Text content (OCR)
A2-Central
2
A journal and exchange of Apple II discoveries
Open-Apple
iF,
August 1990
Vol.6, No.7
ISSN 0885-4017
newstand price: $2.50
photocopy charge per page: $0.15
Apple speaks (and listens)
How visible is the Apple II community's dissatisfaction with Apple?
Well, David Reed’s report on the National Apple Users Group Confer-
ence in the July issue of Byte magazine (“What's New” Midwest region-
al supplement) mentioned the visibility of the “Apple II Infinitum”
campaign at the Chicago meeting, and Apple II Product Manager Rajiv
Mehta’s attempts to confirm the viability of the Apple II during his
speech at the event. Apple apparently feels the outcry of Apple II
users reflects an image of the company that they cannot accept, and
(as reported in “Miscellanea” last month) has assigned seven-year
Apple veteran Ralph Russo a highly-placed position within Apple to
evaluate and respond to the problem.
During June 15-17, Jim Merritt (leader of Apple’s Apple II “Green
Software” group responsible for applications, utilities, and develop-
ment environments) attended Apple Fiesta, a regional Apple II and
Macintosh conference hosted by the AZ-Apple user group in Phoenix,
Arizona. Merritt was “between fires” at Apple and was asked by Ralph
Russo to combine the trip to Apple Fiesta with a short vacation. He
brought a notebook and spent much of his time wandering the con-
ference and gathering comments and suggestions regarding Apple
products, primarily the Apple II product line, as well as giving two
talks regarding future directions at Apple.
Merritt portrayed the Russo organization as a “potent but small
core” within Apple dedicated to cutting through Apple’s bureaucratic
net and finding ways of getting the most out of the Apple II product
line. The group was described as being “small, but not too small”;
large enough to get the job done, but not so large as to have bureau-
cratic problems of its own.
Apple is working on more hardware and software for the
Apple Il as well as the Macintosh. Among the primary technology
directions Apple is concerned about are “media integration” (Merritt
used this more descriptive term in place of “multimedia’), data
exchange between the two dominant Apple product lines, and univer-
sal access regardless of the dominant sensory mode of the user.
Media integration consists of the ability to use the computer with
other devices meant to convey information by various methods: hard
disks, video monitors, CD-ROM drives, laser disc systems, and so on
(see “Picture This”, July 1990).
Data exchange will focus on the ability to move data files between
the Apple II and Mac platforms so that integrating the two types of
systems within an environment will not cause an unnecessary burden
on the user. The audience voiced its concern that such exchanges be
facilitated in both directions, unlike some past attempts (such as the
third-party Works-to-Works Transporter utility for Apple File Exchange).
Apple has managed to converge the line of peripherals for both sys-
tems over the past few years so that the Apple II can work with the
same 3.5 drives, SCSI hard disks, printers, and so forth as the Macin-
tosh, which will help in making the systems more compatible when
combined in a single environment.
Universal access refers to the broadening of Apple's “Human Inter-
face” design to allow computer users who may not favor (or be able
to use efficiently) specific input/output methods to have alternatives
available. One example is the Apple Iigs Video Keyboard v.1.0B1 (cur-
rently a beta product available to APDA members for $20, part
#A0028LL/A) a New Desk Accessory which provides an on-screen key-
board accessible via the mouse (or other compatible pointing device).
All of these directions seem to be logical choices for Apple if
it is to solve the problem of integrating Macintosh and Apple II sys-
tems among its customers. And it is unnecessary to be paranoid
about Apple’s motives; you can read them by tallying the number of
systems “in place” in Apple’s important K-12 education market. Using
rough figures based on a survey published in the October 1989 issue
of Electronic Learning, of the 59% of the Apple computers in K-12
institutions (a figure close to the 60% figure Apple has used), 97% are
Apple II systems. If Apple wants to sell Macs to this market, it had bet-
ter demonstrate that it isn’t declaring its current systems obsolete
after only a few years use.
About the Apple Ilgs specifically, Merritt stated that he felt its sys-
tem software represented the best built-in “toolbox” for developers
(‘in my humble opinion”) of any personal computer. Regarding the 8-
bit Apple Il systems, Apple is looking into ways to bring the desktop
metaphor to those platforms.
Apple is also investigating the codification of their text (MouseText
and file card) interfaces for Apple II systems so a more consistent
interface (across applications) for text applications can be document-
ed in the same manner as the desktop metaphor is for applications
using the graphics display.
Merritt also affirmed that Apple is indeed looking into new Apple II
CPUs, though he cautioned that the decision of bringing a new CPU to
market depended on more than just technological issues.
In his opening remarks in one speech, Merritt emphasized that the
company very much wants to get back the advantage that it had when
people bought Apple computers partially because of a “grassroots”
feeling that the company and its products were accessible. Apple
wants the feeling of using Apple products to be a positive and enjoy-
able experience from the time the computer is taken from the box.
Apple realizes many Apple Il (and even Macintosh) users may not
have felt this way in recent years. To help correct this, Apple wants to
improve user group and developer relations, including an elimination
of the schism that seems to exist between Macintosh and Apple II
WHILE SEEKING HER PC-BASED RECIPE INDEX, LORRETTA IMDDVERTENTLY LOADS A CADICAM
PROGRAM. INSTEAD CF WAKING CHERRIES JUBILEE, SHE BUILDS A SUBOCEANIC DVING PROBE
6.50 A2-Central
factions. Also, extending John Sculley’s comments about the work on
lowering the cost of Macintosh systems, Apple is investigating ways to
lower the cost of many of its products, including peripherals and
Apple Il systems. Merritt touched upon the announcement of the
lower cost Macintosh systems in response to perceived markets, and
explained that part of the function of the Russo organization is to
insure that Apple II alternatives are not overlooked. The intention is
that Apple should be concerned with user satisfaction with their prod-
ucts regardless of the platform used.
It’s too early yet to perceive what affect the Russo organization will
have on Apple’s real-life actions. But the presence of representatives
at the regional Apple II conferences indicates that Apple has heard
the complaints about its past policies, and is now looking to gather
reasonable suggestions for improving things. When Jim Merritt gave
his closing speech, his notebook was close to full.
One very strong positive response to criticism is demonstrated
by Apple's announcement of changes (effective June 25th, 1990) to
the Apple Partners and Associates developer support programs. With
the new changes, Apple Associates will receive invitations to the
annual Apple Worldwide Developer's Conference, and will also be eli-
gible for a limited number of discounted hardware purchases to ease
their development efforts. Formerly, the invitations and hardware dis-
counts were limited to Apple Partners only.
In addition, a often-requested option of Apple Il-only program pric-
ing has been introduced; you can become an Apple Il-only Apple Part-
ner for $300 a year (Apple Il and Macintosh price remains $600) and
an Apple Associate for $150 (Apple II and Macintosh remains $350).
