[comp.sys.amiga.tech] another 1.4 wish

xanthian@well.UUCP (Kent Paul Dolan) (08/20/89)

This one just bit me, and the folks at C-A should have known better!

Please change the name of "ram disk:" to either ram_disk: or just ram:;
changing the name to include an embedded space for 1.3 broke
working software!  For example, the Lattice C++ hello.sh demonstration
script won't work if the compiler is installed in ram:.  Instead,
a message to the effect: "parameters beyond file name ignored" comes
up, and is promptly followed by a delightful requestor to "Insert disk DISK".

It is sure nice to have icons with multi-word labels, but what I as a CLI
user think of a file system design that allows embedded spaces in names
cannot be expressed without profanity, and lots of it.

well!xanthian
Kent, the man from xanth, now just another echo from The Well.

jms@tardis.Tymnet.COM (Joe Smith) (08/21/89)

In article <13227@well.UUCP> xanthian@well.UUCP (Kent Paul Dolan) writes:
>This one just bit me, and the folks at C-A should have known better!
>Please change the name of "ram disk:" to either ram_disk: or just ram:;
>changing the name to include an embedded space for 1.3 broke
>working software!

I didn't allow that problem to bother me.  Put the following in your
startup-sequence, and it won't bother you.

	makedir ram:t
	assign t: ram:t  ;So nested EXECUTEs won't write to your boot disk
	relabel RAM: RAM_DISK
-- 
Joe Smith (408)922-6220 | SMTP: JMS@F74.TYMNET.COM or jms@tymix.tymnet.com
McDonnell Douglas FSCO  | UUCP: ...!{ames,pyramid}!oliveb!tymix!tardis!jms
PO Box 49019, MS-D21    | PDP-10 support: My car's license plate is "POPJ P,"
San Jose, CA 95161-9019 | narrator.device: "I didn't say that, my Amiga did!"

andy@cbmvax.UUCP (Andy Finkel) (08/22/89)

In article <13227@well.UUCP> xanthian@well.UUCP (Kent Paul Dolan) writes:
>This one just bit me, and the folks at C-A should have known better!
>
>Please change the name of "ram disk:" to either ram_disk: or just ram:;
>changing the name to include an embedded space for 1.3 broke
>working software!  For example, the Lattice C++ hello.sh demonstration

You can relabel the ram disk to be anything you like; however
you have now been warned that you have broken software...a space
has always been legal in disk names.  And, you've been warned
by a piece of software that you can always refer to by the name
RAM:  or relabel to something else,

This happened as a side effect of fixing the ram-handler for 1.3...
pre-1.2 RAM: didn't have a proper volume node, so lots of programs
had to a special case the ram-handler.

>It is sure nice to have icons with multi-word labels, but what I as a CLI
>user think of a file system design that allows embedded spaces in names
>cannot be expressed without profanity, and lots of it.

Either refer to it as RAM: or relabel it in your startup sequence.
-- 
andy finkel		{uunet|rutgers|amiga}!cbmvax!andy
Commodore-Amiga, Inc.

  The Law of Software Envelopment
  
  Every program at MIT attempts to expand until it can read mail.
  Those programs which cannot expand are replaced by ones which can.

Any expressed opinions are mine; but feel free to share.
I disclaim all responsibilities, all shapes, all sizes, all colors.

swan@jolnet.ORPK.IL.US (Joel Swan) (08/22/89)

In article <525@tardis.Tymnet.COM> jms@tardis.Tymnet.COM (Joe Smith) writes:
>In article <13227@well.UUCP> xanthian@well.UUCP (Kent Paul Dolan) writes:
>>This one just bit me, and the folks at C-A should have known better!
>>Please change the name of "ram disk:" to either ram_disk: or just ram:;
>I didn't allow that problem to bother me.  Put the following in your
>startup-sequence, and it won't bother you.
>	relabel RAM: RAM_DISK
>-- 
>Joe Smith (408)922-6220 | SMTP: JMS@F74.TYMNET.COM or jms@tymix.tymnet.com

I agree with Kent completely.  I also relabel RAM:, but I relabel it
to simply "RAM".   However, the best world would be to simply drop the
second word "DISK" from its name in the first place.  First, most other
devices are only 3 letters (DF0:, DH0:, VD0:, etc) and RAM as its name 
would fit better. (I realize device names and volume names are different, but
they CAN be the same if we want them to, right?)
Second, 3 letters is easier and faster to type, especially when it
contains no spaces (hence, no quotes).

BTW- I have a hard time finding a spot in my startup-sequence to relabel RAM.
It seems if I do it too early in my SS, I get an error code.  I also must do it
a few commands ahead of my loadwb if I remember right.  I think it would give 
me validation errors or something.
-- 
 - Joel E. Swan
[  swan@jolnet.UUCP                   <>  PLINK ID: Amiga*joel         ]
[  "Amigas.... for the rest of us."   <>  CI$     : 74746,3240         ]
[  "...peace with God through our Lord Jesus Christ."    Romans 5:8    ]

shadow@pawl.rpi.edu (Deven T. Corzine) (08/24/89)

On 22 Aug 89 01:48:15 GMT, swan@jolnet.ORPK.IL.US (Joel Swan) said:

Joel> BTW- I have a hard time finding a spot in my startup-sequence to
Joel> relabel RAM.  It seems if I do it too early in my SS, I get an
Joel> error code.  I also must do it a few commands ahead of my loadwb
Joel> if I remember right.  I think it would give me validation errors
Joel> or something.

