March 2004

• Quick question SSL Mats Lidstrom
• Player Justice (was: Character Restraint & Capture (bounty hunting)) Damion Schubert
• Player Justice Paul Schwanz
• • Player Justice Damion Schubert
  • • Player Justice Freeman, Jeff
  • Player Justice Koster, Raph
  • • Player Justice Michael Sellers
    • • Player Justice Koster, Raph
      • • Player Justice Matt Mihaly
      • Player Justice Freeman, Jeff
    • Player Justice Paul Schwanz
• Quick question SSL...solution? ceo
• The Economic Model - as a Game Matt Chatterley
• • The Economic Model - as a Game j0hn@mistydale.com
• Parser engines Brian Hook
• • Parser engines Hans-Henrik Staerfeldt
  • Parser engines Malcolm W. Tester II
  • • Parser engines Mike Rozak
    • • Parser engines Miroslav Silovic
      • • Parser engines Mike Rozak
        • Parser engines Artur Biesiadowski
      • Parser engines Jason Slaughter
      • • Parser engines Mike Rozak
        • • Parser engines Travis Casey
      • Parser engines Nathan Yospe
  • Parser engines ext.Christer.Enfors@tietoenator.com
  • • Parser engines Sean Middleditch
  • Parser engines T. Alexander Popiel
  • Parser engines Lars Duening
  • Parser engines Brian Hook
    Thanks to everyone for their replies.
    
    It looks like my fundamental assumption -- that it just isn't
    necessary to have a very elaborate parser -- is probably the most
    accurate. I doubt that better parsers were somehow eluding the
    technical capabilities MUD server authors =)
    
    My original parser was a very simple "force the user to adapt to me"
    style coupled with "terse and effective style". Everything was
    decomposed into the basic format:
    
    <verb> [<dobj>] [[<prep>] <idobj>]
    
    It handled a lot of the sentence simplification through plain text
    substitution instead of some elaborate pattern matching. I also
    just flat out ignored all prepositions, originally thinking that the
    parser would be simple enough that differentiating between "look
    under box", "look in box", "look on box" and "look with box" would
    just map to "look box" and get handled.
    
    As for the pattern matching system some have mentioned, that's
    basically what Hugo, TADS and Inform all do. A verb is specified
    with an ordered set of grammars that in turn map to a specific
    function, e.g.
    
    Verb get
    "get" "out" --> Exit
    "get" "in" --> Enter
    "get" <multi> -- Take
    "get" <multi> "from" <obj> -- RemoveFrom
    etc. etc.
    
    Then I realized that at some point complexity is unavoidable,
    specifically with disambiguation (which sword? which key?). I'm
    really not fond of unintuitive (to non MUD players) forms like:
    
    > get 2 sword
    > get 2.sword
    
    That just strikes me as...messy. Also, handling group w/ exceptions
    was proving cumbersome when trying to hack into my simplistic
    parser, i.e.
    
    > get all swords except the rusty sword from the backpack
    
    So I'm writing a parser that lies somewhere between the full on
    complexity and robustness of an IF-level parser, and the simplicity
    of a typical MUD parser.
    
    I still require most sentences to decompose into verb, direct
    object, preposition and indirect object, but I don't use a grammar
    to dispatch, I just have a high level parser for each verb type.
    
    Thanks for all the responses!
  • • Parser engines Helps Family
    • • Parser engines Travis Casey
    • Parser engines Robert Zubek
  • Parser engines Brian Hook
• Meet & Greet IV Anthony R. Haslage
• N00b question Matthew Turland
• • N00b question Peter Harkins
• [good] N00b question Manuel Lanctot
• Player Justice Brian 'Psychochild' Green
• • Player Justice John Buehler
  • • Player Justice Damion Schubert
    • Player Justice Jester
    • • Player Justice Baar - Lord of the Seven Suns
      • • Player Justice cruise
        • Player Justice Jeff Fuller
        • • Player Justice Travis Nixon
          • Player Justice Tom "cro" Gordon
        • Player Justice John Buehler
        • Player Justice Eric Random
• Player Justice Matt Mihaly
• • Player Justice Brian 'Psychochild' Green
• [good] Parser engines Manuel Lanctot
• • [good] Parser engines Travis Casey
  • [good] Parser engines Mike Rozak
• Character Restraint and Griefing mark
• Cultural impact on Muds Ola Fosheim Grøstad
• 2d mapping in SQL Matt Chatterley
• • 2d mapping in SQL szii@sziisoft.com
  • 2d mapping in SQL Byron Ellacott
  • • 2d mapping in SQL Matt Chatterley
    • • 2d mapping in SQL Byron Ellacott
      • • 2d mapping in SQL Matt Chatterley
  • 2d mapping in SQL Ola Fosheim Grøstad
• Why significant rewards for quests are a bad thing Ola Fosheim Grøstad
• • Why significant rewards for quests are a bad thing Brian Miller
• Character Restraint & Capture. (long) Jester
• MUD/MMO Service Tools Jim Purbrick
• [Biz] Bots Open Door to Gaming History Michael Tresca
• [Announcement] Mesh Project Jon A. Lambert
• Sun's Sim Server and Gordon's 10 Reasons (the first one :)) ceo
• • Sun's Sim Server and Gordon's 10 Reasons (the firstone :)) Koster, Raph
  • Sun's Sim Server and Gordon's 10 Reasons (the firstone :)) Michael Sellers
  • • Sun's Sim Server and Gordon's 10 Reasons (the firstone :)) ceo
  • Sun's Sim Server and Gordon's 10 Reasons (the first one :)) Luca Girardo
  • Sun's Sim Server and Gordon's 10 Reasons (the firstone :)) Mike Rozak
• DGN NEWS: Master's thesis on quest systems for MMORPGs William Saar
• • DGN NEWS: Master's thesis on quest systems for MMORPGs Douglas Goodall