This brings the cost of these programs back into range for many
smaller Apple II developers, many of which had complained the origi-
nal prices were disproportionately high given the dominance of Macin-
tosh specific information supplied as part of the program. (For more
information about the programs, you can call the Apple Developer
Hotline at 408-974-4897.)—DJD
Apple acquires Claris
Possibly in line with some of it’s stated aims, Apple has decided to
acquire Claris Corporation as a wholly-owned subsidiary. Claris had
existed as a majority-owned subsidiary of Apple since it was “spun
off’ by Apple in 1987 (see “Ask Uncle DOS", June 1987, p. 3.37);
Apple has chosen to bypass the option of allowing Claris to become a
separate publicly-owned company. The current CEO of Claris, William
V. Campbell, will remain but now reports to John Sculley.
According to John Sculley, Claris will focus on software develop-
ment and marketing in three areas: continuation of current “flagship
applications” (is it absurd to assume the dominating integrated appli-
cations AppleWorks and AppleWorks GS will be part of this group?),
extensions to Apple’s system software to enrich the Apple software
development environment and “promote the creation of innovative
Macintosh applications”, and products to “bridge Macintosh with other
computing environments’. Consideration will be given to licensing
technologies Claris may develop to software developers for incorpora-
tion in new applications.
It’s interesting to consider why Apple may have made this move.
Part of the original justification for Claris’s formation was to reduce
any perception of a “head to head” competition between Apple and
third party application developers (Apple, of course, retains propri-
etary control of its system software); retaining ownership of Claris
seems to negate that intent of the “spin-off”. There is the possibility
that Apple feared an independent Claris might start developing ver-
sions of its major applications for Microsoft's graphics-based operat-
ing system for IBM PC compatibles, Windows 5.0. Is Apple positioning
to reduce its dependence on Microsoft as a source of application soft-
ware for the Mac, or is Apple trying to “rein in” Claris, or was Apple
fearful of the possible acquisition of a “public” Claris by a competi-
tor?
Again demonstrating mixed signals on its commitment to the Apple
Il, Apple’s press release regarding Claris gave a good example of the
work the Russo organization has cut out for it. In the three items
delineating Claris’s part in Apple's strategy, only the Macintosh was
explicitly mentioned by name (twice). It is just such omissions (even
assuming it was an oversight) that allows the persistence of specula-
Vol. 6, No. 7
tion regarding the future of the Apple II, because the Apple Il is not
overtly mentioned as is repeatedly allowed for the Mac.
It’s probably going to be the job of the Russo organization to point
out such inequities to Apple’s top management. If Apple’s job is to
sell computers (and it must be assumed that is what it’s all about),
then they should insure ail their literature reflects optimum promotion
of both Apple Il and Mac platforms.
Apple claims to be interested in the Apple II customer. Making sure
the Apple II is mentioned in comments involving future strategy would
be a simple place to start demonstrating commitment to that interest.
That has been accomplished verbally in recent weeks, but Apple's
message still comes across differently in print.—DJD
Object Module Manager
The inherent problem with 8-bit software on the Apple II has always
been the lack of resource management. To be fair, this was a limita-
tion in all of the popular 8-bit personal computers, but only the Apple
Il has survived long enough for the problem to become painfully obvi-
ous versus its present-day competitors. (GS/OS applications using the
ToolBox on the IIgs do not have this problem; they have a full-fea-
tured Memory Manager and other protocols available to arbitrate
things.)
The problem is most apparent in Applesoft, where short machine-
language programs are often used to enhance the language. The eter-
nal question is where to put such routines in the Apple Il memory so
that they stay safe.
Roger Wagner Publishing's ToolBox series for Applesoft takes one
approach: it attaches the routines to the parent Applesoft program
itself so that whenever the program is loaded, the routines come with
it. But what about a program where it is more desirable to have a
series of installed enhancements that can be accessed by several
modular “subprograms”?
Tom Weishaar (with some help from Mark Simonsen of Beagle
Bros) described a protocol of adding routines to ProDOS BASIC using
the EXTRNCMD vector at $BE06 in the BASIC.System global page.
EXTRNCMD can be altered to point to a user-installed external com-
mand which executes as if it were a BASIC.System command. Tom
used this to implement a ProDOS “TYPE” command that types the
contents of a file (usually a text file) as displayable ASCII characters;
the program was presented in a Softalk magazine article and later
incorporated as Chapter 20 of ProDOS Inside and Out. (Incidentally,
the listing in the book ended up missing a couple of lines thanks to
yours truly; you can write me at A2-Central to request the corrections
before you try entering the program.)
The larger problem the protocol did not solve was the management
of multiple routines in memory. Removing a routine (other than the
last one added) and recovering the memory it used was not feasible
without a more complicated “master routine” that could recognize
when it was “safe” to add or remove a routine, and then identify and
relocate the remaining routines. The Morgan Davis Group has now
provided such a utility, the Object Module Manager ($24.95 from The
Morgan Davis Group, 10079 Nuerto Lane, Rancho San Diego, Calif.
92078-1736, 619-670-0563 or FAX 619-670-9643).
The centerpiece of the Object Module Manager is the Loader,
which manages loading the modules from disk (relocating memory
resident modules to close any gaps in memory, if need be), communi-
cation between the modules, and purging of modules no longer need-
ed. The Loader is installed by executing the system file OMM.LOADER;
if the Loader is already installed, a duplicate installation will be abort-
ed with a “DUPLICATE FILE” error.
The Loader adds seven commands to BASIC through Applesoft’s
ampersand command “hook”. Each command is of the form
“& <command> <list>” where “<command>” is the actual Object
Module Manager command name and “<list>” is data that can be
passed to the command.
The LOAD GET command accepts one or a series (separated by
commas) of module names and loads them from disk:
August 1990
& LOAD GET “nodulel”
would load the module named “module!” from disk, and
& LOAD GET “nodulel”, "nodule?"
would load “module1” and “module2”. The Loader includes a “trace”
mode (enabled as a default) that will display information regarding
each module file as it is loaded: an index number, an ID number,
default address (for relocation calculations), length, date, title, and
version. The ID numbers for each module must be distinct; attempt-
ing to load a module with the same ID as a previously loaded module
will abort with a “DUPLICATE FILE” error.)
LOAD FRE accepts a module index number or ID number and
removes the associated module from memory, relocating other mod-
ules to fill any gaps in memory use. With trace mode on, a report on
the “freed” module will be displayed as it is purged. The index num-
ber is distinguished as a value that will always be 255 (the maximum
number of memory resident modules allowed) or less, while the ID
will always be 256 or greater. The Object Module Manager Loader
itself will always have an index number of 0, and
& LOAD FRE 0
will therefore unload the Loader and all modules.
LOAD TRACE and LOAD NOTRACE enable and disable the “trace”
mode respectively. LOAD PRINT prints the “trace” information for all
installed modules. LOAD PEEK allows a program to determine infor-
mation about loaded modules:
& LOAD PEEK 0,N
will return the count of available modules into the numeric variable N,
and
& LOAD PEEK 1,M,A$
returns the “trace” information for the module with the ID (or index)
number contained in the variable “M” into the string variable A$.
