Loading ...
Sorry, an error occurred while loading the content.

Re: [baseball-databank] Re: Digest Number 790

Expand Messages
  • Theodore Turocy
    ... Chadwick s development is largely feedback-driven. The current state of Chadwick reflects my needs -- I needed a set of tools to manipulate data, and I
    Message 1 of 10 , Jun 13, 2006
    • 0 Attachment
      On 6/8/06, cjhaddick <cjhaddick@...> wrote:


      Professor T and All:

      These are very good points, and I can see where situational scoring,
      Diamondware/Retrosheet notation could be of an advantage and,
      consequently, where Scorepad type notation could be slower.

      Here are, in my view, the three chief drawbacks of the current
      Chadwick package, some of which the professor and I have talked
      about briefly off the board. First, the current Chadwick version
      requires manual input of not only lineups, but entire rosters,
      before scoring a game starts. You cannot write - in or add non-
      roster players (such as late callups) if you are using Chadwick in
      real time. You had better have all possible players manually
      entered up front or you will be unable to complete the game notation.

      Second, the current version of Chadwick does not support pitch
      detail. Scorepad, e.g., does, and will be coming out with an
      upgrade so that even pitch location can be input.








       

      Chadwick's development is largely feedback-driven. The current state of Chadwick reflects my needs -- I needed a set of tools to manipulate data, and I wrote them. The entry facilities are certainly not complete, for example -- but they do what I need them to do.

      It is very helpful to receive specific, detailed feature comments and requests like this. If I don't hear anything, I assume that either (a) Chadwick satisfies people's needs, or (b) Chadwick is something people aren't interested in.  In either case, it doesn't make sense for me to implement features, since I don't know if, or to whom, features will be helpful.


      Third, and here's an interesing one, let's call it the "Turocy
      Learning Curve." I bet you could sit the professor down and throw
      all kinds of game situations at him and could come up with the
      proper Diamondware/Retrosheet notation lickety split. The problem
      is that when infrequent users are confronted with infrequent
      occurences, e.g., catchers interference, dead ball runner advances,
      etc., this tends to stall the speed of the system. In fairness,
      learning is not hard, and most of it is common -sense intuitive once
      you find what you are looking for, but this is a little daunting up
      front.



       
      A long-standing weakness for the DiamondWare notation is that it's not fully documented. Over the last ten years, Retrosheet has developed a set of "coding standards," but these aren't even really written down anywhere in one place (that I've seen at least). One of the hardest things about writing the parser in Chadwick was figuring out just what *were* the legal play notations!

      The folks at XMLTeam have done some work (with some help from me) at specifying the notation as a formal grammar, which is at least a step in the right direction. But, public documentation of how to code plays has always lagged.


      To be honest, I was almost completely swayed by Scorepad, but the
      professor has really forced me to re-think, and I have to say now
      Scorepad and Chadwick are in a fair fight, at least in my mind.

      Perhaps the answer is continued work and improvements on Chadwick,
      maybe even some built in help features, glossaries, and maybe long
      term, a windows type interface. Another improvement would be some
      licencing relationship with MLB whereby Chadwick users could quickly
      download and update current MLB rosters, to allow easier work with
      players, lineups, and subs.




       
      The chance of a licensing agreement with MLB is zero, since it's philosophically absolutely at odds with the Open Source principles Chadwick is based upon. However, there is nothing stopping anyone from maintaining and distributing such files themselves, and I'd be quite happy to make them available at the Chadwick website.


      If I have a pencil, there is no doubt I use Reisner's forms and
      Retrosheet notation. If I want to go the laptop or desktop route,
      it is still a close call.


       
      One of the reasons I never finished the pitch-by-pitch input was that I was toying with adopting Reisner's situational scoring squares for input. That requires a lot more custom interface coding, of course, and I just never have had the time or impetus to finish it.

      TT


    Your message has been successfully submitted and would be delivered to recipients shortly.