I guess many are not so keen to join my project because they think opensource makes it very hard to protect their ideas. I'll offer an explanation of why it's feasible:
1) Firstly, the source code will be under a special license that *prohibits * copying and modifying the source *except* under the official project. This eliminates a lot of potential copycats who simply take the source, improve on it slightly, and then call it their own. 2) The second type of competitors may steal our *ideas* and then create their own products. 2a) Remember that AGI is a large project with many modules working together. If we make it easy for people to simply join the project (remember that we reward our contributors financially), why would they take the troubles to start new ones? Pause here and think about how likely that'd happen... (my impression is that most PC operating systems nowadays are descended from a few sources such as Unix and Linux, which is testimony to how impractical it is to start a new OS from scratch. The same may apply to AGI projects, perhaps even more so, if you think about the scarcity of people genuinely interested in AGI and are willing/able to work on it.) Also note that AGI ideas / algorithms are unlikely to be re-usable by narrow-AI projects. So if somebody wants to steal our ideas, they're doing that to start their AGI projects, and obviously they have some improvements in their mind, so why don't they just join us and improve on our project? That saves the trouble of duplicating other stuff. 2b) Another thing we can do is to patent our designs and algorithms (so there will be a grace period of 1 year between disclosing something on our forums and patenting it). A potential problem is that during the early stages, our ideas will be rapidly evolving so it will not be easy to decide what to patent or not. So we risk not protecting some early ideas... but also remember that *ideas* cannot be patented anyway. And even in closed companies, leaking of secrets is easy due to the flow of people. I mean, *normally*, opensource + commercial is an *unsound* business model, but in the special case of AGI it may be sound, due to the large-scale and intensive research required. All in all, we aim to be the *best* AGI project and our reward system will be very generous and fair. That'll make it quite unnecessary and unlikely for people to start competing projects, *given* the great difficulties of starting them. Have I stated my case clear enough? Are more people convinced? Or have counter-arguments? YKY ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=e9e40a7e