The interface to each module's commands is through the
LOAD CALL command. LOAD CALL requires three pieces of data:
the ID (or index) number for the desired module, a message code,
and a function number. There also may be extra data items to pass to
the called function; these can be added after the three required
items.
Object Module Manager requires that the modules it uses be relo-
catable; that is, the modules must be constructed in such a fashion
as to allow the OMM.LOADER to move them about in memory. To this
end, each module must be of type REL (relocatable object file) and
designed with a specific header for recognition and manipulation by
the Loader. To allow OMM.LOADER's relocator to operate, any object
code that might require modification during relocation must be orga-
nized and identified as a contiguous block, and any data identified as
a separate block, rather than intermixing the two (the data block can
contain 65xxx instructions as long as no absolute address references
to points within the module are used).
Modules may send messages to each other; by this method a mod-
ule can ask another module to execute a task needed for the originat-
ing module to complete a function. By using messages, the “sending”
module does not have to know details about the internal structure of
the “receiving” module. Each module must respect a minimum set of
commands; the big three are MSG_INIT (initialize the module),
MSG_INFO (return module information string), and MSG_QUIT (shut
down module). If the module executes ampersand commands itself
(reasonable to assume if the module is to have a function), a
MSG_AMPR (execute ampersand service routine) command must also
be supported. In addition, the Object Module Manager itself may send
other messages: MSG_USER (intermodule service request), MSG_REL1
(alert module that it is about to be relocated), MSG_REL2 (alert mod-
ule that it has just been relocated), MSG_KILL (notification that a mod-
ule is about to be removed), MSG_DIED (notification that a module
has been removed), MSG_BORN (notification that a module has just
been installed), MSG_IDLE (idle event; not yet implemented). Most of
these are designed to allow modules to recognize when management
functions may affect them; interdependent modules will need these
whereas other modules may be able to ignore them.
Several service routines are built into the Object Module Manager
that a module may access: OMM_GETID returns the index of a module
based on a supplied ID number, OMM_XOAMP executes the “original”
A2-Central 6.51
ampersand vector (the “&” vector in effect before the Loader was
installed), OMM_FREE frees a module, OMM_PUTWORD stores a word
value to an Applesoft numeric variable, OMM_PUTSTR stores a string
to an Applesoft string variable, OMM_GETSTR evaluates the string
expression at the current position of the Applesoft text pointer and
returns its descriptor, OMM_PADDEC prints a word in decimal (right
justified), and OMM_C2PSTR copies a string into a Pascal format string
(with a leading “length” byte).
The package comes with ORCA/M and Merlin interface files contain-
ing “equates” for the Object Module Manager. You can use the
equates with any source code that you develop to create your own
modules. The use of assembly allows you to code for whatever pro-
cessor you have available (6502, 65C02, 65802, 65816, ...) but it will
be your responsibility to make sure the proper environment is pre-
sent.
Modules you create should allocate any buffer space you may need
internally; the GETBUFR/FREBUFR mechanism of BASIC.System is not
guaranteed to work between Object Module Manager service calls due
to the dynamic management of the modules. The communication
between modules also can’t take place by absolute reference since
the modules may be relocated when other modules are deleted.
There is a paucity of “standard” modules supplied; one sample
module with source code is included along with documentation on
disk. Apparently the product is aimed squarely at programmers who
are interested in converting existing routines to take advantage of the
Object Module Manager environment, or in developing new modules
for distribution for use use with the environment. An Object Module
Manager compatible adaptation of Morgan Davis's AmperWorks
(minus the & FRE command) is included for purchasers of Modem-
Works or MD-BASIC (AmperWorks is included with those products, but
not sold separately) who wish to use the AmperWorks commands in
the Object Module Manager environment. You will need to own one of
those products for the documentation of AmperWorks. (Although MD-
BASIC requires a llgs, notice that Object Module Manager and Amper-
Works do not.)
To distribute programs you create with Object Module Manager,
you must obtain a license from the Morgan Davis Group. For commer-
cial programs or a single-site license, the fee is a one-time payment of
$100.—DJD
Miscellanea
“Can the Machine Maim the Message?" That's the title of an Aca-
demic Computing (Jan. 1990) article by Marcia Peoples Halio that has
been generating some controversy lately.
In the June 29 issue of Personal Computing, a letter titled “Scared
by ‘McStudents’” from Joe Celko (pp. 15-16) refers to the article.
According to the letter, freshman students at the University of
Delaware (where Dr. Halio is the assistant director of the writing pro-
gram of the English department) were allowed to choose between
Macintosh and IBM PC lab sections for their composition classes, The
results were that students using the PCs wrote at an average grade
level of 12.1 on the Kincaid scale versus an average of 7.95 for the
students using the Macs. Several other quotes indicate other weak-
nesses in the performance of the students using Macs versus those
using the PCs, including simpler and less diverse sentence structure,
more spelling errors, and “fluffier” topics (rock music versus capital
punishment, for example).
Many Macintosh (and even IIgs “desktop”) afficianados I have men-
tioned this to have gone absolutely ballistic, refering to numerous
“studies” that demonstrate just the opposite. | am aware of one such
independent study (contracted with Diagnostic Research, Inc., by
Apple) entitled “Macintosh or MS-DOS: a synopsis of what MIS man-
agers and other professionals in Fortune 1000 companies have to
say” that came to the conclusions that (for the companies studied)
learning the Macintosh system was 34% easier than MS-DOS, and that
Mac users reported using one-and-a-half times the number of software
packages that MS-DOS users reported. Higher productivity figures
were also reported for Macintosh.
6.52 A2-Central
(Those who read Bruce Tognazzini’s “The Human Interface” col-
umn might expect me to also use some of his figures; alas I cannot.
Although his January 1989 column asserts his figures are based on
“scientific conducted” studies, his refusal to discuss Apple's testing
adequately to allow the basic scientific principle of independent rep-
etition and evaluation of the experimental method forces me to con-
sider his assertions with the same skepticism as | do “flying saucers”)
Until | see the original article (an interlibrary loan is being facilitat-
ed as | write this), I tend to believe that comments from “Gloom and
Doom: Computers and Composition” in the June 1990 mini‘app‘les
Newsletter (The Minnesota Apple Computer User's Group, Inc., Box
796, Hopkins, Maine 55343) are insightful. The author, Emmeline
Gragerford, looks at Dr. Halio’s article and offers some theories about
the “fluffy Mac phenomenon”.
Several of the observations reflect what | think is the most impor-
tant key to interpreting the results: the results are influenced by the
temperment of the student's judgement which resulted in their selec-
tion of a CPU, Students who select the Mac primarily for its visual
interface appeal or ease of use versus the “user hostile” image of MS-
DOS are already indicating a preference for a certain work environ-
ment (and possibly work ethic?). That theory could be tested by forc-
ing students from each group to work with the “other” computer and
see if the differences correspond to the individuai students rather
than the CPUs (descriptions of the study imply that the same students
continued to use the same CPU). If the difference is the students, it
would explain the differences from corporate studies, where the envi-
ronment is likely to sort out work habits in short order.
