I agree with you, sometimes when you hit a limit, you better go check something else, i use to find solutions while doing something else, totally different.
Even i have never coded an AI, i can imagine how diffuclt and tricky it can be, keep it up, i know you will finish it, but just a question, will this AI be carried in all the wellets ? or will it have a central brain ? will users be able to rename the AI ? or provide some modifications to it ?
The Ai is meant to be decentralized, but to work within given rules. Each node acts like a neuron, broadcasting data to others over the network. Yes, i think i will allow each Ai node to be named by the owner, however i will enforce versioning to prevent attempts at manipulation.
There are currently a couple of sticking points that i am trying to solve hence the delays in releasing an alpha version.
1) random forest decision trees (was using a crap implementation, decided to go pro) for individual nodes and better simulations
2) how to put together a decentralized neural net that can tolerate nodes coming and going ?
I am curious though, do users prefer PoW or PoS ? The code is capable of both or hybrid.
I think that hybrid is the best solution: I've seen coins have sudden issues with PoW and the chain kept going because there was PoS; and I could imagine a similar, viceversa situation.
I'll begin testing more once i complete the current set of changes i am working on.
QUICK UPDATEI want to start by apologizing to users who were looking forward to the testing version. A lot of the initial ideas have changed greatly during the course of development to the point that i had to hold back everything else and start afresh. While it's bad news in terms of time, it's really super news in terms of what Oracle will evolve to become. I'll just briefly describe some of the major changes that i am currently proposing and developing and the rationale behind them :-
Addition of transaction comments In the ideal future, Oracle will be used for payments across a wide range of goods and services, I've added an optional field to transactions that users can attach information about transactions to over the network. Unlike the standard wallet only transaction comments found in most bitcoin clones, these comments are actually relayed between peers.
Use caseSay @usernameX proposes a new service or idea and asks for donations/holds an ICO. I could send coins to his address with additional information included such as my bitcointalk username.
Or if @usernameY is selling some items, i could send them payment with my name, goods required, quantities.
Crypto Directory This is an extension of the transaction comments. Users will be able to send each other encrypted messages over the network in transactions. You can only read encrypted messages of users in your local directory.
Users are identified by name and address and or both. This is not only limited to users. A service such as node hosting or mixing service could advertise themselves over the network and leverage the directory to be able to contact their customers without needing to know any of their personal info.
Native Assets on the Blockchain I have been tinkering around with the chain while studying how it functions as well as concepts that were never fully explored vis a vis blockchain/network expansion and top level application layering. I find the concept of secure, mandatorily funded assets with real cost to be an achievement worth pursuing. I have seen quite a few coins tote the "assets" or "blochain 2.0" moniker, but fail to gain serious useage, this is due to many varying factors, but i have chosen to address the biggest (IMO) issues, which are value and utility. Rather than reliance on external services, web or otherwise, i am implementing native asset issurance, management and trading as a part of not only the network and blockchain, but as part of the reference client. The uses cases are more than i can write out here, but i'll just point out that rather than assets issued on a paltry amount of token, would you as a potential user/trader prefer real cost to them therefore ascribing a real minimum value for each asset unit you hold ?
More on this later....
Proof Of Stake ProposalI'd really like your input on this as this is something that became an issue when an Ai instance choked while trying to determine PoS , PoW and Money Supply modifiers at a checkpoint. (Block day 48 me thinks)
What do you think is a good rate of PoS? Should we use coin age ? How much time should pass before a coin can gain stake ? should we have a minimum amount that can stake ?
Should we allow users to issue assets that gain stake , should we impose limits on this ?
Is negative proof of stake something you would object to as a means of supply control ?
User Interface Modifications I am doing some major reworking of the UX and UI, mostly to accommodate the expanded scope of Oracle such as assets, internal exchange and also to ensure a truly seamless cross platform environment.
i am still rather surprised but the rejection of the Ui i previewed, I really like the see through aspect, though the color may need changing. We will user response to the Ui i will preview soon.
Sundry and pet projects/ideas Most of these you already know, i.e the custom address thing. A new one however came to the fore-front. If it gets enough support, and is truly feasible, our reference client will not only be cross platform, but also allow use of multiple blockchains and wallets from the one wallet. This is a major overhaul of the code and may take some time, but i already have a git branch with some of the basic changes already applied. The eventual idea is for a user to simply click a few checkboxes in settings, enter some variables and poof, any existing or new coin based on basic bitcoin code base will be accessible from the wallet. The initial version, if this goes ahead will host Oracle , Bitcoin and maybe LTC to showcase it's functionality.
More on this in the future
Branding What to call the base token of Oracle ?
Open Source I still have zero interest in making this open source.
Website and Mailing list I will be setting up a new website for Oracle, testnet nodes and update the mailing list soon so we need not only communicate via this forum.
Oracle Ai Kernel The kernel has seen some more update to it, however due to the massive changes in the blockchain structure, i have had to reset the blockchain's genesis block a few times this weekend, leading to the sample size of data being too small, so rather than a lone test, the next test will include some users , hopefully people who do not mind running the wallet 24/7 and mining with a single thread for the duration, so that the sample space becomes larger and we can evaluate the accuracy of classification and predictions made by the AI.
