since decades :) Yet another reason to stick to them.

Billy Bissette

unread,
Sep 29, 2008, 5:45:36 PM9/29/08
to
Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
news:slrnge0u7...@atos.wmid.amu.edu.pl:

> At Sun, 28 Sep 2008 21:50:58 -0500,
> Billy Bissette wrote:
>
>> The intuitive understanding of the game world is that when I want
>> my character to use an item, I want him to use an item. I don't
>> care whether he quaffs it, reads its, or waves it. I want it used,
>> I order it used, and if possible under the circumstances it gets
>> used.
>
> Why do you want to use a potion, and not quaff it? Why would you use
> a scroll, and not read it? Why would you use food instead of eating
it?
> Didn't your mom teach you to not play with food?
>
> I believe your "intuition" in this case comes from experience with
many
> primitive, in terms of item interaction, games -- like platformers,
> first person perspective shooters and, sadly, even adventure games.
> But this concept is unnatural and you have specifically trained to
make
> it seem intuitive to you.

What is the difference between hitting "q" to drink a potion and
"u" to drink a potion? Because "q" can stand for "quaff"? Why not
"d" for "drink"? Oh, because "d" is probably "drop".

If it is the wording itself, how many games that use "q" for
"quaff" even display the action as "quaff" anywhere other than the
command display screen? Most probably say something like "You drink
the icky green potion."

I used the word "use" to make the distinction to the readers of my
post.

I personally have no problem with hitting the same key to use
(drink) a potion as I do to use (read) a scroll. I doubt many
people would get horribly confused.

As for the imaginary action itself, that doesn't really matter.
I don't really care whether the game claims I read a scroll or
lick it and stick it to my body like a temporary tattoo. I just
want to use it for whatever effect it produces. I hit an arbitrary
key designated to use an item (and "q" is just as arbitrary as "u"),
the game reports back the results of trying to use that item, and
it probably even tells me how I used it (read it, drank it, scratched
the shiny silver bit off with a quarter. whatever).

Damien Neil

unread,
Sep 29, 2008, 5:47:02 PM9/29/08
to
Raymond Martineau <bk...@ncf.ca> wrote:
> I support a single 'u'se command. As you know, there isn't much
> difference between 'e'ating and 'q'uaffing, and extremely limited
> difference between 'z'apping and 'i'nvoking both wands or staves.
> Roguelikes mostly don't have an exclusive 'e' and 'z' - you won't
> need to eat a wand, and or zap a charge from food.

In NetHack , quaffing always takes one turn. Eating generally takes
more than one turn. That's a *huge* difference.

Plus, of course, you can polymorph into a metalvore and eat, say, a ring.

- Damien

Billy Bissette

unread,
Sep 29, 2008, 6:01:14 PM9/29/08
to
Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
news:slrnge20j...@atos.wmid.amu.edu.pl:

> At Mon, 29 Sep 2008 17:51:38 +0200,
> Christophe wrote:
>
>> You don't need a strong semantics on the interface to make the player
>> aware of the strong difference between the items themselves.
>
> [...]
>
>> Frankly, those elements are much more important for the player than
>> the shortcut key used to use them.
>
> The point I'm trying to make is that using physically different
movements
> for them emphasizes that they are different and actually helps us to
> categorize them as separate classes, by providing "groudning" for the
> pretty abstract rules of the game.

They *aren't* that different.

Scrolls and potions are one-shot items. (Generally scrolls are in
Roguelikes. Hrm, I wonder how many Roguelikes ever bother to explain
to the player that scrolls are one-shot items?)

Wands, staves, and rods are magical objects that can be used
repeatedly.

Wands and staves had charges.

Variations depend upon the game world, though it is fairly
consistent.

All of the above attempt to do one specific thing when you use
them via whatever action triggers their primary usage. That action
is generally considered magical in nature.


