::scr Ramblings of a Classic Refugee or How I Learned To Stop Worrying and Love OS X

[Chris Devers]

On Tue, 5 Feb 2002, Simon Wistow wrote:

> Effectively DB file systems and also individual files would have the
> equivalent of DTDs (the file that describes an XML file for the buzzword
> protected) but for lovely shiny binary files. Al Snell would be proud. 

So what you're saying is that the binary format (ascii is binary, so this
includes text based formats too) can be formatted arbitrarily, but files
in these formats should be distributed with a universally machine readable
format description document. 

Yeah right. 

It's a good idea, and I'd like to see it happen, but I doubt it ever will. 
I can't see Microsoft ever publishing the specs for any of the formats
that the Office suite produces. I can't see much value in sending html
document around if they need the overhead of being bundled with descriptor
files that explain both ascii (surely an oo-inheritance scheme would cut
down on repeating info like that over & over) and html itself (along with
notes about common tolerated errors, like missing closing tags). I can
only assume that having to check one of these DTDs before rendering every
image file is going to slow things down unbearably. 

It seems like in most cases, file formats are either:

  * open [ascii, jpg, etc] in which case the code to read & write such
    files can be embedded into the system or major applications, which
    should be much more efficient than scanning a DTD every time anyhow

  * or proprietary [doc, xls, ppt] in which case some organization is 
    making money off the closed nature of the format and are not going 
    to be willing to publish it without government coersion or maybe
    coersion due to having the format reverse engineered on them

So you're looking at a middle ground there, where it's all open (sounds
nice), it's all dynamically read, rendered, and written (sounds heavy, but
could be feasible), and it will be actively resisted by major players
(which could be enough to, if not scuttle it, at least curtail it). 
 
> Of course there are *some* issues but if I hand wave sufficently then I
> can gloss over them.
 
Wave faster, your fingers aren't blurry enough yet :) 
 



--
Chris Devers

"Okay, Gene... so, -1 x -1 should equal what?" "A South American!"    
[....] "no human can understand the Timecube" and Gene responded
 without missing a beat "Yeah.  I'm not human."