Tim,
   The idea is to do things in 2 parts with FileStream (Standard or
otherwise) being the link between the FileSystem and Codecs. I envisage 2
packages one for the filesystem framework and one for the Codec framework.
Other packages will add codecs to the framework so Ogg would be a separate
package (as would MP3, JPEG, etc).

I will search the archives for the innumerable discussions but could I ask
you what you think is important? 
I have a URI implementation that I like and Michael Rueger is sending me
another. The main difference between mine the others I have seen is that I
separate out the extension allowing mapping of filename and extension on
systems that don't support it natively (i.e. RiscOS).

My plan is to have nearly everything written in Squeak with very basic
primitives, that way it allows more control.

Fundamentally there are 3 classes:
File handles file access
Directory handles directory access
FileSystem handles the raw access and provides a unified interface for Files
and Directories.

Russell


> -----Original Message-----
> From: squeak-dev-bounces at lists.squeakfoundation.org [mailto:squeak-dev-
> bounces at lists.squeakfoundation.org] On Behalf Of Tim Rowledge
> Sent: Friday, 17 December 2004 9:07 AM
> To: squeak-dev at lists.squeakfoundation.org
> Subject: RE: About a new roadmap for 3.9
> 
> "Russell Penney" <russell.penney at tincanct.com> wrote:
> 
> > > > I am up for cleaning my reader however I was planning to
> > > write a URI
> > > > based file framework to replace what lives under FileStream and I
> A small bit of advice - don't start off by tying them together. A
> better filesystem is a big thing and needed much more generally than
> for a music player.
> 
> I assume you've exhaustively read the archives for the innumerable
> discussions we had on the subject of file system rewrites? Some of it
> actually included some good ideas. I think you should find some useful
> code relating to URIs etc if you track it down.
> 
> The key problem with something like a new file system is not the large
> amount of work it involves (been there, done that several times too
> often) but the work of getting people to adopt it. You'll need all the
> good wishes that will surely be beamed your way from those of us
> wanting an improved FS.
> 
> 
> tim
> --
> Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
> Useful random insult:- Settled some during shipping and handling.
> 
>