Let's develop an open-source media archive standard

From: melamy_at_earthlink.net <(melamy_at_earthlink.net)>
Date: Thu Aug 12 13:24:34 2004

you have to think beyond emulators and think of just getting the data. You
started this whole thing out by asking for ideas and concepts to make sure
that disks could be made in years to come. I challenge you to expand the
horizon and suggest that what you need is a way to get the data out easily
and make disks.

Personally, I have zero use for a image file format that can only be used
to make disks and makes getting data out of it a pain in the backside.
Maybe in years, I will need to make a floppy disk, but I believe that
getting the data out (files or whatever you choose to call them) is more
important and will be that way in the future also. Where small things can
get lost, having a "standard" image file format that let's me do both makes
more sense to me.

As for emulators, Microfast by Simeon Cran uses an import from ms-dos and
the emulator runs great. The original BabyBlue card wrapped the cp/m
program in a ms-dos wrapper and then you didn't even have to think twice
about running a compiler, linker, etc. Would it do direct disk access? No,
but that is what the original platforms are for. Most programs went through
the OS for disk access because they were file oriented. I can't recall how
Fred arranged his emulator, but I have a copy of that to.

Your arguement about emulator programmers wondering "why" about binary is a
non-issue. The spec defines ascii and that is the way data is represented.
They will probably more appreciate the the standard then complaining about
such things. If they do, then they certainly don't need to write anything.
A user could easily write the utility or someone who is part of the
standards group will make software templates and simple utilities
available. Writing the utilities would be something that I wouldn't mind
helping with for example.

The image file that you are discussing is only good as a "how to" to write
arbitrary data to arbitrary media. It is only good for creating a final
media image because the format being contemplated (mixing data into the
physical spec) will preclude the file from standing on its own. I was
proposing identifying data blocks as files if that is what they were. The
data blocks would have been sequential and trivial to retrieve. That is
hardly what I would call defining a file system as part of the image file.

best regards, Steve Thatcher

Original Message:
From: Vintage Computer Festival vcf_at_siconic.com
Date: Thu, 12 Aug 2004 10:25:20 -0700 (PDT)
To: cctalk_at_classiccmp.org
Subject: RE: Let's develop an open-source media archive standard

On Wed, 11 Aug 2004, Steve Thatcher wrote:

> what is wrong with making things easier? If a small amount of effort
> could be put out to make the image file have a more useful purpose then
> to ONLY be able to create media, why not? Your view seems to be to make
> all the emulators tyhat are currently available not work anymore because
> now they have to be modified to access an image file. My idea is to not


All emulators for 8-bit machines today that I'm aware of use disk images.
Emulators like SIMH can read in individual files by, for instance, turning
the file into a virtual paper tape reader.

Your point, again, is taken, but so is Dwight's. I think it depends on
what emulators you spend most of your time playing with which defines your
outlook on this.

One problem I did think of after I sent my last message concerning this is
that if you do start including filesystems, people (probably emulator
writers in particular) will start to wonder why you didn't store the data
in binary. "Why," they ask, "do I have to convert the data from ASCII
when the file is fully capable of holding binary data?" Then you might
get some joker making unauthorized extensions to the spec with binary data
fields. Perhaps I'm getting carried away here, but I do see this as
potentially problematic.

> If you thought of an emulator is to BE the machine in absolutely all
> respects then yes who ever is writing it needs to understand the file
> structure intimately and we need to add even more to the specification
> to allow writing to an image file...

Isn't the specification itself a "how to" for writing an image file?

Also, the image file we are discussing is *NOT*, again, intended to serve
the emulators. Emulator usage is a SECONDARY concern.

Sellam Ismail                                        Vintage Computer
International Man of Intrigue and Danger               
[ Old computing resources for business || Buy/Sell/Trade Vintage Computers 
[ and academia at www.VintageTech.com  || at http://marketplace.vintage.org
mail2web - Check your email from the web at
http://mail2web.com/ .
Received on Thu Aug 12 2004 - 13:24:34 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:34 BST