Andy Holt wrote:
But, none of
this answers the query of "Why file type information
has migrated into the file name"... :> (persistent, eh? :>)
Probably because in RT-11/CPM/DOS/(most unix) there is/was no where else to
keep it (except within the main body of the file itself which would then
mean that any application that touched the file would have to be aware of
it).
Note my point that the OS shouldn't *care* (IMHO) about the type.
E.g., nowadays, this information is typically used by The Desktop.
So, why can't The Desktop assume responsibility for maintaining it?
Using whatever mechanisms *it* deems efficient.
To me, having
"something" (IMO, *above* the filesystem) that
tracks the "type" of a file -- for the benefit of users *of*
that file -- is a desireable adjunct *to* the OS.
There is a lot to be said for the NTFS approach of allowing any number of
file attributes to be stored along with the file (pretty much a
Hmm.. I was unaware that NT did this. I thought it just embellished
the ACL's, etc.
generalisation of the Mac "forks") and
accessible to anything that "needs to
know". But then hardly any software actually supported this feature and
copying a file could silently lose vital information.
[though, having made these arguments, I am
building support
for that mechanism *within* my OS -- since it is a key service
that all apps *must* use, in my case :< ]
Yep, I think we agree more than we disagree - but a polite public argument
can always be a pleasant relaxation.
As I have stated previously, "collecting" is not my interest.
Rather, I see older machines as "lost ideas". Some ideas
probably *should* have been lost (e.g., Harvard architecture,
non-binary radixes, etc.) but others appear to have been
*unfortunately* lost.
My point in starting this thread was to decide which category
this idea falls into :-/
I'm just amused by the way that 30+ years after
Unix drasticly simplified
the powerful Multics filesystem, most of it has been reinvented.
Exactly! Given another 10 years, we'll return, once again,
to the "mainframe with terminals" model that the industry
revisits almost as regularly as the four seasons! (though
the role of The Mainframe will now be played by The Internet)
Now if we just added-in the automatic migration of
files that George 3 had
40 years ago and automatic file generation management as ICL had then and
VAX had from its outset ...
[but then I'm drifting off your topic, again]
[and I've no idea why my ISP's mail server decided to flag your message as
SPAM!]
Maybe it *is*!! ;-)