[comp.sys.mac.digest] Info-Mac Digest V7 #139

Info-Mac-Request@SUMEX-AIM.STANFORD.EDU (The Moderators) (08/09/89)

Info-Mac Digest             Tue,  8 Aug 89       Volume 7 : Issue 139 

Today's Topics:
                            Administrivia
             Adobe Illustrator vs. 32-bit Color QuickDraw
                       Appletalk example eanted
                         Business calculator
                          dLib-demo-3.hqx01
                   Don't need a programmer's switch
                          Fonts techie info
                           HYPERCARD EVENTS
                         InfoBackup 1.0(Beta)
                    LZW Algorithm Explained (long)
                        Mathematical programs
                            Moire-3.hqx01
                        Query 340MB MiniScribe
                          Radius Accelerator
                       Searching/indexing files
                     Simple matrix manipulators?
           Wanted:Comments on Macintosh Programming Primer.
                  writing software for a FDHD drive

Your Info-Mac Moderators are Bill Lipa, Lance Nakata, and Jon Pugh.

The Info-Mac archives are available (by using FTP, account anonymous,
any password) in the info-mac directory on sumex-aim.stanford.edu
[36.44.0.6].  Help files are in /info-mac/help.  Indicies are in
/info-mac/help/recent-files.txt and /info-mac/help/all-files.txt.

Please send articles and binaries to info-mac@sumex-aim.stanford.edu.
Send administrative mail to info-mac-request@sumex-aim.stanford.edu.
----------------------------------------------------------------------

Date: Tue, 8 Aug 1989 14:49:19 PDT
From: The Moderators <Info-Mac-Request@sumex-aim.stanford.edu>
Subject: Administrivia

They said it couldn't be done... but, thanks to Glenn Trewitt, we now have
a way of keeping track of the use rates for all of the files in the archives.
Aside from being just plain interesting, this file helps us decide what to
delete when the archive gets too full. Check out /help/popular-files.txt.
Warning: the Times Read column saturates if the file is read more than once
in a fifteen minute interval. So truly popular files are even more popular
than they appear to be.

Also, sumex will be down Tuesday, August 22.

Bill

[Archived as /info-mac/help/popular-files.txt; 138K]

------------------------------

Date: Tue, 8 Aug 89 09:47:48 PDT
From: dplatt@coherent.com (Dave Platt)
Subject: Adobe Illustrator vs. 32-bit Color QuickDraw

	"Users of Apple's 32-bit QuickDraw have encountered some problems with
	Version 1.8.3 of Adobe Illustrator 88.  It seems that Illustrator
	crashes everytime you quit the application, but one user who reported
	the bug to the company said that a fix for the problem is not expected
	until the end of the year."

Hmmm.  This sounds as if it might possibly be the same problem I encountered
when adapting my MandelZot fractal program for 32-bit Color QuickDraw.
The system would crash on exit 100% of the time, if it was running
under the single-Finder and if it had ever created a custom palette and
applied it to an on-screen window.  The crash would not occur under
MultiFinder, or if I commented out the SetPalette() call that applied the
palette to the window.

The crash occurred after the application itself had been deleted from
memory, as the Finder was being re-launched.  I infer that it may be
occurring when the new 32-bit Color QuickDraw's Palette Manager
attempts to repopulate the device's CLUT with the default set of
colors... but I wasn't able to prove this hypothesis.

I found a fix for the problem... and I do _not_ understand why the fix
works.  The fix is simply this:  I made sure that the code-segment
containing the THINK C "MacTraps" glue-code has its "locked" bit set in
the application file.

I cannot for the life of me figure out why this should matter!  The Segment
Loader is supposed to lock down all CODE segments after loading them,
and unlocks them only upon receiving an _UnloadSeg call from the
application.  The segment in question was never unlocked when it was in
memory, as far as I could tell when I poked around with TMON... it was
loaded when the application was launched, was locked by the Segment
Loader, and never moved nor was unlocked thereafter.