That is is called a "Staff of Magic Mapping," weighs ten times as
much as a potion, stacks different than potions, and tends to burst
into flames when I stand next to a dragon is generally enough for
me to recognize it as a different kind of item than an "Icky Green
Potion."

Billy Bissette

unread,
Sep 29, 2008, 6:04:26 PM9/29/08
to
Christophe <chris.c...@free.fr> wrote in
news:48e0f98a$0$30967$426a...@news.free.fr:

> Well, because potions can break because of cold, scrolls can burn
> because of fire. Also, it has been mentioned that scrolls cannot be
> used without a light source. Try to explain to the player why he can
> lose his "item of healing" to an ice elemental attack but not his
> "item of teleportation", while the first one can be used inthe dak and
> not the second. And this is even before you decide to add a "Potion of
> Phase Door" in your game.

Heaven forbid they wonder why cold can destroy a potion but extreme
heat cannot. :)

Billy Bissette

unread,
Sep 29, 2008, 6:11:35 PM9/29/08
to
Darren Grey <darrenj...@gmail.com> wrote in news:3bc7f2f4-7dc5-44f1-
b4e4-186...@t54g2000hsg.googlegroups.com:

> On Sep 29, 8:21 am, Radomir 'The Sheep' Dopieralski
> <n...@sheep.art.pl> wrote:
>> At Sun, 28 Sep 2008 21:59:33 -0400, Raymond Martineau wrote:

>> > The best example is Nethack seperating 'W'earing armour, and
'P'utting
>> > on accessories. There isn't much technical difference between the
>> > two, and merging them together isn't likely to confuse or annoy
>> > new/veteren players.
>>
>> There is a huge technical difference between the two, and merging
them
>> together is likely to confuse and annoy both new and veteran players.
>> Unlike you, instead of just stating this, I will also provide the
>> reasoning that leads to it, although I might be repeating what I
>> already wrote in the original post. This is OK, because I think that
>> an example might make it easier for you to understand.
>
> Man, I can't believe you're actually defending that... For most
> players wearing or equipping items is a global action that can produce
> different effects from a variety of items. The vast majority of games
> (mainstream and roguelike) have a single command or interface for
> equipping things. To separate it into numerous commands becomes an
> annoyance - and an illogical one at that. The simple fact is that
> many people choose not to play Nethack because of its rather horrible
> inventory interface - shouldn't those real complaints about the game
> count more than conjectured justifications?

Indeed, maybe there should be more separate commands. Why not a
specific key to put on a helmet versus putting on a chest plate?
Those are two different actions, affecting two different items, and
putting them in two different locations. Same for gloves, boots,
belts, shields, weapons, rings, earrings, nose rings, toe rings,
thumb rings, cloaks, capes, money pouches, backpacks, ...

The only down side is that it will confuse and annoy more new
players, but I'm sure there will be experienced players to defend it
once they learn it.

Martin Read

unread,
Sep 29, 2008, 6:18:35 PM9/29/08
to
Billy Bissette <bai...@coastalnet.com> wrote:
> Scrolls and potions are one-shot items. (Generally scrolls are in
>Roguelikes. Hrm, I wonder how many Roguelikes ever bother to explain
>to the player that scrolls are one-shot items?)

Spellscrolls and potions being one-shot items is a fairly standard fantasy
gaming trope.
--
\_\/_/ turbulence is certainty turbulence is friction between you and me
\ / every time we try to impose order we create chaos
\/ -- Killing Joke, "Mathematics of Chaos"

Paul Donnelly

unread,
Sep 29, 2008, 3:57:39 PM9/29/08
to
Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> writes:

> Not really. People who have learned to drive a car have no problem to
> go back to riding a bike -- except for maybe some high-level thinking.
> Similarly, you can learn both ASDF and Dvorak keyborad layout and use
> both of them, switching when necessary -- one does not interfere with
> the other as long as there is enough context to know which one to use.

