Re: A Web of Ledgers [via Interledger Payments Community Group]

On 9 October 2015 at 20:21, David Nicol <davidnicol@gmail.com> wrote:

> The concept doesn't exist in a vacuum
>
>
> http://womeninbusiness.about.com/od/glossaryofaccountingterm/g/definition-ledger.htm
>
> implies, for instance "chart of accounts" and "journal"
>

That makes sense.  Originally ledgers were in books, now they are becoming
digital.


>
> it seems to me that the proper focus is the messages between the
> ledgers, in general enough way that the internal details of ledgers --
> the Dogecoin block chain on one end of the spectrum, pen and ink on
> the other -- don't matter.
>

Messages are certainly a key concept.  But so is the underlying ledger
itself, how it is defined, how it is shared etc.  Importantly the
declarative nature of the web allows a ledger to be embedded in a
document.  However as you say the same concept could apply to crypto coins,
pen and ink etc.


>
> I like DHT hashes for identifiers; DHT hashes of public keys even
> more. By using Distributed Hash Table instead of Domain Name Service
> to identify and connect to a ledger service, several administrative
> functions (and associated secutity issues) just disappear >poof<.
>

DHT are indeed good for identifiers.  But I would caution against, while
including DHT, *excluding* other styles.  Could we not make DHT ids into
URIs and allow ledgers to be system independent?


>
> Double Spending is obviously resolved by requiring per-spend
> round-trips to a ledger service or its authorized delegate; Forgery is
> obviously resolved by relying on public key signing.
>

Double spending is an important orthogonal problem.  There's going to be
lots of approaches.


>
> The devil is in the details of course; the above points are what I've
> been thinking about, and would welcome venture backing to take a year
> off and focus on, but have not been aggressively pursuing outside of
> statements such as this one.
>

I've also spent the last year working on such things as a personal points
system for my own productivity.  It works really well, im very happy with
the results.  Naturally as I trust myself, the double spend problem solves
itself :)


>
>
>
>
>
>
> On Fri, Oct 9, 2015 at 12:54 PM, Melvin Carvalho
> <melvincarvalho@gmail.com> wrote:
> >
> > I would like to see a ledger defined somewhere.  In particular the
> minimal
> > fields that can be used to create a ledger, which I think are:
> >
> > <identifier> : <amount>
> >
> > And an extensibility mechanism.
> >
>