What the heck... I now leave this segment locked in the application
file, to ensure that this problem does not occur.  I've forwarded the
application to the Palette Manager team at Apple to assist them in
reproducing the problem;  I haven't heard any news from them.

If somebody out there has encountered this problemw with Adobe Illustrator
1.8.3, s/he might want to try making a copy of the application, and then
using ResEdit to set the "locked" bits on each of the application's
CODE segments in turn, one at a time.  The problem might just go away!

Dave Platt    FIDONET:  Dave Platt on 1:204/444        VOICE: (415) 493-8805
  UUCP: ...!{ames,sun,uunet}!coherent!dplatt     DOMAIN: dplatt@coherent.com
  INTERNET:   coherent!dplatt@ames.arpa,  ...@uunet.uu.net 
  USNAIL: Coherent Thought Inc.  3350 West Bayshore #205  Palo Alto CA 94303

------------------------------

Date: Tue,  8 Aug 89 09:11:00 -0400 (EDT)
From: "Norman William Franke, III" <nf0i+@andrew.cmu.edu>
Subject: Appletalk example eanted

I`d also like some example AppleTalk code, if possible. Though I`d prefer
Pascal, LSP if it matters.

Thanks,
Norman Franke
nf0i+@andrew.cmu.edu

------------------------------

Date: Sat, 5 Aug 1989 19:58:08 EDT
From: "Juan M. Courcoul" <PP838474%TECMTYVM@icsa.rice.edu>
Subject: Business calculator

On Thu, 27 Jul 89 11:15 EST Dan Henderson said:
>In response to the request for a business calculator on the Mac, I
>have seen a DA called 12-C that is a duplication of the HP 12C in DA form. It
>comes in two parts (at least when I saw it), an INIT and a DA.  I haven't seen
>it in about a year, but it was great.  If you run across it, or anyone happens
>to have it, I would love to get a copy again! ( or any other RPN business
>calculator).

Included you will find a stuffed and binhexed copy of the HP 12 Calc DA. I
haven't the faintest idea who was the Mac wizard who cooked it up, but it IS
a neat piece of code. Seems to work fine in most machines (couldn't test it
on SE/30's, cause I don't have any available), even though it's a program
dating from 1984-85 (ancient history, in the Mac scale). I guess it comes to
show how stable a program can be, if the developer follows the rules.

Enjoy,

Juan


[Archived as /info-mac/da/hp-12c-calculator.hqx; 14K]

------------------------------

Date: Mon, 7 Aug 89 14:24:59 EST
From: munnari!munmurra.cs.mu.OZ.AU!jkjl@uunet.uu.net (John Keong-Jin Lim)
Subject: dLib-demo-3.hqx01

This is a demo of d'Librarian 3.0, a shareware disk cataloguer
for the Mac. Online help is available, and a DA for quick
searching of missing files is included. This demo limits you
to recording 255 files.

	John Lim 

[Archived as /info-mac/demo/dlibrarian-30.hqx; 131K]

------------------------------

Date: Tue, 8 Aug 89 12:22:36 PDT
From: USERQKMP@cc.sfu.ca
Subject: Don't need a programmer's switch

To the dude that lost his programmer's switch:
A better solution than hunting down a switch would be to install the
Programmer's Key INIT (part of TMON but available widely...like in the
info-mac archives) which changes the power-on key on a SE/II to a 
debugger/shut down/restart/etc. key.  Great utility.
Alex Curylo Simon Fraser University (604) 298-8913

------------------------------

Date: Tue, 8 Aug 89 09:57:03  19
From: "Darren" <dstalder@gmuvax2.gmu.edu>
Subject: Fonts techie info

I am designing a postscript font to go with a bitmap font that I have.
I would like to keep the original bitmap font but couple it with the
postscript one.  Can anyone tell me how to modify a font such that it
has a pointer to a postscript font?  I know that the flag word in the
FOND for a bitmap-only is $6000 and for a bitmap with postscript is
$0800.  Is there somewhere in the FOND (in the table maybe?) that the
name is kept or is it just derived from the name of the font?

