Re: [urn] Comments on draft-ietf-urnbis-rfc3187bis-isbn-urn-01

[SM <[e-mail]>]

Hi Alfred,
At 15:00 19-10-2011, Alfred wrote:

>Yes, this (and a couple of other) "MAY" are not appropriate usage of
>RFC 2114 language.  A lot of these indeed should be downcased; my
>apologies for the editorial oversights.

Don't worry about the editorial oversight. :-)

>According to my understanding, the last sentence above should better
>say:
>
>    "In such cases, it will likely be possible to find the digital copy
>     with the ISBN of the printed original."

I suggest:

   In such cases, it is likely that digital copy can be found by using the
   ISBN of the printed book.

> > In Section 4.1:
> >
> >    "Each product form (e.g. hardcover, paperback, PDF) MUST have its
> >     own ISBN."
> >
> > Why is this a MUST?
>
>That's a requirement from ISO 2108.
>(Section 4.1 and its subsections make essential content of the
>ISO standard readily available to the reader of this memo.)

Larry Masinter suggested alternative text.  He also commented about 
the usage of "MUST".  Drop me a note and I'll expand on why the 
question was asked if it is not clear.

> >    "Each product form MUST have a separate ISBN, but digital
> >     manifestation will not be long-lived."
> >
> > Why is this a MUST?
>
>See above -- as per ISO 2108.

Quoting text from Larry:

   "ISO 2108 requires that each product form (hardback, paperback, PDF) have
    its own ISBN."

The above avoids the RFC 2119 keyword while still stating the ISO 
2108 requirement.

>Suggested replacement:
>
>      "Some users may prefer a modern manifestation although it might not
>       have the original look and feel, while other users want the
>       original manifestation that is authentic but might require digital
>       archaeology for access."
>
>Is that more reasonable?

Yes.

Coming back to "MAY", it's an optional feature in an 
implementation.  The text explains a user's preference and it does 
not have anything to do with an implementation.

>Suggested replacement text:
>
>      "...  Usage if "URN:ISBN will support information architectures
>       that enable persistent access to the relevant intellectual content
>       (work), independent of its form, although ..."

That sounds better.

>Due to the persistency of assigned ISNB-10 and derived UNRs, most
>text regarding ISBN-10 cannot be considered strictly "historical"
>or "background" -- it is still of practical relevance for users
>and resolution services.  Therefore IMHO separation of specifics
>from RFC 3187 into a separate section does not make much sense;
>the IANA registration of the namespace still needs to cover both
>versions of ISBNs.

I don't feel strongly about this.  I suggest getting more people to 
review the draft to gauge clarity.

>Similarly, placing examples into an appendix will likely not be
>of much help for readers not very acquainted with bibliographic
>identifiers, and would perhaps violate the requirements from
>RFC 3406 and its planned successor, the rfc3406bis draft.

Ok.

Regards,
-sm