W3C home > Mailing lists > Public > www-archive@w3.org > November 2008

Re: ZIP-based packages and URI references into them ODF proposal

From: Marcos Caceres <marcosscaceres@gmail.com>
Date: Fri, 28 Nov 2008 20:45:45 +0000
Message-ID: <b21a10670811281245l30548bdfu48588cbe0eefb966@mail.gmail.com>
To: "Larry Masinter" <masinter@adobe.com>
Cc: "Arthur Barstow" <art.barstow@nokia.com>, "Jon Ferraiolo" <jferrai@us.ibm.com>, "Richard Cohn" <rcohn@adobe.com>, "Bill McCoy" <bmccoy@adobe.com>, "Henry.Story@Sun.COM" <Henry.Story@sun.com>, "Michael Stahl" <Michael.Stahl@sun.com>, "www-archive@w3.org" <www-archive@w3.org>, "Svante Schubert" <Svante.Schubert@sun.com>, "eduardo.gutentag@oasis-open.org" <eduardo.gutentag@oasis-open.org>, "Philippe Le Hegaret" <plh@w3.org>, "Carl Cargill" <cargill@adobe.com>, "Stephen Zilles" <szilles@adobe.com>, "www-tag@w3.org" <www-tag@w3.org>

On Fri, Nov 28, 2008 at 6:14 PM, Larry Masinter <masinter@adobe.com> wrote:
> The architectural principle that seems to be missing the orthogonality of content-type, protocol, and identification -- that HTML shouldn't specify HTTP or be dependent on it as the transport protocol, that the packaging system shouldn't depend on HTTP being the protocol being used to access the package or components within it.
>

Agreed; as a design principal, aspects should be designed as
orthogonal. However, this often does not reflect practice because, as
HTML5 shows, the overall running system must overcome incorrect or
idealistic assumptions that were made when each layer (protocol,
content-type, identification) was architectured.

For instance, HTML5, IIRC, defines handling malformed URIs because it
is not defined in the URI spec. HTML5 also defines content-type
sniffing for when the MIME type of a resource is unknown because
servers are often misconfigured or label resources with the wrong type
or no type at all. In fact, even though you say that "HTML shouldn't
specify HTTP or be dependent on it as the transport protocol", HTML5
does redefine how aspects of HTTP should be interpreted by the
browser: section [1], for instance, notes, "...the above algorithm is
a willful violation of the HTTP specification".

The point being that: yes, orthogonality is fundamentally important.
But the system also needs to be architectured to cope with
inadequacies or inconsistencies of the layers it depends on. Maybe a
better principle would be: design orthogonally but assume your
dependencies are broken. Or, maybe, design for the problem you are
trying to solve, but don't assume you've have solved any other
problems.

Anyway, my proposed solution was supposed to show that HTTP could
solve the problem of addressing inside packages without requiring any
extension to HTTP's URI scheme.

Kind regards,
Marcos

[1]
http://www.whatwg.org/specs/web-apps/current-work/multipage/infrastructure.html#content-type-sniffing:-unknown-type

-- 
Marcos Caceres
http://datadriven.com.au
Received on Friday, 28 November 2008 20:46:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 November 2012 14:18:19 GMT