One last thing to people in the know font-wise:  Is there a set of
numbers reserved for private developers or do you have to be a font
company to be able to have font numbers that won't conflict with other
people?
--
                  Torin/Darren Stalder/Wolf
Blessed         Internet: dstalder@gmuvax2.gmu.edu
  Be!           Bitnet:   dstalder@gmuvax
                ATTnet:   1-703-883-5747
      Hail      uucp:     multiverse!uunet!pyrdc!gmu90x!dstalder
        Eris!   Snail:    1350 Beverly Rd., Suite 115-223/McLean, VA 22101/USA
DISCLAIMER: I have enough problems getting credit what I do do for
            work, much less this stuff.

------------------------------

Date: Tue, 08 Aug 89 21:17:30 SST
From: TNG TH <ISSTTH%NUSVM.BITNET@forsythe.stanford.edu>
Subject: HYPERCARD EVENTS

I need help, again!
I have written a SuperCard 1.0 Project containing some 10 Meg of data.
What I need is a way to detect in the background that there are no events
for a period of time and the project runs itself again.
Having an on Idle handler in the project does not work. How can I know if the
user has move the mouse, or click the mouse, or press the keys etc? Is
there something equal to a mouseIdleTick function?
Please help me.

------------------------------

Date: Mon, 07 Aug 89 09:49:37 GMT
From: "J.M.L.Martin" <LUCTHSCH%BDILUC11.BITNET@forsythe.stanford.edu>
Subject: InfoBackup 1.0(Beta)

Dear fellow-MacIsts,

Enclosed is a beta version of InfoBackup 1.0, a quick & dirty utility to backup
and restore the 'GET INFO' comments in a desktop file. A recommended procedure
for maintenance of the desktop file is now:

a) run InfoBackup to backup the Get Info comments
b) run BundAid to fix the BUNDLE bits
c) rebuild the desktop file in the well-known 'Option-Cmd-Quit' or 'Option-Cmd-
mount disk' manners
d) run InfoBackup to restore the comments.

If no files are copied, moved, created or deleted in between any of the four st
eps, all Get Info comments should appear in the right places (It might also be
true under certain other circumstances, but this cannot be guaranteed).
Bug reports or comments are welcome at LUCTHSCH#BDILUC11.BITNET.
The program is charityware (concept invented by Steve Christensen of Superclock
fame): if you find it useful, donate any sensible amount to any relief fund
without political or religious biases (such as Unicef, Oxfam, or the Red Cross)

                                Happy Mack-ing,

                                Jan M.L.Martin
                                Quantum Chemistry
                                Department SBM
                                Limburgs Universitair Centrum
                                Universitaire Campus
                                B-3610 Diepenbeek, Belgium

Disclaimer: IBM just changed its name to I'd Buy a Mac

[Archived as /info-mac/util/infobackup.hqx; 20K]

------------------------------

Date: Tue, 8 Aug 89 11:00:55 edt
From: abboud%cuavax.dnet@netcon.cua.edu (Hisham)
Subject: LZW Algorithm Explained (long)

This is in response to Rocky Olive's mail, inquiring about LZW algorithm.
     
> Attention programmers:
>     
> I was reading the /info-mac/art/gif/gif-format.txt document about
> gif files, and I was doing great until the part about LZW compression.
> I've heard of LZW compression before, but I've never seen anything
> that explained the algorithm (in detail).  I'm contemplating the
> conversion of some image files to .gif format and I really would like
> to know if anyone has (or can write up :-) ) some description and
> examples of the LZW compression and maybe even how it relates to gif.


I implemented the LZW algorithm last year in assembly language (on an
IBM-PC, sorry!!) to capture and store screen snapshots.  I am not that
familiar with the GIF format, but I'll try to explain the LZW algorithm as
clearly as I could. 

First, let's forget about the W in LZW for a little while.

LZ stands for A. Lempel and J. Ziv, the two original authors.  You can think
about this algorithm as more like an "adaptive character set".  Let's take an
example, and assume that our "character set" has 12 bits per character,
instead of the familiar eight.  The set, or table, can thus have 4096
characters.  The first 256 are usually (surprise!) the same as the standard
ASCII code on your machine.  Now, let's compress the following string: 

	this is the test

We start scanning this string and run accross "t".  It already has an entry
in the table, so we output the code for "t" and continue scanning.  Next we
have an "h" so we now have "th".  "th" is not found in the table, so we make
an entry, code 257.  Then "h" is used to start the next string.  Since "h"
is in the table, we output the code for "h" and continue scanning.  And so
forth.  After a while, you will have entries in the table like "th" and
"is", so the next time you run accross these combinations, you output the
corresponding code.  An entry in the table does not have to be two
characters only, and can easily exceed thirty characters per entry. 

Once the table is full, a "clear code" is issued, the table gets reset, and
we start filling it all over again.  When decompressing, the reverse process
is used until the decompressor runs accross the clear code, at which time it
also resets the table and continues. 

This method has advantages and disadvantages.  It is a single-pass algorithm
and produces very efficient compressions.  But, when used in communications
for example, if a single bit gets flipped the wrong way while transmitting,
the entire message following it gets scrambled.  Also, the file you are
compressing must be long enough to take advantage of the adaptive algorithm,
otherwise there is a (slight) chance it may actually grow in size! 

The LZW algorithm is also very flexible (read: non-standard).  Examples:

    o	The table size is usually 12 bits, but it does not have to.  For 
	example, on the IBM-PC program I worked on, things were memory-
	resident, and with that stupid 640K limit, memory is at a premium.
	So we reduced the table size from 12 to 11 bits.

    o	The clear code is usually a null character (ASCII 0), but does not
	have to be.  Pick one at your convenience.  In the GIF format for
	example, the clear code is 2**<number_of_bits_per_pixel>.

    o	You do not have to assume that the first 256 characters are the
	same as the ASCII code.  In the GIF format, only the
	<number_of_bits_per_pixel> is assumed, i.e., if you have 4 bits
	per pixel (16 colors), you can start building the table at code 5.

Then the W (in LZW) came into the picture.  It stands for Terry A. Welch.
Before him, output codes were fixed-size 12 bits.  So Mr. Welch suggested
that by varying the output codes size, we can obtain significantly better
compression.  For example, for codes 257 to 511, only 9 bits are output. 
For codes 512 to 1023, only 10 bits are output, etc. 

The complete description of the LZW algorithm is described by Terry Welch in
"A Technique for High Performance Data Compression", IEEE Computer, vol 17
no 6 (June 1984).

I hope this helped.

					Hisham.


Hisham A. Abboud
Computer Center/Academic Services
The Catholic University of America, Washington, D.C. 20064

Bitnet:	   ABBOUD@CUA				   |  "God bless he who expects
Internet:  ABBOUD%CUAVAX.DNET@NETCON.CUA.EDU	   |   nothing, for he shall
    or	   ABBOUD%CUAVAX.DNET@192.31.193.2	   |   never be disappointed!"

------------------------------

Date: Tue, 8 Aug 89 13:01:30 EDT
From: Michael_Webb@ub.cc.umich.edu
Subject: Mathematical programs

Dear Netlanders:
 
A while ago, I requested info from you concerning Math analysis 
and graphing programs available for the Mac.  I have received a better than 
expected response, and I now wish to summarize to you.
 
I am only going to give pointers to software.  I have not done an 
analysis of any of the software described.  I cannot make an informed 
decision about any of them, and any decision made will be based on my 
own needs.  If I have left out any software, it is because I didn't know 
about it.
 
Theorist by Allan Bonadio Assoc.  The same people that brought 
you Expressionist will be bringing you a Mathematica-type program (as I 
am told) this fall.  It will be more "mac-like" and smaller.  (415)282-5864.
 
Maple.  Don't know anything about it yet.  Contact:
Ms. Iris Pietsch
Symbolic Computation Group
Department of Computer Science
University of Waterloo
Waterloo, Ontario
Canada N2L 3G1
(519)885-1211 ext. 3055
maple@watdcs.BITNET
 
Mathematica, of course.  No address.  Wolfram Research Inc.
Try Shawn@wri.com, head of technical support.
 
Igor by Wavemetrics.  Sounds neat, like a 2nd generation, souped 
up and more powerful version of Cricket Graph.  (I don't know if they 
would agree with that, it probably can't be compared with CG).  I've read 
some promising reviews.
Wavemetrics
PO Box 2088
Lake Oswego, OR  97035
(503)635-8849
Applelink D1832
 
GraphiC(TM) 1.0 by Taliaferro Software.
This is not an application, but a library of graphing functions you 
call from your own code.  Designed for THINK C, no mention of whether 
or not it is 4.0 compatible.  "Similar to what you might find on a mainframe 
computer".
Applelink D4320.
Scientific Endeavors Corporation (615)376-4146.  Don't know 
relation between Tailaferro and SEC.
 
 
If anybody knows of other software, please contact me.  I appreciate 
the great volume of mail and responses received from you.
 
$ continue with <standard disclaimer>
 
          --------------------------------------------------
          |                                                |
          |       Michael Webb                             |
          |       University of Michigan Physics Dept.     |
          |       1038 Randall Laboratory                  |
          |       Ann Arbor, MI  48109                     |
          |                                                |
          |       Michael_Webb@ub.cc.umich.edu             |
          |                                                |
          --------------------------------------------------

------------------------------

Date: Mon, 7 Aug 89 14:58:34 EST
From: munnari!munmurra.cs.mu.OZ.AU!jkjl@uunet.uu.net (John Keong-Jin Lim)
Subject: Moire-3.hqx01

Moire 3 is a screensaver for the Mac. Needs system 3.2 or later.
This release is shareware ($10), and the following new features have
been added :

	- Configurable sleep and wake rectangles.
	- Improved colour animation controls and better colour
	  selection and output.
	- Redesigned Control Panel interface.
	- New shape - Dyson's Folly (thanks to Scott Storkel)
	- Run on Shut Down option for those people who *never*
	  turn off their macintoshes (safe - parks your hd's first !)
	- Improved background performance, especially under MultiFinder.

This release consists of 3 files, Moire cdev, Moire docs and the 
cdev=>INIT converter to reduce Moire's size for those with low-end systems.

Thanks to all the people who put in suggestions. Sorry if i didnt have
time to put it in this version - wait for version 4 !

	John Lim

[Archived as /info-mac/cdev/moire-30.hqx; 88K]

------------------------------

Date: Tue,  8 Aug 89 02:53:25 -0400 (EDT)
From: "Andy A. Lee" <al1f+@andrew.cmu.edu>
Subject: Query 340MB MiniScribe

Hello All,

     In September issue of MacUser, Hard Drive International has 340MB
MiniScribe hard drives priced at $1795 (internal) or $1895 (external).
According to their advertisement, these drives come complete with UniMac(tm)
formatting & partitioning software and all necessary cables/mounting tools.
And these are 16ms drives!

     Before I jump on the phone, I would like to hear from those who has
purchased this package. More specifically, I would like to know if these
drives are quiet, and if it is possible to use other partitioning softwares
instead of UniMac(tm).

     I need a RELIABLE, true SCSI partitioning software that is capable of
password protection, and (perhaps) options to resize partitions on-the-fly.
I'll summarize replies... thanks!