Another factor Gragerford discusses is the propensity to play with
aspects of the Mac display’s flexibility, such as in trying various com-
binations of fonts. This resembles the “Computer Perseveration Syn-
drome” (A2-Central, April 1989, p. 5.17) where completed projects
are often modified endlessly because the computer makes it easy. In
truth, content must have priority over (visual) style when writing, but
it's easy to forget when you know there is a particulary expressive
font in the machine somewhere.
It's certainly not fair to blame the Mac for the work habits of its
users, the survey results seem to be interpreted as meaning the Mac
is not as satisfactory a tool as the PC in some cases. (This interpreta-
tion applies to other programs which use a graphics-based interface
such as applications for GS/OS on the IIgs, Windows 3.0 for MS-DOS,
the Commodore Amiga, and others.) It is more reasonable to blame
the software tool in use; possibly writing tools used for composition
Vol. 6, No. 7
courses should be confined to more basic functions (and a minimum
of distracting features), or the students should be advised how to
concentrate on the use of the program as a writing tool.
The “computer as writing tool” message was driven home to me in
a panel at the 1990 Chicago WindyCon science fiction convention.
Editors advised prospective writers to throw their fancy fonts and
graphics in a trash can when writing, because Real Editors want plain,
easy-to-read type when evaluating the content of a story. In many
cases, such as teleplays, there is a specific form expected, and sub-
missions that don't conform may go unread, The fancy stuff can be
saved until later; that's why we still write the newsletter using the
Apple II text screen, and treat typesetting as a separate project.
There have been more changes at the top levels of Apple. On
June 28 Lotus Development Corporation announced that Donald P.
Casey would become vice president of their spreadsheet division
effective July 15. If you don’t have your Apple (re-)organization chart
handy, Casey occupied the position of vice president, Networking and
Engineering; the link between Ralph Russo and John Sculley.
Effective July 1, Robert Puette is the new president of Apple‘s
Apple USA division, responsible for the division’s sales, marketing,
and support functions and channels. Puette came to Apple after 24
years at Hewlett-Packard Company where he led Hewlett-Packard’s
start-up personal computer business. He arrives as an individual with
proven management experience.
Cirtech has announced Duet, a coprocessor that will allow
the Iigs to run Macintosh programs. The card fits into any standard
slot on the IIgs and contains a 68020 microprocessor with up to 8
megabytes of RAM and a custom ROM. The system recognizes stan-
dard Apple peripherals using the 65816 for input/output processing,
Cirtech claims the system will outperform a Mac Ilcx.
A 68882 math coprocessor can be added as an option to a socket
on the Duet card. The card comes standard with one megabyte of
RAM, and Cirtech anticipates having the card ready to ship by Decem-
ber 1990. No price has been determined yet.
By the way, Cirtech has a new address: Cirtech (UK) Ltd., Monks-
ford Stables, Newtown St. Boswell's, MELROSE TD6 ORU, Scotland,
0835-23898 or FAX 0835-22471.
Shrinkit for the igs, a desktop-based version of the standard
Apple Il archiving program, has been completed by Andy Nicholas
and is in distribution via national and local information services. This
Ask
This is the “be humble and admit your mis-
takes” issue...
Before I wrote the answer to “Bad Driv-
ing?” (June 1990, pp. 6.38-39), I had received
three separate reports of problems with 5.25
drives within two weeks, and had seen an
anomaly with the AMR 3.5 drive (different
behavior on the ROM OI and ROM 03 lligs)
myself. However, after my solicitation in “Ask
Uncle-DOS” in the last issue there have been
only two further reports of problems with 5.25
drives, and neither follows the description
given for the earlier glitches. At this point, 1
have to assume Apple's changes to the drive
port don’t represent a significant compatibility
problem for 5.25 drives. I apologize for indi-
cating Apple could have foreseen the few prob-
lems that did arise, since the incidents seem
to be unusual.
In “Wipe it clean” (p. 6.46), the BASIC exam-
ples used to check the /RAM volume for files
are incorrect; not only will the BLOAD com-
mand gag on /RAM's shortened directory with
an “END OF DATA’ error, I got the pointer to
the file count wrong.
The following corrected line will prevent the
“END OF DATA” error by limiting the number of
bytes read from the /RAM volume directory
“file”:
PRINT CHRS (4);"BLOAD /RAM, 32000, SHS, L$100"
and this is the corrected line to verify the num-
ber of files:
‘IP (PEEK(8192+37)) + 256 * PEEK(8192+38)) THEN
PRINT “/RAM NOT EMPTY!”
One of the items overlooked in the answer
to “Don’t touch that memory’ (p. 6.46) is that
ProDOS Technical Note #26 does specifically
mention that if /RAM is enabled, all space
above $800 in auxiliary memory may be used
by an application after /RAM is removed (and
/RAM should be re-installed upon completion).
If /RAM is not enabled, then memory above
$800 can be used at the programmer's discre-
tion, but the areas indicated as “reserved” in
the Apple documentation must be left alone.
Application programmers should realize that if
their application finds /RAM is disconnected,
someone may already have used some of the
auxiliary memory; if they happen to overwrite a
driver that is supposed to continue to reside in
auxiliary memory for use by several applica-
tions, disaster may result. What is needed is
some type of “globally respected” memory
management protocol for auxiliary memory.
In contradiction to the reply to “BASIC net-
working’, there is a way to access the possible
“illegal” filenames of an AppleShare volume
from ProDOS 8...sort of. Matt Deatherage at
Apple Developer Technical Support pointed
me at the right section of the AppleShare Pro-
grammer’s Guide for the Apple II Family,
Beta Draft (available to APDA members for
$50, part #A2G0051/A), which has supplanted
the AppleShare Programmer's Guide for the
Apple ligs.
When you are using ProDOS with Apple-
Share installed, many commands to interface
with the AppleTalk protocols are made avail-
able through a new MLI command ($42). One
August 1990
IIgs version will work with the GS/OS extended file structure when
creating and extracting file archives in the Apple I] NuFX format. The
IIgs version of Shrinkit will extract files from Binary Il “envelopes”
(even those “squeezed” with Floyd Zink’s Binary Library Utility) as well
as America Online’s ACU format. It will also create and extract
AppleSingle files.
An exciting new capability is the extraction of files from several
non-Apple II formats such as Stuffit (Macintosh), ZOO and ARC (MS-
DOS, Amiga, and Atari ST), and compress (unix). Users can employ
this feature to extract and convert graphics with the aid of utilities
such as SHRConvert ($15 from Jason Harper, 1480 Michelle Ct. #A,
Colorado Springs, Colo. 80916); programmers will be able to use it to
access public source code from foreign environments for conversion
and use on the Apple Il.
