[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [xri] versioning of authirity section
Lindelsee, Mike wrote:
>Even though XRIs don't always need to resolve, I tend to think of the modal case as being URIs that do resolve. If versioning can be applied to name authorities, I definitely agree that we need to pay close attention to the details of what that means and how resolution is specified/handled.
>
>
Agreed.
>Would you go so far as to version parts of the name authority? For example:
>
> host[;2002-04-01T20:10:23Z].foo.biz/this/resource
>
>My sense is that using versioning on parts of a DNS name might complicate things more that it would be worth. Such a feature might be easier to implement with name authorties that don't use DNS names (since we are creating a new resolution mechanism for those anyway). Thoughts?
>
>
This is not as unreasonable as it may seem at the surface. The DNS has
this capacity today with serial numbers on zones (commonly expressed as
[\d]{4,4}[\d]{2,2}[\d]{2,2}[\d]{2,2} or yyyymmddnn). So, for larger,
distributed zones, you could even have:
host[:2002030402].foo.bar[:2002020145].biz/this/resource
trouble is, in general, that only the zone authority can express such an
XRI. Others _could_, but only through digging (pun intended) deeply
with DNS administrative tools (getting zone serial numbers). I do not
think we should deliberately exclude such a structure from XRI.
As for non-DNS names, as you point out, things could be simpler (esp.
for HTTP-based resolution)
--- peterd
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]