My old Moyo Go blog!

About the development of Moyo Go Studio, software to (help) play the Oriental game of Go. Go is a two-player zero-sum game of perfect information. It is considered much harder than Chess. Currently, in spite of enormous effort expended, no computer program plays it above the level of a beginner.

Sunday, February 11, 2007

Open-Sourcing Moyo Go Studio

A win-win situation for both me and my customers? (not to mention the last nail in the coffin of my enemies :-)

Quite a number of Delphi-programmers are customers of mine, and quite a few Delphi programmers make popular Go software.

Some of Moyo Go's file import converters are already made by a customer of mine, and I see no reason why we can't accellerate development and bugfixing by making most of the source available. Some of the Gnu Go interface code is inspired by Delphi code of a Go-programming collegue of mine, and I know of at least one other Delphi buff involved in Go programming. With the freeware version yielding thousands of users now, there must be at least a dozen Delphi coders out there, interested in working on a large, exciting, complex and cool Go software project. Moyo Go offers everything from assembly language to pattern recognition to hashing to SQL databases to PDF generation to things like Go influence functions and GTP comms. It is an ideal project for anyone interested in becoming a professional programmer or those that are in need of a serious hobby :-)

Moyo Go uses a great deal of commercial libraries, but they can be legally redistributed as dll's, since there is generally no need to modify them. And my most precious IP - the sanctum sanctorum - the pattern system, is divided into two parts - pattern harvesting and pattern recognition, and the harvesting part is the really important one. I can safely omit that from the distribution.

The core of the matching part is written in assembly, but regardless of that, I don't fear competitors too much. They'll have to work for a year just to catch up with me, and in that time several people will have worked on the OS-ed code. Often, having the code (especially when it's in a programming language you don't use yourself) is of little use, for a competitor.

So I think I'll OS Moyo Go. The only thing is - I will only do it when there are several, reasonably capable Delphi programmers available who seriously want to dedicate time to maintain, bugfix and develop the software. Moyo Go offers a wealth of learning experience in the field of shrinkwrapped software development (of a large project) for Windows and Delphi programmers can grow to become Delphi guru's, by working on the code.

There are many things to be done. One rather daunting task is to convert the code to Free Pascal, so that Moyo Go will run natively on Macs and under Linux. Another is to develop independent modules for everything from file import convertors to TsumeGo. We can talk about my ideas, in that case, but you'll be at liberty to do whatever you deem interesting/useful when working on your own modules.

I still want to be able to sell Moyo Go, so you have to understand that the payment for your efforts will be mainly glory, experience (I'll offer suggestions to your work) and a better Moyo Go Studio, also for the benefit of the Go community at large (there will of course be updates to the freeware version, with credits to your work). It is not bad to have this kind of work on your CV either.

So, what do you think of this idea? Let me know if you're interested!
Email me on frank@moyogo.com

Labels: