Re: XyWrite versus Command.com.

["Michael Edwards" mje@xxxxxxxx]

                         Michael Edwards.

----------------------------------------
[George Scithers:]

>Somewhere I think I read that XyWrite can operate without Dos's
>Command.com -- comments, anyone?
----------------------------------------

   I used DOS for many years, and it would surprise me if this were literally
true. But if you had enough memory (or used little enough memory) that once you
ran Command.com once it stayed in memory, I suppose it's possible it might (at
least until that area of memory was overwritten with something else) continue to
run from the copy in memory and not read the actual disk file again.
   On my desktop 486, which runs on DOS 6.22, I've often noticed that when I
run a file a second time it actually doesn't read it from the hard disk the
second, third, and subsequent occasions. (I can tell because of the light that
flashes whenever the disk is read, and also because the disk makes a
recognizable sound whenever it is read or written to.) Yet the computer is
clearly running the same file, so it has to be held somewhere in memory or
cache. In fact, my 486 seems to be far more efficient in such matters than my
Pentium laptop running on Windows 95 (which I use for all Internet- or
e-mail-related matters).
   I suppose it's possible the statement you read might be literally true if
XyWrite's main .exe file actually contained Command.com, or equivalent code,
within it, but of course I don't know yet whether that's true. But I feel
pretty sure that Command.com, or equivalent code in another file, has to be in
*memory* continuously, even if the computer can survive a while without reading
a copy from a disk.

   If it's important, there is one way I can think of for testing the matter
once and for all. (I say "if it's important" because it's a bit of a bother -
not something I'd do out of passing curiosity.) Prepare two floppy disks thus:
On one put the boot-up files: Command.com, IO.sys, and MsDOS.sys. Make sure you
can boot up off this disk by actually doing so and seeing that things work okay.
On the other disk, put whatever XyWrite files you need to run the program, but
NOT Command.com, etc.
   Boot up off the floppy disk, then once you have the command prompt, change
floppies and run XyWrite. This should work okay, because I'm pretty sure it is
routine for Command.com to be held in memory. XyWrite should work okay; but the
crunch will come when you quit that program and try to return to the DOS prompt.
In my experience, things should come to a halt because the computer needs a
system disk, and will give you a message to that effect, and it will be
necessary to insert the floppy disk containing Command.com. It will run again,
and then you'll be back to where you were after initially booting up.
   I don't really understand technically what's going on here, and why
Command.com appears to remain in memory while the other program is running, but
when you quit it appears to be lost, and must be read from the disk again. I
don't know if this always happens, and there may be many variable factors that
change this behaviour. Possibly computers and programs and DOS versions vary in
this, and it may be that you'll come back to the DOS prompt without needing the
DOS floppy disk again - but at least this experiment should test the matter and
resolve it one way or the other.
   If I'm not mistaken, many of the old computers without hard disks ran this
way: they had two floppy drives, and one floppy had the system files, and the
other had whatever program you wanted to run, and also held the saved copy of
the file you were working on. If you were prepared to change floppy disks as I
just described, this arrangement would work quite well with just the one floppy
drive.
   Obviously, you could experiment with various ways of using the computer
under this arrangement, and see how it behaves. By effectively quarantining
Command.com on a separate disk, you ensure the computer can read the file only
when you choose to let it do so, and this gives you control of things to
experiment with. (To be sure, it's still on the hard disk, but if you booted
off the floppy, for that session the computer thinks Command.com is on the
floppy, and it won't know about the hard disk copy.)
   Hope this helps. (If I've got things wrong, I'm sure other list members
can correct me, as I suppose this list has a higher concentration of DOS users
than computer users generally do.)

             Regards,
             Michael Edwards.