If memory serves, I first came across PUSD in the griefing thread a while back. I then borrowed it gratuitously for various other threads, so I guess I''m partly to blame

The way I envisage a "closed" information system (rather than the usual "open" system where there''s no check on where information came from) working is by having an in game "information object" (info for short) class. When a player learns something in game, he acquires an info. This sits in an inventory screen somewhere, and can be "given" to other people. Which actually means giving that person a copy. The hard part is figuring out an implementation that captures all relevant information - particularly if you extend the info class so it includes things like character skills, spells, etc., not just passwords, maps, etc.

If you do extend info, then you probably want to include possibilities of imperfect copying between characters, and of recording info in "physical" form - scrolls/books etc - this would make it easier to enforce a limit on "headspace" info inventory capacity. Naturally, info in physical form can be given away without being copied, or can be stolen. Written info should be memorisable without chance of failure by the author, but not necessarily for others.
Optionally, written info could be of two types - personal (no penalties for author, severe learning difficulty for others) or general (flat moderate penalty for everyone and take longer to "write")

As to implemantation, info probably wants to be some sort of game code fragment or fragment in some embedded scripting language. Obviously, info shouldn''t be created every time a player drops an item (or should it?) otherwise both character and computer memory will fill up rapidly... On the other hand, if info on abandoned items doesn''t take up much inventory, it could be workable, and allows the player to "forget" where the character left a given item. Of course, taken to extremes, info becomes a direct copy of the game data files - which may actually be the best way of implementing it - give each node of your datafile an associated info size and just keep track of which sections to point at. The player will soon wise up and start deleting "free" info - info that is automatically learned 100% accurately just by entering the appropriate area of the gameworld. Possibly the player should have the option to have most "free" information lost automatically (rather than memorising the wall textures, lights, etc, the character just keeps a map of room sizes/locations). This sort of implementation also allows the option of recording "home videos" as in game info objects. Maybe worth allowing a "recreate" option to let the player wander round a reconstruction of an area from info...

Hiding something just means that the info on its location is hard to learn automatically (just by being in LOS) and probably also that the info size is larger than if it weren''t hidden.

Anyway, I think interested people can come up with any other details of the implementation for themselves.

quote:

---

Original post by rmsgrey
If you do extend info, then you probably want to include possibilities of imperfect copying between characters, and of recording info in "physical" form - scrolls/books etc - this would make it easier to enforce a limit on "headspace" info inventory capacity. Naturally, info in physical form can be given away without being copied, or can be stolen. Written info should be memorisable without chance of failure by the author, but not necessarily for others.
Optionally, written info could be of two types - personal (no penalties for author, severe learning difficulty for others) or general (flat moderate penalty for everyone and take longer to "write")

---

Your description of the info system idea is dead on. I hadn''t thought about information degrading or writing and recorded copies of things, this is pretty interesting. The thing is, do you count all significantly degraded info as false, like a lie? As to the information of overflow, you can ignore a lot of actions (dropping items) and only stick to major events, like killing important monsters and characters or groups of monsters and characters, getting important items, completing major quests, etc. Other stuff doesn''t matter as much, even though you could have temporary info that dissappears after a day or so (representing short term memory.)

quote:

---

Original post by krez
[i am worried about a situation where a player sees another player bury something in game, and then tells a third player where it was buried OOC... now even if the first player asks the second player, and extracts truthful answers, he will find that that character saw him bury it there, but that he didn''t dig it up or tell anyone else where it was... so now his treasure is lost with no way of tracing it ever. then again, i imagine this wouldn''t happen often anyway.

---

This is a problem when you have perfect mind reading... If worse comes you worse you could also treat OOC chat logs as info also. Even though they wouldn''t mean anything to NPC characters because they don''t describe events in a way they can comprehend, it would help players in situations like the one you described.

quote:

---

Original post by Impossible
Your description of the info system idea is dead on. I hadn''t thought about information degrading or writing and recorded copies of things, this is pretty interesting. The thing is, do you count all significantly degraded info as false, like a lie? As to the information of overflow, you can ignore a lot of actions (dropping items) and only stick to major events, like killing important monsters and characters or groups of monsters and characters, getting important items, completing major quests, etc. Other stuff doesn''t matter as much, even though you could have temporary info that dissappears after a day or so (representing short term memory.)

---

If you get an important item, you probably don''t need a separate info for your character to know he''s got it - at worst, a check through the physical inventory screens should find it... and quest completions etc. are probably of more interest to the game engine than the player character. If you really want to preserve that sort of info automatically, then have the character keep some sort of personal diary, in which he records significant events.