Muscle memory does not work that way. If some cars were operated like

Radomir 'The Sheep' Dopieralski

unread,
Sep 30, 2008, 2:57:37 AM9/30/08
to
At Mon, 29 Sep 2008 17:01:14 -0500,
Billy Bissette wrote:

> Scrolls and potions are one-shot items. (Generally scrolls are in
> Roguelikes. Hrm, I wonder how many Roguelikes ever bother to explain
> to the player that scrolls are one-shot items?)

I remember Rogue (or was it Hack?) mentioning something about the
scroll turning into dust after reading it.

Krice

unread,
Sep 30, 2008, 3:00:09 AM9/30/08
to
On 30 syys, 01:11, Billy Bissette <bai...@coastalnet.com> wrote:
> Indeed, maybe there should be more separate commands. Why not a
> specific key to put on a helmet versus putting on a chest plate?

Use vs. separate keys can have less dramatic results, because
let's not forget that the world offers more than two ways to
do things. I don't have use command in Kaduria, but then I
haven't mapped the entire keyboard for commands either.

Soyweiser

unread,
Sep 30, 2008, 8:18:24 AM9/30/08
to
On Sep 30, 8:57 am, Radomir 'The Sheep' Dopieralski

Which I still think is great. In a complicated game such as the
average roguelike (not the CbreakRLs) it is odd that scrolls and
potions disappear after use. But from the gameplay perspective it
makes sense, else you would have even more junk items, or break the
game balance. (In NonExistRL corpses leave skeletons when eaten. A
quaffed potion becomes a empty potion, tins leave empty tins. Gifts
from gods come in wrappers, A wall that is dug creates a lot of sand
and stones etc. NonExistRL also has a GarbageDemon, garbage bins, the
special garbage dump level, and overweight gelatinous cubes.)

But I would still like to see used items not disappear. Provided you
could still do interesting stuff with the garbage remains. Example: An
empty potion, can be filled with water, water can be made in a potion
of healing using an alchemist kit and some healing leaves. And empty
tins can be refilled, or melted down for the metal in the tin etc. But
this gets closer to something like Dwarf Fortress or a MMORPG crafting
system. Which are normally not used in RL's.

--
Soyweiser

Raymond Martineau

unread,
Sep 30, 2008, 12:10:36 PM9/30/08
to
On Mon, 29 Sep 2008 06:42:58 +0000 (UTC), Radomir 'The Sheep'
Dopieralski <ne...@sheep.art.pl> wrote:

>At Sun, 28 Sep 2008 21:50:58 -0500,
> Billy Bissette wrote:
>
>> The intuitive understanding of the game world is that when I want
>> my character to use an item, I want him to use an item. I don't
>> care whether he quaffs it, reads its, or waves it. I want it used,
>> I order it used, and if possible under the circumstances it gets
>> used.
>
>Why do you want to use a potion, and not quaff it? Why would you use
>a scroll, and not read it? Why would you use food instead of eating it?
>Didn't your mom teach you to not play with food?

When most normal people "use" food/potions/books/gloves, they would
eat/drink/read/wear them.

In Rogue, it's possible to write a script involving a unified use
command - if you select a given item, the script can always determing
the correct action since there is only one way to handle a given
object (almost like all the platform games you mention.)

In Nethack, a unified use command would not work. For wands, you can
either 'z'ap them or break them. For potions, you can either drink,
throw or dip them. For markers, you can write a new scroll, or write
grafitti. (Speaking of which, Nethack has an 'a'pply command - this
is an appearance of a catch-all statement if none of the other
commands apply.)

>
>> That I might use the same key to use a potion as I would to read
>> a scroll doesn't make the potion less of a potion or the scroll less
>> of a scroll. I'm not going to go "It's pitch dark. Why can I still
>> use this potion but I cannot use this scroll?"
>
>You are not? What's stopping you? There is no mechanism that would hint
>you that you cannot use a scroll in the dark, other than actually trying
>it, is there? In similar manner, why should mummy not be able to "use" a
>potion?

When you attempt to read a scroll in the dark in Angband, the 'r'ead
command will stop right away. The unified use command doesn't have to
be different - it can simply not list the items in question if they
can't be used, and if you attempt to use them anyway, you get an error
message (i.e. you can't read the scroll in the dark.)

>
>> From what I recall, it took a while before the functionally
>> equivalent 'p'ray and 'm'agic were finally made interchangable in
>> Angband. You still had one function mapped to two keys, but at least
>> the game stopped complaining if you hit 'p' while playing a magic
>> user or 'm' while playing a priest.
>
>The next step is making "a magic user" character class, and getting
>rid of "priests" and "mages". Of course Angband's spells and prayers
>weren't different from each other in the first place -- so it seems
>good that two mechanisms that have no practical or semantic difference
>have been merged.

This was actually done in CthAngband and Sangband. However, there are
independant spell trees and trying to max them out and use them at the
same time would fill your inventory.

Angband won't merge the two classes, since they are modelled after a
red-box style of Magic users and clerics - one uses arcane magic
(which is traditionally an towards agression and power), and the other
uses divine magic (traditionally towards healing and defense.) The
spell progression between these two trees may ultimatly give the same
power, but will affect how the character is played.

Radomir 'The Sheep' Dopieralski

unread,
Sep 30, 2008, 12:56:14 PM9/30/08
to
At Tue, 30 Sep 2008 12:10:36 -0400,
Raymond Martineau wrote:

> On Mon, 29 Sep 2008 06:42:58 +0000 (UTC), Radomir 'The Sheep'
> Dopieralski <ne...@sheep.art.pl> wrote:
>
>>At Sun, 28 Sep 2008 21:50:58 -0500,
>> Billy Bissette wrote:

>>> That I might use the same key to use a potion as I would to read
>>> a scroll doesn't make the potion less of a potion or the scroll less
>>> of a scroll. I'm not going to go "It's pitch dark. Why can I still
>>> use this potion but I cannot use this scroll?"
>>
>>You are not? What's stopping you? There is no mechanism that would hint
>>you that you cannot use a scroll in the dark, other than actually trying
>>it, is there? In similar manner, why should mummy not be able to "use" a
>>potion?
>
> When you attempt to read a scroll in the dark in Angband, the 'r'ead
> command will stop right away. The unified use command doesn't have to
> be different - it can simply not list the items in question if they
> can't be used, and if you attempt to use them anyway, you get an error
> message (i.e. you can't read the scroll in the dark.)
>
>>> From what I recall, it took a while before the functionally
>>> equivalent 'p'ray and 'm'agic were finally made interchangable in
>>> Angband. You still had one function mapped to two keys, but at least
>>> the game stopped complaining if you hit 'p' while playing a magic
>>> user or 'm' while playing a priest.

The problem with just quietly removing the "readable" items from the list
is that you still don't know why they have been removed, and often that
they have been removed at all. This leads to those surprising moments of
"wait a minute, didn't I have a scroll of teleportation?", followed by
"it's displaying in the inventory, why isn't it displaying in the use
command" and finally "right, I forgot that they wrote in the manual that
you cannot use scrolls in darkness".

This is a problem with error reporting (and other forms of feedback)
common to situations that introduce layers of indirection -- ones that
select one of several possible commands, for example.

Billy Bissette

unread,
Sep 30, 2008, 7:31:55 PM9/30/08
to
Soyweiser <soyw...@gmail.com> wrote in news:dfd76e41-e70c-40f7-b822-
5de55b...@l64g2000hse.googlegroups.com:

> But I would still like to see used items not disappear. Provided you
> could still do interesting stuff with the garbage remains. Example: An
> empty potion, can be filled with water, water can be made in a potion
> of healing using an alchemist kit and some healing leaves. And empty
> tins can be refilled, or melted down for the metal in the tin etc. But
> this gets closer to something like Dwarf Fortress or a MMORPG crafting
> system. Which are normally not used in RL's.

That would depend a lot upon the game, particularly the number of
items found and their usefulness. Inventory limitations are a killer.

One of Angband's big problems is considered "Too Much Junk," as even
the majority of supposedly "useful" items are considered junk by the
average player. You've got people squelching entire groups of objects,
including some magical ones, just because whatever is possible in that
group just isn't going to be useful for them.

On the other hand, if you have a fairly unlimited inventory and
decent uses for nearly everything, then leftovers can be useful.

Billy Bissette

unread,
Sep 30, 2008, 7:53:50 PM9/30/08
to
Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
news:slrnge4mh...@atos.wmid.amu.edu.pl:

> The problem with just quietly removing the "readable" items from the list
> is that you still don't know why they have been removed, and often that
> they have been removed at all. This leads to those surprising moments of
> "wait a minute, didn't I have a scroll of teleportation?", followed by
> "it's displaying in the inventory, why isn't it displaying in the use
> command" and finally "right, I forgot that they wrote in the manual that
> you cannot use scrolls in darkness".

Don't remove them. Let the player pick them, and then explain the
failure.

If you want to use color codes, display items that are only currently
unusable in a different color, and still allow them to be picked
(giving the player a way to find out why they are currently unusable,
if they haven't already noticed the reason.)

Radomir 'The Sheep' Dopieralski

unread,
Oct 1, 2008, 7:53:03 AM10/1/08
to
At Tue, 30 Sep 2008 18:53:50 -0500,
Billy Bissette wrote:

> Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
> news:slrnge4mh...@atos.wmid.amu.edu.pl:
>
>> The problem with just quietly removing the "readable" items from the list
>> is that you still don't know why they have been removed, and often that
>> they have been removed at all. This leads to those surprising moments of
>> "wait a minute, didn't I have a scroll of teleportation?", followed by
>> "it's displaying in the inventory, why isn't it displaying in the use
>> command" and finally "right, I forgot that they wrote in the manual that
>> you cannot use scrolls in darkness".

> Don't remove them. Let the player pick them, and then explain the
> failure.

It still has the taste of letting the user fill in three pages of
forms just to inform them, at the end, about some error, and clear
the form...

Billy Bissette

unread,
Oct 1, 2008, 2:53:13 PM10/1/08
to
Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
news:slrnge6p4...@atos.wmid.amu.edu.pl:

> At Tue, 30 Sep 2008 18:53:50 -0500,
> Billy Bissette wrote:
>
>> Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
>> news:slrnge4mh...@atos.wmid.amu.edu.pl:
>>
>>> The problem with just quietly removing the "readable" items from the
>>> list is that you still don't know why they have been removed, and
>>> often that they have been removed at all. This leads to those
>>> surprising moments of "wait a minute, didn't I have a scroll of
>>> teleportation?", followed by "it's displaying in the inventory, why
>>> isn't it displaying in the use command" and finally "right, I forgot
>>> that they wrote in the manual that you cannot use scrolls in
>>> darkness".
>
>> Don't remove them. Let the player pick them, and then explain the
>> failure.
>
> It still has the taste of letting the user fill in three pages of
> forms just to inform them, at the end, about some error, and clear
> the form...

That's why I like the idea of color-coding the item display to show
what will work and what for some reason will not.

Another idea is to put some kind of tag in the item display list
that shows what action will be taken or why it currently cannot be
done.
Like:
a) [drink] potion of healing
b) [read] scroll of phase door
c) [aim] wand of striking
Or:
a) X[full] potion of healing
b) X[blind] scroll of phase door

Though for the latter, you'd need to prioritize failure conditions,
else you might end up with a string of them displayed. Like if the
character were blind while it was dark, you'd just display "blind".
(Heck, the character wouldn't even know it was actually dark until
their blindness was cured anyway.) Not sure how to rank others
though, like blind versus confused.

