Cryptocurrency: -- The New Social Contract
zen at freedbms.net
Thu May 28 17:16:11 PDT 2020
On Thu, May 28, 2020 at 06:32:17PM -0300, Punk-Stasi 2.0 wrote:
> so, I was wrong and now stand corrected. Apparently a system based on accounts works fine. A counterargument of sorts is that keeping the whole history makes the system more robust, but it doesn't seem like the increased amount of 'security' can be clearly gauged. The costs imposed by a bloated chain on the other hand are easier to see. Something that always bothered me with chain verification is that it only makes sense if you have the 'authentic' client. So in the end the security of the whole system depends on getting the right binary, or the right sources, etc, off the arpanet...
Yes the following is most likely stating the obvious - just making sure we're all on the same page here:
The present status quo for the current dominant player (BTCes, muffas) may have devolved to "the right binary" (I'm not personally familiar though sorry), but at least in principle it's the protocol that is 'canonical'.
And so, logically, if the protocol is 'secure' then conformant implementations must also have the same (in respect of that protocol) security characteristics.
Lack of competing implementations and lack of protocol improvements thereafter disclose other issues or problems, such as lack of development community, dysfunctional community due to greed (e.g. many irrelevant competing alternatives/variations), failure of anyone to either identify and/or improve upon protocol deficiencies, etc.
More information about the cypherpunks