W3C home > Mailing lists > Public > spec-prod@w3.org > October to December 2011

Re: References Re: What are the requirements/problems? Re: Working on New Styles for W3C Specifications

From: Harald Alvestrand <hta@google.com>
Date: Tue, 13 Dec 2011 07:34:59 -0800
Message-ID: <CAOqqYVEoKVx2rgGA4AgQu1h6YOVvh4Rzgs3adK8CrtHoVopr0Q@mail.gmail.com>
To: Marcos Caceres <w3c@marcosc.com>
Cc: Julian Reschke <julian.reschke@gmx.de>, duerst@it.aoyama.ac.jp, Charles McCathieNevile <chaals@opera.com>, "chairs@w3.org" <chairs@w3.org>, "spec-prod@w3.org" <spec-prod@w3.org>
On Tue, Dec 13, 2011 at 03:39, Marcos Caceres <w3c@marcosc.com> wrote:
>
>
> On Tuesday, December 13, 2011 at 11:18 AM, Julian Reschke wrote:
>
>> (a) Following the link might get you a 404. Yes, cool URIs do not
>> change, but unfortunately some orgs do not get that.
>
> Some go out of business, so sometimes it's not their fault… sometimes things just change (e.g., companies change names and their old domains are forgotten about).
>> When that happes,
>> additional information like the full title, the organization, the date
>> and the authors can help finding the document somewhere else.
>
> Question is still if you need all four. Certainly, all four provide a lot of redundancy and fallback, but that comes at the cost to the Editor: I'm personally tired of having to keep my references up to date (it's very time consuming), so I'm simply not doing it any more unless it can be proved that it is not possible to find a document with just the old URL and the title… I've actually kept the organization in my specs, but that's it. I also see others moving towards this model (e.g., HTML5, DOM4, which have kept author and org, but have dropped showing dates, status of document, and "available at" etc. which is seen in "classical" referencing).

This is a tools question. XML2RFC (popular in the IETF) provides a
bibliography with both versioned and stable identifiers for the main
bodies whose work is popular to reference in the IETF. The "printed"
version will give correct information *at the time of publication*.

The references need to show what the spec author was referring to when
he wrote the text he wrote. If a spec is updated after that, either
the reference will still be valid (no problem), or it will be broken
because of the update - in that case, the people reading the spec need
to know that the text they are reading does NOT refer to the newer
specification.

>
> This is what we used to see:
> [XML]
> Extensible Markup Language (XML) 1.0, Tim Bray, Jean Paoli, C. M. Sperberg-McQueen, Eve Maler, and François Yergeau eds. W3C (World Wide Web Consortium). Available at http://www.w3.org/TR/REC-xml/.
>
>
>
> This is what we see today in HTML5 (note the version drop also!):
> [XML]
> Extensible Markup Language , Tim Bray, Jean Paoli, C. M. Sperberg-McQueen et al.. W3C.

This removes the information needed for the disambiguation I refer to above.

If ever XML goes to version 3.0, and removes or modifies stuff that
HTML5 depends on, there will be no way for a reader to tell for
certain what HTML5 intended.

That's bad.
Received on Tuesday, 13 December 2011 15:35:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 10 March 2012 06:19:18 GMT