Gelatinous Mutant Coconut

unread,
Oct 2, 2008, 1:50:25 AM10/2/08
to
>   Though for the latter, you'd need to prioritize failure conditions,
> else you might end up with a string of them displayed.  Like if the
> character were blind while it was dark, you'd just display "blind".
> (Heck, the character wouldn't even know it was actually dark until
> their blindness was cured anyway.)  Not sure how to rank others
> though, like blind versus confused.

Well, usually you CAN take any normal type of action while confused,
you just don't always get the results you want. I would prioritize
listing effects that prevent you from taking the action at all before
listing ones that just have a negative effect on the outcome. (You
might need to make a separate hierarchy for each action, depending on
what status effects you have.)

Radomir 'The Sheep' Dopieralski

unread,
Oct 2, 2008, 5:38:26 AM10/2/08
to
At Wed, 1 Oct 2008 22:50:25 -0700 (PDT),
Gelatinous Mutant Coconut wrote:

I'm not sure that adding even more code and indirection really helps
here :)

Jeff Lait

unread,
Oct 2, 2008, 9:32:02 AM10/2/08
to
On Sep 29, 11:54 am, Gelatinous Mutant Coconut

Actually, POWDER side steps the whole problem by going noun-verb
rather than verb-noun. This is the other axis of this debate which is
largely ignored.

In POWDER, you first select the item. This then gives a context menu
of possible actions on that item - quaff, dip, throw, drop, etc. PC
users can directly press the shortcut key with the item highlighted
without bringing up the menu.

Personally, the only way I want to equip armour is by dragging it onto
my paperdoll - any random set of keypresses is a second-best approach.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)

Radomir 'The Sheep' Dopieralski

unread,
Oct 2, 2008, 9:40:30 AM10/2/08
to
At Thu, 2 Oct 2008 06:32:02 -0700 (PDT),
Jeff Lait wrote:

> Personally, the only way I want to equip armour is by dragging it onto
> my paperdoll - any random set of keypresses is a second-best approach.

That's artifical too! You should have a controller in form of two
sleeves with sensors and stuff, and just put your hands into the
sleeves... :D

Soyweiser

unread,
Oct 2, 2008, 10:12:36 AM10/2/08
to
On Oct 2, 3:40 pm, Radomir 'The Sheep' Dopieralski <n...@sheep.art.pl>
wrote:

> That's artifical too! You should have a controller in form of two
> sleeves with sensors and stuff, and just put your hands into the
> sleeves... :D
>

Play WiiRL, the newest sensation on you Wii.
- Realistic play
- Slightly less boring than putting on your clothes
- Now with shock controller, you feel the hurt*
- You will finally know how a Tengu really looks
Note: demigodhood not included

*:Nintendo Inc is not responsible for any damage done to rings.

--
Soyweiser

Billy Bissette

unread,
Oct 2, 2008, 7:14:21 PM10/2/08
to
Radomir 'The Sheep' Dopieralski <ne...@sheep.art.pl> wrote in
news:slrnge95k...@atos.wmid.amu.edu.pl:

> At Wed, 1 Oct 2008 22:50:25 -0700 (PDT),
> Gelatinous Mutant Coconut wrote:
>
>>>   Though for the latter, you'd need to prioritize failure conditions,
>>> else you might end up with a string of them displayed.  Like if the
>>> character were blind while it was dark, you'd just display "blind".
>>> (Heck, the character wouldn't even know it was actually dark until
>>> their blindness was cured anyway.)  Not sure how to rank others
>>> though, like blind versus confused.
>>
>> Well, usually you CAN take any normal type of action while confused,
>> you just don't always get the results you want. I would prioritize
>> listing effects that prevent you from taking the action at all before
>> listing ones that just have a negative effect on the outcome. (You
>> might need to make a separate hierarchy for each action, depending on
>> what status effects you have.)
>
> I'm not sure that adding even more code and indirection really helps
> here :)

Prioritized conditions is something you really should deal with
regardless of your command structure. Unless your system is simple
enough that it isn't an issue, which means it wouldn't be an issue.

