[Trans] Directory instead of .well-known for URL structure

[Jacob Hoffman-Andrews <[e-mail]>]

The latest draft adopts a /.well-known/ path for CT as a way to get 
around BCP 190 (URI Design and Ownership: 
https://tools.ietf.org/html/bcp190#section-3).

Personally I think BCP 190 makes it needlessly painful to specify 
HTTP-based APIs using techniques that are very common among 
practitioners. However, given that it is still considered best practice 
for IETF documents, I propose that CT should use a different workaround, 
one used very successfully by ACME: Directory URLs.

In ACME, a CA defines one URL, a Directory, which provides a JSON 
dictionary showing where to find the other URLs. For instance:

$ curl https://acme-v01.api.letsencrypt.org/directory
{
   "fUKGu3pAntw": 
"https://community.letsencrypt.org/t/adding-random-entries-to-the-directory/33417",
   "key-change": "https://acme-v01.api.letsencrypt.org/acme/key-change",
   "meta": {
     "caaIdentities": [
       "letsencrypt.org"
     ],
     "terms-of-service": 
"https://letsencrypt.org/documents/LE-SA-v1.2-November-15-2017.pdf",
     "website": "https://letsencrypt.org"
   },
   "new-authz": "https://acme-v01.api.letsencrypt.org/acme/new-authz",
   "new-cert": "https://acme-v01.api.letsencrypt.org/acme/new-cert",
   "new-reg": "https://acme-v01.api.letsencrypt.org/acme/new-reg",
   "revoke-cert": "https://acme-v01.api.letsencrypt.org/acme/revoke-cert"
}

To use an ACME CA, rather than configuring a base URL, an ACME client 
configures a directory URL. All other URLs can be found by looking up 
fields in the directory.

Adding a directory to rfc6962-bis provides the most minimally invasive 
way to satisfy BCP 190. Well-known URLs aren't really meant for this 
type of application. To quote RFC 5785 
(https://tools.ietf.org/html/rfc5785#section-1.1):

    1.1.  Appropriate Use of Well-Known URIs

        There are a number of possible ways that applications could use
    Well-
        known URIs.  However, in keeping with the Architecture of the World-
        Wide Web [W3C.REC-webarch-20041215], *well-known URIs are not
    intended**
    **   for general information retrieval or establishment of large URI**
    **   namespaces* on the Web.  Rather, they are designed to facilitate
        discovery of information on a site when it isn't practical to use
        other mechanisms; for example, when discovering policy that needs to
        be evaluated before a resource is accessed, or when using multiple
        round-trips is judged detrimental to performance.

        As such, the well-known URI space was created with the expectation
        that it will be used to make site-wide policy information and other
        metadata available directly (if sufficiently concise), or provide
        references to other URIs that provide such metadata.

(emphasis mine)

Beyond that, requiring log operators to organize their log under 
/.well-known/ would pose unnecessary difficulties. Service providers 
intentionally restrict the /.well-known/ URL space. Depending on 
deployment model, log operators might have to bypass such restrictions 
or get special exemptions.

Thanks,
Jacob