openmoney specifications

John Waters 5 years ago

Perhaps this statement should be clarified.

John Waters 5 years ago

Yes, I get that. I was wondering why “openmoney” is lower case when the use of

John Waters 5 years ago

Is term “promises” intended to mean something more than “commitments”?

John Waters 5 years ago

Thanks.

John Waters 5 years ago (edited)

At this time, the openmoney-network interface allows the steward to log in only

John Waters 5 years ago

My point was that (in contrast to this specification) stewards currently log in

John Waters 5 years ago

Again, this is currently done using username and password.

Michael Linton 5 years ago

right

John Waters 5 years ago

I assume this is what’s done at present since upon registration of a new steward

John Waters 5 years ago

DL confirmed/explained this in his email earlier today.

John Waters 5 years ago

Why is a namespace globally unique?

A namespace can be identified uniquely by the namespace in which it is enclosed/embedded (and in which that in turn exists).

e.g.

paris.france.eu

is very different from

paris.texas.usa

Otherwise the variety management problem becomes too great - and my understanding was that the recursively-nested namespaces were intended (at least partly) to mitigate that.

Otherwsie, how can we avoid the Googlish/eBayish name collision problem where self-registering latecomers end having to try hundreds of options (all of which already exist) before caving in and accepting an horrendous autogenerated suggestion such as “fred3896”.

Maybe the initial namespace assigned to a new steward should be generated automatically from some additional details provided - such as town, county/state/whatever, country, continent, postcode/zip/whatever, etc.

In this case, new steward “fred” would have a much better chance of avoiding collision with other Freds, and would have far less difficulty in finding an acceptable username if such a clash were to occur.

(If GDPR is an issue - and anywhere in Europe it will be, as will something similar be in most places outside the US - then responsibility for maintaining the user’s details can be delegate to a third party well set up for that purpose. MailerLite’s API is very easy to use and seems to provide everything that might be wanted to offload such a burden.)

Alternatively, a completely random but unique initial namespace could be generated as a starting point for the new steward - after which it could create its own within existing namespace paths while maintaining responsibility for and control of whatever is enclosed within its own namespace(s).

Michael Linton 5 years ago

paris.tex.us and paris.texas.us etc etc are as unique as .us is

John Waters 5 years ago

That’s what I had been assuming, but then how can the following be true?

“Namespace stewards have the ability to move child namespaces, currencies or accounts from their namespace to another.”

If paris is a child of tx.us and paris.fr.eu already exists, the Texan child paris cannot be moved to the namespace fr. eu - in either case paris is a namespace so cannot be globally unique.

I suggest using the term Fully Qualified NameSpace

(FQNS) (cf. FQDN). Just as an FQDN is globally unique, so would an FQNS be. But subdomains can reappear anywhere, and for an unlimited number of cases. The same is true of a *nix filesystem, where files or directories of the same name can re-appear anywhere.

However, *nix filesystems offer something more. The same file or directory can appear in multiple directories, not necessarily by the same name, such that changes to one occur exactly in all others (because they are the same, not just copies).

Michael Linton 5 years ago

FQ for the time being I think

John Waters 5 years ago

John Waters 5 years ago (edited)

Does this mean that the steward of a particular namespace is notified when anoth

John Waters 5 years ago

OK. That helps.

John Waters 5 years ago

How does a steward know which other stewards and namespaces exist? How does a st

John Waters 5 years ago

If namespaces were not globally unique, that would not necessarily cause a probl

John Waters 5 years ago

OK. That’s not at all clear from this specification.

John Waters 5 years ago

How does the steward (A) of the child (namespace, currency or account) request p

John Waters 5 years ago

OK, but I’m sure something more sophisticated can be worked out down the line.

John Waters 5 years ago

What can be done within a private namespace?

John Waters 5 years ago

Should that read something more like “All stewards of a currency get notified wh

John Waters 5 years ago

That’s reassuring. (I often get confused by others’ sentence constructions so h

John Waters 5 years ago

To disable all accounts using that currency, regardless of ownership?

John Waters 5 years ago

OK, that’s what I would have expected.

John Waters 5 years ago

Should that be something like “All stewards of a currency get a copy of all jour

John Waters 5 years ago

What can be done with a private currency? Is this simply a temporary state whil

John Waters 5 years ago

OK, that makes it much clearer. Thanks.

John Waters 5 years ago

I assume this “set of stewards” includes all stewards with the ability to make a

John Waters 5 years ago

OK. Again this may need to be reworded slightly.

John Waters 5 years ago

… I assume the “account steward” is the steward with administrative/control priv

John Waters 5 years ago

That’s what I assumed.

John Waters 5 years ago

(This brings me back to the question of why the CSV export doesn’t include all o

John Waters 5 years ago

OK. I think a slower cobble would be a lot more useful.

John Waters 5 years ago

I assume this means that three copies of the journal entry are saved, each in a

Michael Linton 5 years ago

currency steward can only see what accounts are doing in that currency - not the