If being confused, blind, or in the dark prevent you from reading a
magic scroll, then the game should report those failure conditions in
an appropriate order whether you hit a dedicated "read scroll" button
or a generic "use item" command.

Andrew Doull

unread,
Oct 10, 2008, 6:45:31 AM10/10/08
to
On Sep 28, 8:41 pm, Radomir 'The Sheep' Dopieralski
<n...@sheep.art.pl> wrote:
> In her paper "Object Concepts and Action", published with other
> similar papers in book "Grounding Cognition", Anna M. Borghi is
> presenting her theory, together with some experimental evidence
> about how we represent artifacts in our minds: as references to
> actions that can be performed with those objects. These actions
> are usually very simple movements -- and they are remembered as
> simulations of those movements. Assuming this theory is true, I
> can see another important function of so many separate commands
> in roguelike games: they work to better differentiate the items
> in our minds by providing different affordances tied to motions
> of our fingers on the keyboard.
>
> Without these different commands items just feel like all those
> objects you collect in a platform game -- they give you points,
> sometimes they give you an additional life or some special move
> or attack, but most of the time you don't need to look at them,
> you just collect them all. Items in roguelike games are totally
> different, even if there is some error made by the designer and
> you have a potion and a scroll that do the same thing, they are
> still different in the player's mind. This is an error, because
> you shouldn't suggest things that are not true -- you break the
> intuitive understanding of the game world this way.
>
> Understanding the understanding of roguelike games may lead not
> only to better description of the traditional mechanisms -- but
> also to using these mechanisms better and improving both design
> and interface of the games, leading to better user interface.
>
> Or I might be completely, totally wrong :)

Everyone seems to have gone with the multiple action vs. single action
approach - but no one has consider an effects based command approach.
That is, I want to light a room, and I might have several approachs:
reading a scroll of light, casting a spell of light or running around
the room - lighting all the torches on the walls. Would it make sense
to have a single command to do which of those options is currently
available to me?

I notice you've addressed this by arguing that having multiple ways of
doing this is a design error.

Regards,

Andrew

Radomir Dopieralski

unread,
Oct 10, 2008, 8:53:47 AM10/10/08
to
At Fri, 10 Oct 2008 03:45:31 -0700 (PDT), Andrew Doull wrote:

> Everyone seems to have gone with the multiple action vs. single action
> approach - but no one has consider an effects based command approach.
> That is, I want to light a room, and I might have several approachs:
> reading a scroll of light, casting a spell of light or running around
> the room - lighting all the torches on the walls. Would it make sense
> to have a single command to do which of those options is currently
> available to me?

I would be vary of making such an "autopilot" macros that would allow
you to "run around the room and light all torches". It seems it would
have a smilar effect as using a mouse pointer for specifying commands
for your character: you are no longer "in the body" of the character,
you are merely telling it what to do. This may be also a good kind of
an interface, especially if you are controlling a whole party -- such
approach would make the kind of immersion that lets you "become" your
character impossible anyways. Actually one could argue that the "use"
action is also merely a macro that automatically selects action which
fits selected item best.

It's like you had "jump over the cliff" and "kill that turtle" macros
in the Super Mario game: suddenly it wouldn't be that fun anymore!

> I notice you've addressed this by arguing that having multiple ways of
> doing this is a design error.

Not a design error by itself, perhaps: the mistake is mixing this way
of organizing resources with this particular user interface. Once you
go the "each item can have any effect" way, the traditional roguelike
interface seems to lose a lot of its appeal, to the point of becoming
a burden.

--
Radomir Dopieralski, http://sheep.art.pl

0 new messages