Quote from: brg444 on August 31, 2015, 01:29:33 AM

Bitcoin development is quite unlike any other open source software.

You are not the first one to bring up the Linux parallel. Unfortunately it doesn't apply.

There is some value to the comparison IMO.

Quote from: uxgpf on August 31, 2015, 01:18:18 AM

Quote from: Carlton Banks on August 31, 2015, 12:47:27 AM

Software Development teams can only be a small number of individuals, with a preferably smaller number of members with commit access to the project. The fact that we have a really amiable, pragmatic guy in that role right now has been consistently good for Bitcoin; Wladimir van der Laan is exactly the type of leader I like.

I think Linux has managed quite well with huge number of contributors (thousands?). Linus has been it's benevolent dictator from the day one (although he has been delegating his work to trusted "leutenants" more and more). It seems that large software projects need to have clear hierarchy in order to move things forward (someone has to have the final say on what patches get accepted).

Obiviously some people will sometimes disagree in direction taken and it will result in a fork. But the impression I get from Core is that it has no direction (no one has the final say in what should be done), just bunch of devs with different ideas on what should be done.

Somewhat agree. The hierarchy is definitely important, there cannot be deadlocks to progress, even if that means progress goes in the "wrong" direction. It also demonstrates how a less patient character can also be effective, although whether Torvald's demeanour is what actually motivates people to do the job well is debatable.

The Bitcoin dev team does have a hierarchy, but it's not enforced by a strong character. The strong characters are amongst the immediate lieutenants, not the people with commit access (who I believe are currently Wladimir, Pieter Wuille and Jeff Garzik). When contemplating the alternatives, there is no-one that I think has the right temperament for a role that requires more trust than Torvalds'. We'd possibly already have a solution, but possibly a host of new problems too.