MIME-Version: 1.0
Sender: bernd.jendrissek@gmail.com
In-Reply-To: <558EBCB7.9050100@trek.io>
References: <558EBCB7.9050100@trek.io>
Date: Sat, 27 Jun 2015 19:36:50 +0200
Message-ID: <CAF7PVPrQxyCE-zPZxwXfEkWMb0u2cS_nmavBM0b7n7iQdMpFtg@mail.gmail.com>
From: Bernd Jendrissek <bitcoin@bpj-code.co.za>
To: will binns <will@trek.io>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: bitcoin-dev@lists.linuxfoundation.org
Subject: Re: [bitcoin-dev] Block Size Debate Analogy / Workaround: Bitcoin
 is Like Windows 3.11
Precedence: list

> In some ways, however, to me at
> least - Bitcoin is like Windows 3.11.
[...]
> Now there is a huge debate about if there
> should ever be a Windows 95, XP, Pro, etc., that scales better and makes
> advances over time, but doesn=E2=80=99t support facets of older versions =
as it gets
> updated.

I like your analogy for how it frames blockchain compatibility in
terms of the backward compatibility that hopefully most
computer-literate people already understand, but there's a key
ingredient missing.

It's as if, if everyone in the world did somehow upgrade to Windows
95, it would become forever impossible to take a program written *on*
Windows 95 but *for* Windows 3.11, and successfully run it on a
Windows 3.11 computer. It would be as if cross-compilers from Windows
95 to Windows 3.11 didn't, and couldn't, exist. Any coins that have
post-hardfork coinbase outputs anywhere in their tree of inputs (a
Windows 3.11 program, that's written on a computer that has ever run a
Windows 95 program) can never be spent on the no-change side of the
fork.