Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: definition of clones: Danchess an Crafty

Author: Dann Corbit

Date: 11:40:53 02/15/04

Go up one level in this thread


On February 15, 2004 at 14:35:09, Jaime Benito de Valle Ruiz wrote:

>On February 15, 2004 at 14:26:00, Dann Corbit wrote:
>
>>On February 15, 2004 at 14:21:06, Jaime Benito de Valle Ruiz wrote:
>>
>>>On February 15, 2004 at 14:12:05, Dann Corbit wrote:
>>>
>>>>On February 15, 2004 at 14:08:24, Jaime Benito de Valle Ruiz wrote:
>>>>
>>>>>Does that mean that we can be sued for using ideas such as using bitboards... or
>>>>>a 64 bytes array to represent the board?
>>>>
>>>>You can be sued if you are innocent.  Software is an extrememly complex matter
>>>>and I doubt if 1% of judges or juries would be able to make a competent
>>>>decision.  Therefore the quality of the legal team will probably determine the
>>>>outcome.
>>>>
>>>>>The only two pieces of code that I have ever copied are:
>>>>>
>>>>>  1) A few lines to read the input, because I didn't know how to do this.
>>>>>  2) A branchless assembly version of a 64-bit BSR/BSF for 32-bit processors.
>>>>>
>>>>>The implementation (but not the original ideas) of typical chess engine search /
>>>>>evaluation functions are mine... because I don't like reading other people's
>>>>>code. Do you advice to start sending lots of mails to lots of people asking for
>>>>>the permission to use their ideas if I ever decide to send my program to a
>>>>>tournament (or make it commercial)?
>>>>
>>>>Yes.  It is not a legal necessity, but I think it is a good idea.  Probably, we
>>>>don't write more than one new routine a day.  That's not a lot of emails.
>>>>
>>>>>All this sounds a bit extreme to me.
>>>>>
>>>>>Regards,
>>>>>
>>>>> Jaime
>>>>
>>>>Like any idea, it has both a positive and negative aspect.
>>>
>>>The only problem is: How do I know who should I write for every thing that I do?
>>
>>Write to the person where you got the idea.  If it is from a book or a paper,
>>then to the book or paper author.  If is is from code, then to the one who wrote
>>the code.  If you make it up yourself, you can still be in violation (especially
>>for a patent).  Commercial software systems will do a patent search to see if
>>they violate any patent ideas.
>>
>>For instance, certain types of bitmap indexes cannot be used -- even if you make
>>the idea up by yourself because there are several patents on bitmapped indexes.
>>For that reason, I never use bitmapped indexes.
>>
>>>I know who should I ask for permission if I ever implement the tablebases (and I
>>>will ask), but I'm not sure about the rest of the stuff.
>>>Thanks for the advice in any case.
>>
>>I think that a minimal courtesy that should always be extended is to at least
>>cite the sources from which the information came.
>
>Who owns the patent for bitboards? And what bitmapped indexes are patented?

I am a database guy, who writes SQL engines.  The bitmapped indexes I am talking
about are [for instance] these:
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=36&f=G&l=50&d=PTXT&p=1&p=1&S1=(((bitmapped+OR+bitmap)+AND+index)+AND+database)&OS=+(bitmapped+OR+bitmap)+AND+index+AND+database&RS=(((bitmapped+OR+bitmap)+AND+index)+AND+database)


>I think that using bits to represent squares is not exactly an original idea. I
>started programming in assembler when I was 14 y.o., and to me it seems an
>obvious choice to use 64 bytes registers for a 64 squares board. Rotated
>bitboards is another story, of course, but I don't use them anyway.
>
>Cite the sources and thank all the people who have pubished and helped with
>ideas goes without saying, but I doubt I can remember where did I get every
>single of these ideas in the first place. Is there any page with this
>information anywhere?

No single place to find it, I am afraid.

But it is not too late to start the practice now.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.