Many people have ideas but no idea how to code. I propose a bounty be posted for clean templates of the existing code bases in the community for easy use, along with detailed guides on what can be modified in graduating order from a total noob to semi-competent developers. Thus far NRS,MMC,PTS are great but anyone who wishes to use them will suffer to get things started.
Additional Bounties could be set up for introducing new useable code bases for different applications.
This will help kick start DAC deployment outside of the usual developer circles.
This is exactly what we need. We need templates so that the development process can be decentralized and we can avoid the centralization of knowledge problem that Bitcoin development has.
I have several good to great ideas for DACs. The main problem is there is no template for how to make a DAC. I don't blame Invictus for this because making a DAC is not easy.
But the sooner we have templates to make it easier the sooner we can really push the limits.
What I want to see is a code base which allows for Proof of Stake with voting, to have data storage in the blockchain, to have some sort of M of N transaction capability so that a group of people can hold angelfunds and then vote on whether or not it gets spent by signing off on it or not. Maybe the top 20 shareholders all get a vote via M of N and if 60% are in favor then it's given the green light. Maybe some other advanced scripting should be put into it too specifically to create a sort of decentralized corporation which can allow shareholders to collectively make decisions by PoS voting.
As far as the development kit and code, just make sure that it's high level enough that anyone can deal with it. C++ is not the language which anyone can deal with even if I can deal with it. It should be high level enough that any web app developer can make use of it, so Ruby, Python, C#, Javascript or even HTML. These DACs (or at least the autonomous bounty distribution DAC I'm thinking of) is something which would work well if it could be designed as a web application which people can install on their phones, in their browsers, or download as a binary.
So go with cross platform capability if possible. We need that more than anything else and we need high level, with modularity and security.
Great, i haven't looked into the BTS code properly yet. However if they take to this idea, maybe they should break it up into individual components. eg A set of BTSX bounties and a set of non BTSX bounties. One could be setting up a new github account with clean template of PTS, with a wiki that describes the process of forking and creating an alternate. This will tie in with the genesis initalization guide i wrote. Fot BTS maybe it could be a repository in the same account doing the same, even keyhotee will have a clean template. Compilation of these clean templates, the tools required and guides would be a huge step for the ecosystem. NRS code as well can be added.
Then as we progress, anyone has access and can request pulls that improve the code as a template. So much possibilities.
I've looked at some of the BTS code. It's very clear whats going on, well organized, commented. I think Bytemaster specifically wrote it in a way so that other developers can easily read and understand whats going on so
there.
But it's still in C++ and C++ is intimidating for most developers. Remember this is cryptography, it's not the same as writing a word processor. For that reason I think we need to interface the low level code with high level languages like Python. Fellow traveler did this to a certain extent with Open Transactions. We need something like BTSMadeEasy.
https://bitcointalk.org/index.php?topic=144879.0https://github.com/FellowTraveler/Open-Transactions-old/blob/master/src/otapi/OTMadeEasy.cppanother interesting prospect of standardizing the code is a funded drive toward merge mining. I say this because the fall in hash behind PTS saw a rise in NRS, hat is not the most desirable situation as i prefer both chains to bw working, not one at the expense of the other.
If it is a possibility, it would solve the problem of hash power being divided between chains. This also puts the possibility of modular codebase in reach.
Merge mining might not be such a good idea. I'm not against mining but PTS is probably mined by Botnets. We need new algorithms which cannot be so easily gamed if we are going to use mining at all.
PTS mining is probably centralized now. You cannot mine it on your CPU anymore and for that reason merge mining isn't as attractive as coming up with a better algorithm. Mining as it currently works ultimately only leads to centralized cartels similar to how we have with oil and diamonds.