Re: [Uri-review] Review Provisional registration for app URI scheme

[Carsten Bormann <[e-mail]>]

On Jan 17, 2018, at 17:46, Stian Soiland-Reyes <[e-mail]> wrote:
> 
> On 17 January 2018 at 16:38, Carsten Bormann <[e-mail]> wrote:
> 
>> … which raises the question what “app” stands for.
>> The draft talks about archives and packages, which explains “a” and the first “p”, but I don’t understand the second “p” (pointer?).
>> Yes, I read about the desire to stay compatible with https://www.w3.org/TR/2015/NOTE-app-uri-20150723/, but it probably would help to have a more accurate retronym if that is necessary.
> 
> +1 to make a good bacronym :) "Archive and Packaging Protocol”?

Maybe better: "Archive or Package internal Pointer"?

> But.. I deliberately left out a protocol unless
> https://rawgit.com/stain/I-D/master/app/app.html#rfc.section.3.1
> counts?

Right, protocol is probably misleading.

>> (I’m interested in this as it is sometimes useful to have URIs for references from a CBOR data item to other data items within a bigger, compound data item.  We could define this based on this scheme or in a different way, possibly based on the way the compound data item is formed.)
> 
> Interested in your requirements here, specially if you need more
> flexibility in the authority or path field. I guess this will fit into
> the comments on the Internet-Draft -- but feel free to do Pull
> Requests before that ! :)

I haven’t put much thought into the general problem yet, but this sounds like a neat piece of research to engage in.
(But probably not on this list.  We have T2TRG for that...)

Grüße, Carsten