🚀 go-pugleaf

hi rawn,

you should be aware that iso x3b11.1 is an optic filesystems architecture
for spanning all os's, single drives, autochangers, worms, erasable,
and cd-wo. it has been proposed/approved to ecma this summer and is 
due to be approved by iso sometime next year. 

i dont know anyone besides myself that is working toward a commercial
offering, but ed bayshore(edb@hpgrla.gr.hp.com) can tell you more.

jack...
=============================================================================
 Jack Craig  Microport         SVR4 408-438-UNIX   uunet!mport.com!jack
             Software Horizons SVR3 408-684-1375   uunet!sco.com!blznrzn!jack
=============================================================================

rawn@manduca.neurobio.arizona.edu (Rawn Shah) writes:


>    One problem we came across immediately is that even if the drives were
>compatible (same disk size & conforming to ISO), the actual file system on 
>the disk as saved by the driver software differs; on the Sun its a SunOS 
>file system and on the DOS-based 486, it uses the FAT tables and such.
>So data stored on a disk on the DOS system cannot be read by the MO drive
>on the Sun and vice versa.

>    I had the idea that I might stick a DOS emulator on the Sun and have
>it read the disks in DOS format. BUT as far as I know the DOS emulator
>I looked up (SoftPC for SunOS 4) does not have any drivers for r/w optical
>drives and even if I did NFS-mount the drive (the other option given by
>SoftPC) the actual file format is still SunOS and so the disks wouldn't
>be readable by the DOS machine.

>    As far as I know so far there are no compatibility measures on such
>kinds of drives. Now there're phase change r/w optical drives but I still
>haven't heard of any compatibility between systems either.

I have two solutions for you. One is hacking the mtools a bit. The mtools
implement the DOS file system portably.

The other is using FILEdriver from OSS. This is a WORM filesystem that
has been implemented on DOS, Novell, and SunOS. Of course the file system
has the same structure on all three platforms; even the code derives from
a common source.

The drawback is that while you can run a WORM filesystem on any random
access media, it will assume WORM behaviour: any sector can be written
at most one time. This seems to be OK with your application, though.
Of course, for MOs FILEdriver offers a "wipe" utility to erase the disk.

If you are interested in FILEdriver, call OSS:

	Optical Software Solutions, Inc.
	Keith Ragsdale
	1956 Marion Court
	Lafayette, CA 94549
	Tel (510) 256-1066
	FAX (510) 256-6461
	Email uunet!ossi.com!ossie!keith
-- 
| ...!unido!ukw!lupe  (German EUNet, "bang")     |  Disclaimer:            |
| lupe@ukw.UUCP       (German EUNet, domain)     |  As I am self-employed, |
| suninfo!alanya!lupe (Sun Germany)              |  this *is* the opinion  |
| Res non sunt complicanda praeter necessitatem. |  of my employer.        |

In article <1231@admin.mport.COM>, jack@admin.mport.COM (Jack Craig) writes:
> hi rawn,
> 
> you should be aware that iso x3b11.1 is an optic filesystems architecture
> for spanning all os's, single drives, autochangers, worms, erasable,
> and cd-wo. it has been proposed/approved to ecma this summer and is 
> due to be approved by iso sometime next year. 
> 
> i dont know anyone besides myself that is working toward a commercial
> offering, but ed bayshore(edb@hpgrla.gr.hp.com) can tell you more.
> 
> jack...


	ummm, let me clarify this a little. i am the technical editor
for this standard and have slightly more up to date information.

	ECMA-167 is a file system format standard for random access media,
both write-once and rewritable. it is intended for optical media
but will work just fine for magnetic media as well.

	this standard was approved by the ECMA General Assembly on June 26
this year and will be published shortly (i expect the last galley proofs
very soon now). some time after it is published, with a worst case
of December, we expect it to be forwarded by ECMA to ISO for fast track
processing (meaning it becomes a DIS straight away and the 6 mth ISO
ballot starts immediately -- actually 2-3 weeks later).

	the format is designed for interchange amongst a large number of
systems but most attention was paid to PC-DOS and Posix systems
(and to a lesser extent, VMS). It doesn't mention jukeboxes explicitly,
but several members of the developing committee will be using it in
a jukebox environment; the appropriate hooks are there.

	the work was mainly done in the USA in the X3B11.1 committee.
We interacted a lot with ISO SC15 and ECMA TC15 and with teh
Japanese National Committee (which is why we anticpate little trouble
passing the ISO ballot).

	there are at least three othe rimplementation efforts underway;
two, apparently, quite far along. as jack said, ed beshore (the
x3b11.1 committee chair) is the right contact point. his full info is

Edward Beshore
Phone: +1 303-350-4826
FAX: +1 303-350-4675
email: edb@hpgrla.gr.hp.com

	there is a bunch of stuff available about ecma-167.
there is a high-level overview, a programmer's guide and a
very recent draft of teh standard. mail me for details.


			andrew hume
			andrew@research.att.com