OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   Re: [xml-dev] Re: Where does the "nothing left but toolkits" myth come f

[ Lists Home | Date Index | Thread Index ]


1) I agree that the blanket statement "binary file formats are *broken*" is
inaccurate to the point of absurdity.
2) However, some of these things are not like the others:

On Mon, Feb 07, 2005 at 11:58:22AM -0500, Liam Quin wrote:
>On Mon, Feb 07, 2005 at 10:42:13AM -0500, Elliotte Harold wrote:
>> Binary file formats are *broken*
>Hence the failure of

A protocol specification, not a file format.



>MPEG 3 (mp3)

A good set of examples: these binary formats encode data in particular ways.


Transformative formats for the compression of redundant information, but not
actually data formats themselves.


A protocol specification, not a file format.


Another good example, although (with PS), you could argue that this is a
programming language, not a file format.  PS and PDF can both (I think) be
represented as text, though; they simply have a compact binary alternative
(and the two can be mixed in a single document).


Marginal example.  So far as I know, .exe is very rarely used as a data file

Checking MIME types, other possible examples might include .dvi, .flac,
.pfb, .mif, virtually all the audio/*, video/*, and image/* formats (with a
few exceptions, which are mostly XML, interestingly enough).

3) Almost all the examples are application-area-specific.  I do not know of
a general data-encoding file format in common use with the expressivity of
XML.  If ASN.1 encodings were generally associated with file types (rather
than with protocols), then we might expect to see .ber, .per, .der.  The
other general-purpose data storage formats that seem to appear in MIME types
are mostly found in ... text/*: .csv, .tsv (XML insists that everything is a
tree; those two formats insist that everything is a table, but generally
both are suitable for "generic data transmission" with moderately well-known
semantics for establishing what content actually is).  There don't appear to
be many well-known generic binary database formats, or well-known generic
binary data tree formats.


>characters and text.  But even a "plain US ASCII text file" is in fact
>in a binary format.

Oh, *sigh*.  That's a truly silly argument, in my opinion.  By that measure,
we *already* *have* binary XML.  Okay?

Amelia A. Lewis                    amyzing {at} talsever.com
  "Ruby fruit jungle?"
  "Yeah, women are thick and rich and full of hidden treasures and 
besides that, they taste good."
                -- Rita Mae Brown, "Rubyfruit Jungle"


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS