You are viewing a single comment's thread from:

RE: Hive’s future as a 2nd layer blockchain network

in #hive4 years ago

My only different view point is about the custom_json. IMHO, we should not use the feature so much. Rather, we should be keeping it elsewhere or just on few nodes with special storage. A minimal hash needs to be on the chain. With applications storing more and more JSON we are increasing the chain size.

Sort:  

I don't think on-disk chain size is likely to become a serious issue in the near term, as disk storage is one of our cheapest resources. It also impacts network bandwidth too, of course, but again, I don't see these as the most critical resources. Relatively speaking, I think cpu and memory usage are the big things to worry about now.

At the same time, I do like to optimize things where possible, and I have been considering a new custom_binary operation, as an alternative for apps that want to lower the RC requirements for their users. But this might already be implementable via the current extensions feature, even. I haven't had time to discuss it yet with the other devs here.

IIRC custom_binary operations already exist, though not really used.