8 Apr
2021
8 Apr
'21
9:29 a.m.
On 4/8/21 8:40 AM, Eric Smith wrote:
On Thu, Apr 8, 2021, 09:34 Chuck Guzis <cclist at
sydex.com
<mailto:cclist at sydex.com>> wrote:
There's a big difference--in a WORM, unwritten sectors *mean* something.
For archival purposes, in what way does an unwritten sector on a WORM
mean more than an unwritten sector (with no data field present) of a
floppy? Neither can be accurately archived without representing the fact
that it is unwritten.
Simple--a WORM contains the entire history of information on the disc;
nothing is ever lost. A floppy can have data overwritten--and probably
does (e.g. directories and allocation maps) The only way to update a
WORM is to add to it.
--Chuck
8 Apr
8 Apr
11:38 a.m.
On 08/04/2021 17:29, Chuck Guzis via cctalk wrote:
On 4/8/21 8:40 AM, Eric Smith wrote:
I must be missing something. I understood (for WORM) "unwritten" to mean
"never written" rather than "written but its address is no longer in the
FS catalogue".
So with WORM (whether DEC RVxx style optical disk or CD/DVD-RW in
certain modes) you can (if the firmware allows) read sectors that
contain stale data.
But (afaict) a sector that has never been written is just that:
unwritten. How does it convey any information (other than "I'm still
blank")?
Antonio
--
Antonio Carlini
antonio at acarlini.com
On Thu, Apr 8, 2021, 09:34 Chuck Guzis <cclist
at sydex.com
<mailto:cclist at sydex.com>> wrote:
There's a big difference--in a WORM, unwritten sectors *mean* something.
For archival purposes, in what way does an unwritten sector on a WORM
mean more than an unwritten sector (with no data field present) of a
floppy? Neither can be accurately archived without representing the fact
that it is unwritten.
Simple--a WORM contains the entire history of information on
the disc;
nothing is ever lost. A floppy can have data overwritten--and probably
does (e.g. directories and allocation maps) The only way to update a
WORM is to add to it.
12:27 p.m.
On 4/8/21 11:38 AM, Antonio Carlini via cctalk wrote:
But (afaict) a sector that has never been written is
just that:
unwritten. How does it convey any information (other than "I'm still
blank")?
Let me try...
Because you can't rewrite or update tables without copying them, a WORM
file system that can be updated dynamically has to resort to lists of
previous versions of the tables. So to find the most recent copy of any
list, you read through the list until one of two things happens: You
hit an unwritten sector, which means that the last thing read
successfully is what you're after, or the preallocated list area is
completely full, in which case the list continues in another
preallocated area or is at the absolute end, depending on the
implementation.
In the WORM file system that I'm currently working in, the
implementation is a mixture of both. There are fixed preallocated
areas for directory information, but there are also lists of file data
areas written that can overflow to other preallocated areas.
Overflow of the directory is handled with a tree structure, so it's
pretty much the same thing as a linked list in a practical sense.
I assume that it was done that way so that a bad (error) sector wouldn't
upset the whole apple cart.
By the way, does anyone know the structure of an ".ITI" file?
--Chuck
2:39 p.m.
On 4/8/21 12:43 PM, John Foust via cctalk wrote:
At 02:27 PM 4/8/2021, Chuck Guzis via cctalk wrote:
Files extracted from a WORM file system; OS not relevant according to
documentation that I have (same product offered for several different
platforms). App appears to be some sort of document archiving system.
Said files contain no recognizable cleartext.
Ring any bells?
--Chuck
By the way, does anyone know the structure of an
".ITI" file
The context is a WORM file system, or a particular OS, or an app environment?
1354
days inactive
1354
days old
4 comments
3 participants
participants (3)
-
a.carlini＠ntlworld.com -
cclist＠sydex.com -
jfoust＠threedee.com