As to degraded info, depending on how degraded it is, it probably provides a bonus to attempting to recover the true info - of course, you have to find out that the info is inaccurate to start with! So for buried treasure, there''s normally a possibility of finding it anyway - just by digging a lot of hole - which in game terms translates to a fixed time taken to search in order to learn the info of its location (country->county->local town->precise region->screen->onscreen(x,y)). A degraded version would have an error - probably at the (x,y) level, more degraded at the screen level, more degraded at the precise region level, etc until total degradation gives a random country (server). Precise details will tend to conform to the game''s data structures. Errors at the (x,y) level will produce a lengthy but workable search (several days at least). Info at the precise region level or higher should probably not degrade easily. Someone without the in-game info shouldn''t find the treasure without a very lengthy search, even if he knows the location OOC.

Partial info is also possible - only knowing the location of the treasure to the precise region level, or knowing (x,y) without knowing which screen to look on (real life equivalent would be a map showing an arrangement of trees with an X and no hint as to where the trees are). As with degraded info, partial info should reduce learning times for the correct info - in fact, partial info should provide a better reduction because you don''t need to spend time finding out where the degradation is...

quote:

---

Original post by rmsgrey
If you get an important item, you probably don''t need a separate info for your character to know he''s got it - at worst, a check through the physical inventory screens should find it... and quest completions etc. are probably of more interest to the game engine than the player character. If you really want to preserve that sort of info automatically, then have the character keep some sort of personal diary, in which he records significant events.

---

The quest stuff should be an information item because it''s not only important to the player, but important to NPCs. Really, anything that can be used to gain favor with or manipulate NPCs (and PCs) should be an information object. If you completed some difficult or important quest you''d have a record of it (and perhaps some of the details) and could pass it on to get further benefits, or lie and say you completed it to make yourself look like a hero.

quote:

---

Original post by Impossible
The quest stuff should be an information item because it''s not only important to the player, but important to NPCs. Really, anything that can be used to gain favor with or manipulate NPCs (and PCs) should be an information object. If you completed some difficult or important quest you''d have a record of it (and perhaps some of the details) and could pass it on to get further benefits, or lie and say you completed it to make yourself look like a hero.

---

Most difficult/important quests produce physical items, which act as proof of completion. That''s why the cliche isn''t "Go kill the monster and I''ll reward you" but "Bring me the monster''s head and I''ll reward you"

Bragging rights generally require some sort of independent evidence. Generally a quest will either be to retrieve some item, or to remove some problem. In the former case, the item is evidence enough. In the latter, the cessation of the problem would normally be good enough. In some cases, the proof is a specific info (the secret of the inner chamber of the sacred cave...), in which case, that info is a game object anyway. Players could keep infos like the layout of the final room as souvenirs, but they generally aren''t needed as proof for major quests. Minor stuff, there''s not much point bragging about anyway...

quote:

---

Original post by rmsgrey
Bragging rights generally require some sort of independent evidence.

---

This depends on your game world, in a lot of games there wouldn't be any physical evidence linking you to the quest, also there's no reason you couldn't fabricate the physical evidence (depending on what it was), and come say you completed the quest. Once again, it all depends on how you design your game and if you want to allow for that sort of thing. There's not much of a point to the whole info system if you can only use it for minor things that don't make a difference in the game world, it would be a waste of work.

I wouldn't use the system for a traditional fantasy RPG anyway, but it could be used. I guess we're both thinking about different uses for the same system. I'm thinking it could be used as a gameplay device between both PCs and NPCs. You seem to see it as more of a secure way to tell player's certain things in a multiplayer game.


[edited by - impossible on December 25, 2002 1:07:31 AM]

Of course bolting the info system onto an existing game isn''t going to do much - existing games are designed to work without it, so wouldn''t take advantage of it were it there. If you design the game with the info system in mind, then it''s possible to take advantage of it, and build it into the game properly. For example, finding secrets could require you to talk to the right people (getting (partial) infos) and then spend time searching rather than just look on gamefaqs or in the OSG for detailed instructions...

And the main reason I''ve been focused on player-player interactions in my posting is that, to my mind, NPC-PC communication is easily implemented with existing systems, though the info system should also work, though "realistic" communication would require better AI than anyone''s really developed (though programs like Creatures have come close). Player-player communication though was the major problem being discussed in the thread - particularly distinguishing player and character knowledge. My system is mainly aimed at that as being harder than "good enough" NPC-PC interaction.

quote:

---

Original post by rmsgrey
And the main reason I''ve been focused on player-player interactions in my ily implemented with existing systems, though the info system should also work, though "realistic" communication would require better AI than anyone''s really developed (though programs like Creatures have come close).

---

I think the info system would solve a lot of this, and is better than most (if not all of) the current systems because it''s more flexible than menu based and a lot easier to use than a parser, also it supports lying and rumors better than any other system I can think of.