Re: A Survey of ILP Account Identifiers?

On 12 December 2016 at 00:35, David Nicol <davidnicol@gmail.com> wrote:

> On Sun, Dec 11, 2016 at 2:54 PM, Melvin Carvalho
> <melvincarvalho@gmail.com> wrote:
>
> > Im not sure you can put a 20 byte hash in the ILP ledger, I thought it
> was
> > just URIs, or does it allow plain strings of characters, too?
> >
> > A sha1 is a random string of characters.
> >
> > How is anyone going to tell that the string is a sha1 without some out of
> > band information?
>
> the hashes as such wouldn't cross the boundary between my system and
> ILP; they wouldn't escape as hashes. The connector would have to look
> up what the hash is for, and send the string to a peer that isn't
> using the cloud backing. Alternately, I'd get "Mosler" up and going in
> time to have Mosler-encoded handles (whatever they wind up looking
> like) included as a supported type of string. I haven't actually read
> that far into the ILP spec, yet, to see what is currently declared as
> supported. To actually answer the question, with my current design, a
> string of 32 chars from the normal Crockford Encoding set (ASCII
> digits plus most capital letters) not in an URI format would be
> presumed to be a hash to look up in Mainline as a BEP-44 immutable,
> and the out of band information saying so is essentially this e-mail
> and later restatements of such :)
>

Got it.

So the advantage of http URIs are that everyone knows how to look them up.

The disadvantage with a random string of characters is that it requires
explaining to everyone that wants to use it, which turns out to be quite a
bit of work.

I suppose it all depends on your audience.


>
>
> --
> "Teaching radical novelties is our main safeguard against
> dictatorships" -- Edsger W. Dijkstra
>