* Andy *

+----------------------------+------------------------------------------------+
| Andy A. Lee                |  "We all choose to compromise                  |
| al1f+@andrew.cmu.edu       |   We kill our skills we robotize               |
| CompuServe: 72250,240      |   Acid waste draws kids and flies              |
| Carnegie Mellon University |   And diamonds turn to granite                 |
| Pittsburgh, Pennsylvania   |   Number One, money buys anything anytime..."  |
| (412)268-4932              |                       - Kim Carnes/Donna Weiss |
+----------------------------+------------------------------------------------+

------------------------------

Date: Tue, 8 Aug 89 09:48:08 EDT
From: Michael D. Prange <prange@erl.mit.edu>
Subject: Radius Accelerator

In v7#138 Rich Siegel noted that the most recent Radius SANE init
version is 1.4.  I looked in info-mac/util/radius-software-25.hqx and
found that it has version 1.2 of this init.  Could someone who has the
newest version please upload it?  Thanks.

Michael

------------------------------

Date: Tue, 08 Aug 89 11:17:05 EDT
From: Peter Jones <MAINT%UQAM.BITNET@forsythe.stanford.edu>
Subject: Searching/indexing files

Can anyone suggest some utilities that would allow searching and/or indexing of
files created by MS Word? I've looked at $MACARCH CONTENTS, a list of PDMAC
programs from LISTSERV@RICE, but the short descriptions aren't much help.

Peter Jones     MAINT@UQAM     (514)-987-3542
"All's well that ends." :-)

------------------------------

Date: 7 Aug 89 11:55:24 EDT (Mon)
From: lehi3b15!lafcol!buyskes@lxn.eds.com
Subject: Simple matrix manipulators?

	I'm looking for a fairly simple (and inexpensive) Macintosh
program to help my students learn matrix manipulations.  All I want them
to be able to do would be row operations, but with non-trivial matrices
(i.e., bigger than 3x3, not nice numbers).  I know about APL and some
fancy packages, but is there something simple that doesn't take much
time to learn?  I'm thinking of something like 1)enter a matrix,
2) select a row, 3) drag it onto another row, 4) enter the multiple
of the first row to be added to the second row, 5) and so on.
Any pointers?
 



Steve Buyske                    uucp    : rutgers!lehi3b15!lafcol!buyskes 
Mathematics Department          Bitnet  : BUYSKES@LAFAYETT
Lafayette College             
Easton, PA  18042

------------------------------

Date: Mon, 7 Aug 89 13:56:38 EDT
From: Bob_Voelker@ub.cc.umich.edu
Subject: Wanted:Comments on Macintosh Programming Primer.

     Is anyone familiar with "Macintosh Programming Primer--Inside the
Toolbox Using THINK's Lightspeed C," being offered at a discount by
Symantec to those upgrading to THINK C version 4.0?
     *  Does the book apply to version 3.0 or 4.0 of THINK C?
     *  Will the info in the book become obsolete when System 7.0 comes out?
     *  Any other comments?
     Thanks,
     Bob Voelker

------------------------------

Date: Tue, 8 Aug 89 15:44
From: <HANEWINK%DM0MPB51.BITNET@forsythe.stanford.edu> (Herbert Hanewinkel, MPI Biochemie)
Subject: writing software for a FDHD drive

Hi,
I would like to read floppies on my new IIcx in MFM mode, but the floppies
are not in standard MS-DOS format. So I can't use the file exchange program.

Can someone tell me how to switch the floppy drive between GCR (MAC mode)
and MFM (IBM mode). A piece of C code as an example would be great.


Thanks

Herbert Hanewinkel                        hanewink@dm0mpb51.bitnet
Max-Planck-Institut of Biochemistry
D-8033 Martinsried, FRG

------------------------------

End of Info-Mac Digest
******************************