
	"To do" list for MRBackup:

1.  Pathname inclusion list:  this feature will be mutually exclusive
    of the pathname exclusion list and will take priority.  That is, 
	if both an inclusion and exclusion list are specified, only the
	inclusion list will be used.  The inclusion list will allow backup
	of multiple, disjoint directories.

2.  Add new capabilities to the user initialization file, including 
    window sizes and positions, default date delta for backups, color
	settings, etc.

3.  ERROR RECOVERY: the current release of this program is not very
	tolerant of errors (especially due to output media faults).  This
	is an unacceptible condition which will be addressed soon.  The
	error recovery codes (ERR_NONE..ERR_IGNORE) comprise the bits of
	a "set" of error options which will be selectable via a special
	dynamically created requester ("An error occurred...here are your
	options...").

4.	It already talks, so maybe I'll add some canned music - for those
	who need entertainment while performing the mundane chore of
	backups :-).

5.	User requests: if you have any, write them up and send this back to
	mark@unisec.USI.COM.


	Bugs and User Requests

09/01/87 - Jay Denebeim
	The listing file/device cannot be reassigned if it is in use by 
	another program when MRBackup starts up.  

	Status:	Fixed by MRR in V1.3, 09/03/87.


09/01/87 - Jay Denebeim 
	Add the capability for a file to span multiple volumes.

	Status: Open.  Jay recommended the possible use of unused space in 
	certain system sectors for maintaining this info.  I think I prefer 
	a "volume information file" for recording the characteristics of files 
	with "special" characteristics.  I certainly endorse the idea and 
	recognize the need, though a file larger than ~850K is a whopper.

09/01/87 - Jay Denebeim
	Add recognition of .ARC and .ZOO files as compressed files.

	Status: Fixed by MRR in V1.3, 09/03/87.

09/01/87 - Jay Denebeim (That WAS a long phone call, Jay! :-) )
	There are several problems with the pathname requesters.  It appears
	that changes to the gadget text which are not properly terminated
	with RETURN are left in the gadgets, rather than reverting to the
	original text.  Thus, there is a mismatch between the program data
	and the requester data.

	Status:	Open.


09/01/87 - Jay Denebeim
	Use better a better buffering scheme for compression.

	Status: Done - any buffering is better than none, right?  To
            accomplish this, I had to "absorb" the CopyFile command
			from the library, rewriting it to use a globally allocated
			buffer which can then also be used by compress/decompress.  
			I've done some pretty thorough testing with partial backups 
			and restores and I think all is well. Please do some testing 
			of your own before you commit anything significant to the 
			compressed format.  If you have problems, BY ALL MEANS, LET 
			ME KNOW A.S.A.P!  I'm not asking for any monetary compensation, 
			so you're morally obligated to report bugs back to me. If 
			you don't, your left foot is going to fall off (...if you have 
			one.  If you don't, I apologize for that crass remark - but
			you're still bound :-) ).

09/02/87 - Tony Cabot
	Extended menu selection on flag items gets confused.

	Status: Open, under investigation.  Actually, I'm waiting for someone
            to tell me what idiotic thing I've done or forgotten to do 
			that is causing this not to work.


09/02/87 - Armando Andreozzi
	Resize and/or relocate the STOP gadget as it interferes with the
	operation of the menu.

	Remarks: I agree, I think.  I'm not sure how it interferes with the
	         menu operation, but it probably should be less obtrusive.  
			 I may stick it in the window border or hide it under a rock.
	
	         Also, the STOP gadget should only appear when it is useful 
             (i.e. during Backup/Restore operations) and should be 
			 deactivated at  other times.

	Status: Open.

09/02/87 - Peter DaSilva
	Add AmigaDOS wildcard capability.

	Remarks: Yes Peter, and I'll use your code, too!

             P.S.  Peter informs me that the code isn't actually his, 
			 but I still appreciate the feedback and the idea.

	Status: Open.