Shrinkit for the ligs requires 768K of memory and System 5.0.2 or
greater. It’s free, but if you find it useful, Andy Nicholas would proba-
bly appreciate a contribution. His address is 8415 Thornberry Drive
East, Upper Marlboro, Md. 20772; electronic addresses are SHRINKIT
for GEnie, Shrinklt for America-Online, and 70771,2615 for Com-
puServe.
Proterm v2.2 is available from a new company, InSync. The
new version has been updated from version 2.1 (which was distribut-
ed by Checkmate Technology) to include several minor changes,
including modification of the YMODEM and ZMODEM protocols to
allow their reliable use with the GEnie network, a fix for a file transfer
“lock-up” problem when switching the program disk with a data disk,
full use of the Apple IIgs Memory Manager for memory allocation on
the IIgs, an expanded (from 28 to 36 systems) dialing list, enhance-
ment of the file selection system to handle up to 250 files in a direc-
tory, 19200 baud support in the HST modem driver, addition of a
“defaults” window for entering commonly used settings, selection
between the logon “siren” or the standard Apple II bell signal (or
silence), and a tune-up of the VT-100 emulation.
Owners of previous versions can receive a disk upgraded to ver-
sion 2.2 by sending their original diskette plus $10 in cash or check
to InSync Software, PO Box 22146, Phoenix, Ariz. 85028, 602-992-
5515 (sales) or 602-992-1345 (technical support). Or for $35 you can
receive the entire new package including manual, reference cards,
and disks. InSync says they did not receive a complete list of regis-
tered users of Proterm, so they're asking your help in spreading the
word about the upgrade.
Optical character recognition (OCR) is coming to the Apple
A2-Central 6.53
fle and Ilgs. Beagle Bros programmers Alan Bird and Rob Renstrom
have announced their OCR software, INWORDS, scheduled for release
in September 1990 through a new company, WestCode Software.
OCR software allows the use of scanning hardware to process char-
acters on a document and convert them into a text format for use in
programs that deal with textual information. This allows, for example,
a printed document to be scanned and processed to produce a word
processor file containing the text portion of the document.
INWORDS will scan up to 3000 characters per minute and will
establish the feasibility of OCR for both the Ile and llgs platforms.
John Pothier of Vitesse passed the information to us and revealed
that the software had been tested on Vitesse’s Quickie scanner, but
that the support of other scanning systems was open to WestCode.
For more information contact WestCode Software, 11835 Carmel
Mountain Road, Suite 1304-311, San Diego, Calif. 92128, 619-679-
9200.
Apple's rework program for the ImageWriter LQ ends October
31, 1990. If you have bad line registration or excessive noise, you
need to see your dealer before that expiration date to get the pre-
scribed Apple remedies free of charge.
APDA has announced several new Apple Il products. For those
interested in developing software using the new capabilities of the
DMA SCSI card, the Apple II High-Speed SCSI Card Utilities ($30; part
#A0242LL/A) are now available. Disk #1 contains utilities that ship
with the card; this disk can be licensed through
‘Apple Computer Software Licensing
20525 Mariani Avenue M/S 38-1
Cupertino, Calif. 95014
Disk #2 contains beta versions of device drivers for the Apple
Scanner and Apple Tape Backup 40SC (for development of applica-
tions using those devices).
In addition, final versions of the Apple Iigs Tools and Interfaces are
available; these include the 5.0.2 versions of the interface files for
APW C and the APW assembler as well as updated utilities for dealing
resources. The full package is $50, #AO240L/A; the update from the
previous release is $25, #AO241L/A.
Also available is the Addison-Wesley edition of the Apple Iigs Tool-
box Reference Volume 3, (#A0299LL/B; $39.95) which can also be
ordered from Addison-Wesley or <ahem> us.—DJD
of the extensions is called the ProDOS Filing
Interface (PFI), and it includes an FiNaming call
to allow you to tell ProDOS to accept the
extended naming conventions of AppleShare
volumes, There was one exception versus use
of the GS/OS AppleShare FST: I couldn't get
the use of the *:” (versus “/”) separator to work
for pathnames (I kept getting a $53 error from
ORG $300 location for code
Then we issue a few MLI AppleTalk com-
mands using our parameter list for the FiNam-
ing call. To enable the long naming conven-
tions, we first get and save the current naming
protocol (so that we can restore it when we fin-
ish) and then set the long naming conventions:
the MLI telling me the pathname was invalid). Set¥og_— START _____ Call this to enable
Here’s a sample using the call to set the prefix USING Flaming long naning
to */SV/Users/<Any User>” (the user folder on ma ‘sed to request
our server) and to delete the file “Illegal Name” STR Direct? current settings
from within that folder. Hang on; it's a biggie. JR MI
The assembly language program consists of ue Baa!
four subroutines: SetLong, which gets the cur- DC AF INaniing’
rent “long filename” status and saves it, then STA —ErrorRtn —check error status
enables the use of the extended pathname BCS) ErrExit
syntax; FixLong, which restores the initial sta- WDA NanConvf save current.
tus; SetPfx, which uses the ProDOS MLI to set STA NamConvS «settings
a prefix; and Destroy, which uses the MLI to
delete the named file. Assuming you have the 2 ee
ProDOS 8 Technical Reference (and you STA NanConv?
should have it and the AppleShare Program- BR OMI
mer’s Guide if you are going to use these i wa
commands) most of the source should be a ‘2/FINaning!
straightforward. First, we define a couple of a aa
constants: Bemit ae
PéxBufr QU. $0280 usually systen path RS
MI cu ($ard0 BD
We call FixLong when we exit to restore the
original settings for the long naming conven-
tion:
Fixlong START
USING FINaning
IDA NanConvS restore saved
STR anon? configuration
wa SCO enable mask
ST Direct?
OR MI restore settings
ce Rw
De A2'FINaning’
STA rrorktn
ae
ies
2D
Next we have our data area for the various
values. The AppleTalk parameter list consists
of a byte identifying our call as asynchronous,
the value $33 to identify the call request as
FiNaming, a word (two bytes) to hold any
AppleTalk error returned (an AppleTalk error
will be reflected in the error returned for the
MLI call; our program only checks for the pres-
ence of an MLI error), and two bytes specific to
the FiNaming call. One is the Naming Conven-
tion Flag (NamConvF); bit 7 of this flag tells
6.54 A2-Central
ProDOS whether you want the ProDOS Device
Table disabled (if 0, the table is enabled; if 1,
the Device Table will not be updated as net-
work volumes are mounted or unmounted),
and bit 6 specifies whether the AppleTalk Filing
Protocol Long Naming Conventions are used
(Long Naming used if 1; ProDOS naming used
if 0). The other is the Directional Flag (DirectF);
bits 6 and 7 of this byte determine whether
you set the Naming Convention Flag items as
‘specified from your parameter list, or whether
you obtain the current settings (return current
mode if 0; set new mode if 1). Following these
bytes, we have two bytes for storing the FiNam-
ing protocols found, and for any MLI error
returned:
FiMening DATA AppleTalk pam List
HOD’ async byte
we a3" ‘FiNaming command
DC 12'$0000" —Appletalk result
Direct? we #00" Directional flag
NanCoow? = 0C—s#H'00 Waning convention
Nantows COO!