What I do is start my startup-sequence by installing ConMan and
starting an AShell and killing the original CLI, but then the AShell
mounts RAD:, and attempts to "relabel rad: vd0"...  if it gets an
error on either the mount or relabel, it runs RemRAD, prints an error,
and does a "relabel ram: vd0" instead.  Subsequent references are all
to vd0:, copying files and assigning devices, etc.  This way, it will
use RAD: if it can, (and won't copy files it doesn't need to) and fall
back on RAM: if RAD: fails.  I no longer use the ASDG vdisk, but I was
used to "VD0:" so I used it for the volume label of the main ramdisk.
A "relabel ram: ram" precedes the "mount rad:" so if RAD: works, RAM:
will have volume label ram:.

*I* like it.  :-)

Deven
--
Deven T. Corzine        Internet:  deven@rpi.edu, shadow@pawl.rpi.edu
Snail:  2214 12th Street, Troy, NY 12180       Phone:  (518) 271-0750
Bitnet:  deven@rpitsmts, userfxb6@rpitsmts     UUCP:  uunet!rpi!deven
Simple things should be simple and complex things should be possible.

xanthian@well.UUCP (Kent Paul Dolan) (08/24/89)

Well, I read three responses to my request to change "RAM DISK:" to either
"RAM_DISK:" or back to "RAM:"; I liked best the one that started "I agree
with Kent completely" of course. ;-)

I can state with certainty that with my short term only memory, and return
to read this group only when I'm unemployed but flush, I can be used as a
pretty good approximation of a naive user in cases like this.

I _tried_ RELABELing RAM DISK:, and got back and error message.  Never in my
wildest imaginings did it occur to me that I would have to do it before
the workbench got loaded.  Now I will.

Never the less, Andy and other CATS types, this is a bug, not a feature,
since it is guaranteed to cause grief to old software, and to new users,
and most of all (I'm shouting here, listen up ;-) IT WASN'T NECESSARY!

I hope Carolyn got the copy of my letter to mwm I cc'ed to her; the gist was,
if you don't start making things lots, lots easier for the incredibly naive,
barely-ept user (me for example), your market penetration with the Amiga is
going to stay low, or decrease.  Word of mouth from the brand new owner to
friends and relations is your biggest sales tool, or greatest sales liability,
depending on what impression you make right at the start.  Take a long,
introspective look at what put the Mac in offices all over the country where
it was forbidden by official policy, and made it the success it is today.

The answer?  It did what people wanted, it did it right out of the box,
and it made the user feel in charge of the machine.  Excusing "features"
that cause the user grief on the basis that "the other guy's software is
broken" is the surest way to take Commodore right back to chapter 11.

Don't tell me it isn't your fault.  Don't tell me some arcane way to get
around it.  Don't tell me it is no big deal.  Fix it.

well!xanthian
Kent, the man from xanth, now just another echo from The Well.

(And by the way, if relabel works for RAM:, it should always work, as
should diskchange.)

jesup@cbmvax.UUCP (Randell Jesup) (08/25/89)

In article <13304@well.UUCP> xanthian@well.UUCP (Kent Paul Dolan) writes:
>I _tried_ RELABELing RAM DISK:, and got back and error message.  Never in my
>wildest imaginings did it occur to me that I would have to do it before
>the workbench got loaded.  Now I will.

	I forget about pre-1.3, but under 1.3 it works fine.  The current
Workbench wants you to do relabels through the Workbench Rename after WB is
started (it doesn't notice relabels, of ram: or of any other disk, from CLI).
Works fine.  Relabel before WB comes up, everything works fine.

>Never the less, Andy and other CATS types, this is a bug, not a feature,
>since it is guaranteed to cause grief to old software, and to new users,
>and most of all (I'm shouting here, listen up ;-) IT WASN'T NECESSARY!

	Everyone has known from Day 1 (and many before) that spaces were
allowed in file names and device names.  There is a limit on how much we can
do to make already broken programs work.  Far more important is to avoid
breaking correct programs.

-- 
Randell Jesup, Keeper of AmigaDos, Commodore Engineering.
{uunet|rutgers}!cbmvax!jesup, jesup@cbmvax.cbm.commodore.com  BIX: rjesup  
Common phrase heard at Amiga Devcon '89: "It's in there!"

ajbrouw@neabbs.UUCP (ALBERT-JAN BROUWER) (08/28/89)

In article <13227@well.UUCP> xanthian@well.UUCP (Kent Paul Dolan) writes:
>This one just bit me, and the folks at C-A should have known better!
>
>Please change the name of "ram disk:" to either ram_disk: or just ram:;
>changing the name to include an embedded space for 1.3 broke
>working software!  For example, the Lattice C++ hello.sh demonstration

My solution to this was to take l:RAM-Handler, search for the string
"RAM DISK" and modify the " " to a "_" as to let it read "RAM_DISK".
You can do this with NewZap or any editor that accepts binary files.
Modifying executable files may not be considered to be normal practice,
but I consider this to be the cleanest solution when opposed to some
time-dependant use of Relabel.
Err, BTW, when doing this, try to preserve the "----rw-d" flag pattern.

-- Albert.
## "To err is human -- to moo, bovine" (Nutworks magazine) ##