Re: MIME and the Web

In Larry's document,

Le 29 mai 2010 à 23:56, Larry Masinter a écrit :
> This is a draft  (for TAG ACTION-425) for discussion at an upcoming W3C TAG meeting.

> MIME and the Web


> Where we need to go


> however the following is a partial list of documents that should be reviewed & updated, or new documents written

In this list, I would add a bug tracker with all issues related to mime types.

In the history part of the document, I have the feeling that there is a missing statement along this line:

        The Web is a write-read (order is important)
        metaphor. You could create document using HTTP
        and read them using HTTP. Unfortunately, very
        early, people didn't have most of the time tools
        for writing documents on the Web (using HTTP),
        and they were using authoring tools in their
        operating systems metaphor with different kind
        of identification systems for the type of files.
        The main two ways of creating files which would
        be accessed on the Web were: working in the
        filesystem directly or using FTP for uploading
        the files from a local computer to a server.
        That creates a whole set of issues as in we do
        not have to manage the mime type in the
        transaction but rely afterwards on a mapping
        between the server filesystem and the Web

It is one of my interpretations that might explains the mess and why we are used to have URIs with an ending such as

when this just works

Also in Cool URIs don't change, there is a note on removing file extension

Also Guideline 10: Serve resources with correct content-type and character encoding information

Karl Dubost
Montréal, QC, Canada

Received on Wednesday, 2 June 2010 13:43:29 UTC