Errortn «DC #H'00!
BOD
Finally, we have our “normal” MLI calls that
we need to call from Applesoft in order to deal
with the “long” names. SetPfx will set the Pro-
DOS 8 system prefix to the pathname stored at
$280:
setDix START
USING MPams
IR MI
ORCC set prefix call
DC -Al/prefix’ parameter pointer
SA MUlError
ac
Ras
2D
Destroy will delete the specified file:
Destroy START
USDC MLPams
mR OMI
oe acl’ destroy call
DC Ai'perefix’ paraneter pointer
ST Mirror
ac
Ris
BD
Both MLI calls use a common parameter list:
MiPams DATA
prefix CHO’, pam comt
UC AL’ PfxBufr’_prefix/path pointer
Murer DCEO" for errors
BD
The following Applesoft program uses the
assembly language routines (placed in the file
P8LONG) to switch ProDOS to the long naming
conventions, set the prefix to the user folder,
delete the “Illegal Name’ file, and then switch
back to the original ProDOS settings. It’s crude,
but it should at least give you an idea how the
delete can be done (of course, you do have to
have AppleShare installed or an error will
result). Notice that we POKE the pathnames
into the ProDOS system path buffer at $280
(640 decimal) because BASIC.System's path-
name parser would gag on them:
1000 REM == delete AppleStare file =
1010 0§ = CaRS (4)
1020 PRINT DS;"BLORD POLONG": REM at $300
1030 SL = 768:FL = 809: REM SetLong, Fixlong
‘1040 ER = 838: REM MLI error return
1050 REM — set MLI to use long naning —
1060 CALL SL: IF PEEK (ER) THEN PRINT "MLI
‘Error: "; PEEK (ER): SIP
1070 GONO 2010: REM >>> do your stuff in here <<
1080 REM — set “long name” use back to original —
1090 CALL FL: IP PEEK (ER) THEN PRINT "ML
Error: °; PEEK (ER): STOP
1100 BD
2000 REM — test set prefix and destroy —
2010 GOSUB 10000: PRINT "Error return: “jE: STOP
2020 GOSUB 11000: PRINT “Error return: “;E: STOP
2030 GoTo 1090
9999 :
10000 REM — to set prefix with our code —
10010 SP = 839: REM set prefix entry
10015 EC = 864; REM MLI error return code
10020 PEx$ = "/Sv/Users/<Any User>"; REM our test.
prefix
10030 PORE 640, LEN (PFAS): REM set length
10035 REM next line puts prefix in buffer
10040 FOR I= 170. LEN (PEXS): PORE 640 + I,
ASC ( MODS (PEKS,1,1)): NEE T
10050 CALL SPB = PEEK (EC)
10060 RETURN
11000 RIM — to destroy file with our code —
T1010 DS = 850: REM destroy entry
11020 EC = 864: REM MUI error return code
11030 PFXS = "/SV/Users/<Any User>/Illegal Name":
ROM our test file
11040 PORE 640, LN (PFXS): REM set length
11050 ROM next Line puts prefix in buffer
11060 FOR I= 110 LEN (PFXS): PORE 640 + 1,
ASC ( MIDS (PEXS,1,1)): NEXT I
11070 CALL DS:E = PEEK (BC)
11080 RETURN
This particular program will print the error
result and stop after the set_prefix (line 2010)
and destroy (line 2020) operations. If the error
return is “0”, you can type “CONT” to tell
Applesoft to continue with the next line. You'll
want better error handling in a real-world pro-
gram.
All this assumes you have access privileges
to delete the file (and obviously there's no
easy way to defeat the access privileges). And
it’s still a lot more work than using a GS/OS
“delete” command which can handle the
AppleShare FST’s naming conventions.
In response to the request for computer
travel cases (“Carry me away”, May 1990, pp.
6.30.51), Orbit Systems, 914 Searcy Way,
Bowling Green, Ky. 42103, 502-782-0600 or
800-541-5421 (sales) let us know that they
manufacturer cases for the transit of comput-
ers and other electronics.—DJD
Remotely interested
I read your neat article on “multimedia” (“Pic-
ture This”, July 1990); there are a couple of
things I'd like to discuss,
First, about the VidClip product you men-
tioned: 1 had contacted these folks some time
ago after reading an article about VidClip in
VideoMaker magazine. | wrote to them and they
kindly sent me their instruction manual. The
manual revealed the programs only worked with
Sony equipment capable of receiving CTRL-L
hex codes. It did not apply to any other make of
VHS VCR or camcorder.
Second, there is the subject I'm really inter-
ested in, which wasn’t talked about in your arti-
cle. (No gripe intended. I'm happy to see the
subject discussed at all.) A multimedia area cry-
Vol. 6, No. 7
ing for help that 1 would like to see the Iigs
aimed at is video tape post-production editing
and assembly for home use. | am positive the
IIgs can do this easily with some simple infrared
transducing hardware, I’m certain there are a lot
of ligs people like me out there with VHS VCR’s
and camcorders who would buy whatever it
took to give the IIgs the capability to control
their camcorders and VCRs via infrared codes
or directly through DIN plugs. This would give
us the control capability of our VHS camcorders
and VCR's to do the entire editing and re-assem-
bly tasks on our home videos via the overlay
card. It would also include the necessary gen-
lock ability to add animation, titling, transitions,
etc.
With this kind of capability you could even do
a lot more than patching together home videos
of little Willie, You could cover the entire educa-
tional and presentation gamut.
Would it be so impossible to have the Ilgs
control VHS VCR's in the same way you have
described its ability to control laser discs? Com-
puter control of VCRs is nothing new. Most of
the more expensive VCR's are now coming out
with editing capability. And of course there is
the (sigh...) entire Amiga line.
Also, maybe you've heard of the Direct-ED
machine, which is nothing but a dedicated com-
puter in a small box using an infrared transduc-
er which ‘learns’ a VCR’s command codes and
gives the capability to do post-production edit-
ing with graphics, titling, etc. The graphics
aren't all that good and it’s slow as you said
about VidClip. But so what? It does the job. In
fact, when the Direct-Ed Machine first came out
there was talk that they would produce a pack-
age composed of the transducer and software
for computers like the Apple II's to do the same
thing. | think the ligs with the Video Overlay
Card is capable of doing it better. The only road-
block | see is the lack of the infrared hardware
required to turn IIgs commands into tape VCR
remote commands. If we had software and this,
we'd be off and running. Why are these VCR
infrared remote codes so hard to find? Is there
anywhere a guy can find out what these mysteri-
ous VHS VCR Remote codes are?
Jack Schachtebeck
Roseville, Calif.
As mentioned, the SIA Ill does have a
“decode” mode that will work for SR-based
devices. The trouble is that there are many
remote control systems that aren’t SR-compati-
ble.
