Re: [Tools-discuss] How do we diagnose DOI errors?

[Michael Richardson <[e-mail]>]

Julian Reschke <[e-mail]> wrote:
    >> If you use a specific URL, then it needs to be alive.

    > Yes.

    > But I would call that "robustness", not "portability".

I plug in

      <?rfc include="reference.RFC.8174" ?>

and the tool figures out where to get the data.

When I do:

<!ENTITY RFC2119 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml">

then I'm being very specific, and if the site breaks, or produces 404s for
some reason, then things break for me.
(Yes, I have a document with one of each. Too lazy to change)

    >> While many of the other abstractions can be retrieved from a variety of
    >> sources and can be easily cached.

    > Hmm, no. All the ones that are defined for xml2rfc v3 need a full URI,
    > and all of them can be cached.

My dated experience with caching all of bibxml to my laptop so that I could
compose on long airplane flights was that the ENTITY mechanism never worked
right.  That's probably a bug, but and maybe I'm cargo culting, but that's
why I've come to prefer <?rfc.  if I have to author in XML.
{which I strongly prefer not to now}

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        |    IoT architect   [
]     [e-mail]  http://www.sandelman.ca/        |   ruby on rails    [