- From: Andrew Brown <andrewbb@gmail.com>
- Date: Tue, 29 May 2018 11:00:54 +0700
- To: Melvin Carvalho <melvincarvalho@gmail.com>
- Cc: Web Payments <public-webpayments@w3.org>
- Message-Id: <FD3AE478-93B6-4C3F-B417-0D86E1927AA0@gmail.com>
It’s not necessary to keep a balance. The balance can be calculated by summing the events. Sent from my iPhone > On 28 May BE 2561, at 6:24 PM, Melvin Carvalho <melvincarvalho@gmail.com> wrote: > > > >> On 28 May 2018 at 11:01, Andrew Bransford Brown <andrewbb@gmail.com> wrote: >> That looks like a good control structure to place events into a blockchain-style ledger. > > Indeed. I think since the invent of block chains the terms block chain and ledger have become interchangeable. > > As a first step I'm looking for a ledger, which in blockchain language would be similar to the UTXO (Unspent Transaction Outputs) > >> >> Regardless of where or how you store the events, I suggest these: >> http://promiselanguage.blogspot.com/2017/03/update-to-data-structure.html > >> http://promiselanguage.blogspot.com/2016/07/contract-scripting-language-csl-example.html > > Thx for sharing this, but this is transactions, rather than a ledger. I do want to model transactions too, but my first building block is a ledger. > >> >> Those events have sufficient granularity to describe any currency or barter transaction. They support contract law, plus multilateral agreements. For example, a 1000-person equilateral contract for common property ownership or city residents is possible. >> >> (also see attached for examples of a stock trade and a barter trade) > > This could be very useful as a workflow methodology, I like it! > > A ledger is something like : > > Alice <:amount> 1.2 > Bob <:amount> 1056 > > The important question here is, what is the currency? > > There's two ways to do that, with trade offs. > > 1. Associate each balance with a currency. This has the advantage of flexibility, but the disadvantage of forcing each line to be it's own data structure. > > 2. Associate each ledger with a currency. > > So I am reaching a conclusion that 2 is the simplest, most performant, most easily query-able way to model this in RDF > > This is how I modeled it in webcredits 1.0 [1] which I've tested over many years. It strikes me as an optimal solution as input to webcredits 2.0 which I will tweak slightly with what I've learnt, and to reuse the work here. > > [1] https://webcredits.github.io/spec/#ledgers > >> >>> On Mon, May 28, 2018 at 3:20 PM, Melvin Carvalho <melvincarvalho@gmail.com> wrote: >>> I was pointed to this >>> >>> https://w3c.github.io/web-ledger/ >>> >>>> On 26 May 2018 at 12:09, Melvin Carvalho <melvincarvalho@gmail.com> wrote: >>>> Hi All >>>> >>>> I'm looking at creating a version 2 of my webcredits system and I'd like to reuse any work done by this group, as appropriate. >>>> >>>> I was wondering if a Ledger has been modeled, either conceptually, or, better still, in RDF. >>>> >>>> To my mind a Ledger, in its most basic sense is a list of balances. But there are other items that could apply. >>>> >>>> Would love any pointers to existing work, if there is some >>> >> >
Received on Tuesday, 29 May 2018 04:01:23 UTC