For a more generic remote control solution:
Jerry Kindall (8-bit editor of 8/16 and employ-
ee at Quality Computers) tipped us about a
“hacker project”:
“Dan DeMaggio (another technician at Quali-
ty) and I have been working on hardware and
software for interfacing infrared remotes with
the ligs. We’ve got the “digitizer” working,
which allows you to “record” infrared signals
into the IIgs and save them. The “transmitter”
will allow you to “play” the signals back from
the ligs and control any infrared-controllable
device you have lying around. With appropriate
software (like a HyperStudio XCMD) this little
piece of hardware could do some pretty
astounding things.
“We're also going to try to do some software
which will allow the IIgs to recognize remote
control signals. In other words, you could
August 1990
Program the ligs to treat your remote as a hex
keypad or as a macro keypad, or to just allow
you to advance through a slide show using the
remote.
“The gadget plugs into the game port and
requires about $10 worth of parts, all readily
available at Radio Shack. And while | said it’s
for the Ilgs, it should also work on the Il+ and
Ie.
“Im not sure when we'll be finished with
this, as we're really working on it in our spare
time, but we'll keep y“all posted.”
Dennis Ulm added a comment to the discus-
sion on GEnie with the address of the Direct-Ed
manufacturer: Videonics, Inc., 1370 Dell
Avenue, Campbell, Calif. 95008, 800-338-
EDIT.—DJD
Speaking of portables
Do you think we could convince anyone at
the big Apple to shove a Ilgs into the Mac
Portable case, slap in a 30 megabyte hard disk
and a SCSI port and sell it real cheap? I'd buy it.
Of course | would insist on SIMMs on the moth-
erboard expandable to 4 megabytes RAM.
Michael A. Ament
Antioch, Ill.
Sounds good here. The problem is that
using the Mac Portable’s high-performance LCD
screen and selling the result “real cheap” might
be mutually exclusive. But there may be room
under the Mac Portable’s price tag.
We have also had one reader suggest that
using a common case design for the llgs and
modular Mac systems (the Iicx and Ilci) might
help cut costs since it would eliminate produc-
ing two distinct cases.—DJD
Wipe it cleaner
I'm upgrading my Apple Ile and one of the
components I'll be selling is a 10 megabyte
ProFile hard drive. Before | sell it, how can I be
absolutely certain I've removed all information
from it? It would be a disservice to my clients,
as well as potential legal malpractice, to inad-
vertently allow some clever hacker to expunge
any of those old files. | seem to recall that a
simple re-format or ProSel's “wipe volume”
destroys only the file allocation information, not
the files themselves. For complete peace of
mind, | need to be certain I've returned the
ProFile to its virgin state before | sell it.
Charlotte Roberts
Lafayette, Calif.
Ah, yes, the “dark side” of data recovery...if
you have software that performs a true low-
level format of the hard disk, then that should
erase all data.
IF not, then you can use a simple program to
overwrite all blocks on the disk, The following
Applesoft program will accomplish this:
1000 REM = fill volume =
1010 DS = CHRS (4)
1020 PS = “/HARDI/": REM hard disk volume
1030 sv REM subdirectory/volume
1040 PIAS = PS + "SV." + STR (SV) :SV$ = PTHS
1050 GOSUB 1190
1060 PRINT D$;"CREATE “;PTHS
‘1070 SD = 1: REM subdirectory
1080 PTHS = SVS + “/SD.” + SIRS (SD) :SD$ = PIES
1090 GOSUB 1190
1100 PRINT DS; "CREATE ;PTHS
1110 F = 1: REM file
1120 PTR = SDS + “/FILE.” + SIRS (F)
1130 GOSUB 1190
1140 PRINT DS; "SAVE ";PTRS
1150 IF F< 100 THEN F = F + 1: GOTO 1120
1160 TF SD < 100 THEN SD = SD + 1: GonO 1080
1170 TF Sv < 25 THEN SV = SV + 1: GonO 1040
1180 BD
1190 REM — display progress —
1200 VIRB 1: RTAB 1: CALL - 868: PRINT PTHS
1210 RETURN
To use the program, format your hard disk
(to clean the volume directory information to
reflect a “blank” disk) and run the program
from ProDOS BASIC. Line 1020 should be
changed to contain the volume name of your
hard disk (enclosed in ‘/" characters). If you
have a large hard disk with multiple partitions,
you'll need to repeat this operation for each of
the partitions containing “sensitive” data.
The rest of the program creates subdirecto-
ries on the fly (up to 25 in the volume directo-
ry and up to 100 ‘sub-subdirectories’ in each
of those) and then tries to fill each with copies
of the BASIC program itself, filling the hard
disk. If all goes well, the program should termi-
nate with a ‘DISK FULL" error.—DJD
Debugging help
Is there any software which works like
Maxwell's Demon to debug assembly language
programs under ProDOS?
R. S. Teesdale
Grand Rapids, Mich.
Maxwell’s Demon is a DOS 3.3-based
assembly language debugger that allows you to
execute a machine-language program slowly
(or even one step at a time) while observing
the changes in the 6502 processor's registers
and stack on a specialized display. Debuggers
also usually include a method of viewing and
modifying memory, and of setting “break
Points” to stop execution of a program (run-
ning with the debugger present) when a partic-
ular condition (usually a value of the program
counter) is met.
APDA sells the ProDOS Assembly Tools
(#4222021) for $35 which include a ProDOS-
based debugger called BugByter. It is limited
to the 6502 (not 65C02) instruction set, but
should be usable for most ProDOS 8 programs.
For ligs GS/OS applications, APDA sells
GSBug and Debugging Tools v.4.0BI
(#A0037LL/A, beta; available only to APDA
members) for $30. The GSBug disk includes
an “init” file and application version of the
GSBug debugger, as well as three CDA’s: Load-
er Dumper (displays tool set information),
Memory Mangler (displays memory usage
information from the Memory Manager), and
Scrambler (shuffles memory blocks to identify
problems with memory access in your code).
Note that GSBug is not usable to debug Pro-
DOS 8 applications. —DJD
Fixing PaintWorks fonts
A minor discovery that could help: some
users may have found that the Calligraphy font
(A2-Central disk, February 1989), when used
in PaintWorks Gold, has its delicate curves bru-
tally cut on the right by the text cursor when
you exit the text mode with Escape. The trick is:
type your line with another font (a “straight”
one) then go back to the “Choose Font” option
without touching Escape and choose “Calligra-
phy’. Your line will automatically shift to un-
A2-Central 6.55
truncated characters. Don't forget to leave one
or two extra spaces at the end to take care of
the last character.
Chris Marker
Paris, France
Portable power
1 am designing a “glass cockpit” for a small
homebuilt aircraft which will incorporate naviga-
tional inputs as well as engine sensor inputs.
The display will show a moving-map graphic of
the current position (latitude and longitude)
including airports in the area, as well as bar-
graph information showing engine perfor-
mance. The prototype on the ground is just
begun. | need your help in locating certain
pieces to make this system work in the air.
1am in need of a power supply that accepts a
12 volt input and which outputs the standard
voltages for the Apple ligs: +12 volts at 1 A, -12
volts at 0.25 A, +5 volts at 4 A, and -5 volts at
0.25 A. This need not be an Apple II replace-
ment (in size or power) as mounting and power
requirements have yet to be decided and are
still flexible. Because space and weight are at a
premium in the aircraft a voltage inverter (12
volts DC in; 120 volts AC out) is not an option.
1 will also need a small color video display
that accepts analog video inputs and will run on
+12 volts DC. Digital color inputs are a possibili-
ty, but | would have to use another slot card
(Applied Engineering's IBM video option) which
results in another level of complexity I prefer to
avoid.
An added bonus of using the ligs is the built-
in synthesizer’s wonderful sound capabilities.
High-quality digitized voice messages can be
played into the pilot's headset to inform him of
critical parameter values— a powerful attention-
getter in an environment (the traditional cock
pit) dense with so much critical instrumenta-
tion.
Dave Carpenter
Cupertino, Calif.
This project sounds like it’s over our head
(Ouch!—sorry).
There are linear (analog) integrated circuit-
modules that can provide regulated voltage
available at your local Radio Shack or other
electronics stores. These may work in your sit-
uation, but will require constructing your own
supply. The monitor sounds like a tougher item
to locate. Maybe a reader can provide some
advice on a monitor or applicable power sup-
ply module.—DJD
Dictionary patch
In connection with the letter by Jean-Guy
Mariage in your May 1990 issue, | send you a
patch which allows one to use the extra key-
board characters in one’s custom dictionary.
The patch is for AppleWorks 3.0. | got it from
Alan Bird at Beagle Bros and it works fine.
Words containing “other characters” are not bro-
ken down while checking spelling and they are
checked properly via the custom dictionary (but
they won't be displayed as “suggested
spellings”). The patch goes as follows:
BLOAD SEG. WP, AS300, BSBFCB, 15
PORE 768, 64
PORE 772,126
BSAVE SEG.WP,A$300, BSBFCB, L5
Dr. S. S. Datye
Akureyri, Iceland
6.56 A2-Central
We pass this patch along for those who may
be interested in trying it. In a quick test, we
discovered the patched spelling checker will
ignore the special characters preceding the
first (USA) ASCII character code in a
word.—DJD
A better switch
Do you know about SWCP? It’s a $15 share-
ware program by Tim Grams that lets you switch
slots “on the fly.” It’s great when you want to
use a 5.25 drive only occasionally and have a
card in the slot. It's super.
Merry Perry
Baltimore, Md.
SWCP (software control panel) is a ProDOS
8 program that can execute (and archive) IIgs
Control Panel settings in preparation for run-
ning another program. The program seems to
be primarily oriented for switching (and initial-
izing) slots under ProDOS 8. That may be
enough for most users. Under GS/OS, neces-
sary native drivers may not be loaded since
GS/OS loads the drivers for the devices it finds
during the boot process. Device interfaces you
“switch in” may therefore not have a necessary
driver loaded.
If you want to obtain a version directly from
the author on disk, it is $20 from Tim Grams,
P.O, Box 462283, Garland, Texas 75046-
2283.—DJD
GS/OS fan
Having been a system engineer, | am greatly
AZ Central :
© Copyright 1990 by
A2-Central
Most rights reserved. All programs published in A2-Central are pubic
comin and may be copied and distributed without charge. Appe user groups
and sniicant others may cbtain permission to eprint artes rom time fo time
byspecife wen request
Edited by
Dennis Doms
with hel rom
Tom Weishaar Sally Dwyer Dean Esmay
Joyce Hammond JeffNeuer Jay Jennings
Tom Vanderpool Jean Weishaar
A2-Central_—tted Open-Apple though January, 1982—has been pub-
lished monthly since January 1985. Workwide pies (a US. dolar: aimai
livery incded at ro ational charge) $28 fr 1 year, $54 for 2 yeas; $78 for
3 years. Al back ssues are curently avaabl for $2 each; bound, ndexed ed:
tors of our fist four volumes are $1495 each. Volumes end with the January
issue; anindex forthe prior volume &incuded wih te February issue.
The fulltext ofeach issue of A2-Central s availble on 3.5 disks, along
wit a selection of the best new pubic domain and shareware fs and pro-
‘grams, fr $84 a year (newsletter and disk combined). Single disks are $10.
ease 0
send al
A2-Central
P.O. Box 11250
Overland Park, Kansas 66207 U.S.A.
‘A2-Central is sold in an unprotected format for your convenience. You
are encouraged to make back-up archival copes or easy-to-read enlarged
copies for your own use without charge. You may also copy A2-Central for
distrbution to others. The dstibution fee is 15 cents per page per copy ds
tributed.
WARRANTY AND LIMITATION OF LIABILITY. | warrant that most of the
information in A2-Central is uselul and correct, although drivel and mis
takes are included from time to time, usually unintentional. Unsatisied
subscribers may cancel their subscription at any time and recewve a full
refund of their last subscription payment. The unflled porton of any paid
subscription will be refunded even to satisfied subscribers upon request.
MY LIABILITY FOR ERRORS AND OMISSIONS IS LIMITED TO THIS
PUBLICATIONS PURCHASE PRICE. in no case shal | or my contibutors
bee liable for any incidental or consequential damages, nor for ANY dam:
ages in excess ofthe fees paid by a subscriber.
SSN 0885-4017 (GEnie mail: A2-CENTRAL
Voice: | 913-469-6502
Printed in the U.S.A. Fax: 913-469-6507
impressed with the IIgs system software— it is
very complex, yet easy to use; it is fast (com-
pare it to Windows on a 80286!); and it is virtu-
ally bug free! | would guess that version 5.0.2
could contain over 350,000 lines of code, all
with no significant errors!
For those readers without mainframe operat-
ing system experience, | one time performed an
installation for IBM OS PCP on a 64K (that’s
65,536 total system memory bytes) $360-30.
The system installation took over six hours con-
tinuous run time (tell that to anyone who com-
plains about the GS/OS Installer).
OS for S367/370 was not error free. Release
#15 included a book of known errors that was
over | inch thick (the book of corrected errors
was over 3 inches thick). And OS/MFT or
OS/MVT was never a dependable system. In
normal operation it crashed at least once each
hour! It really did not become dependable until
MVS arrived in the late 70's.
In GS/OS, we have an operating system that
is at least as complex as IBM's OS, and yet it
can be run by complete nincompoops. It does
not crash into the monitor, and it does not trash
disk files or directories. It is state of the art—
reliable, and easy to use. The people at Apple
all deserve a big thanks!
Ken Lessing
Fresno, Ohio
We’re not in a position to judge mainframe
operating systems; we'll accept your comments
without trying to sort out the acronyms.
Anyone interested in learning about working
on a large software project should read Freder-
ick Brooks’ Mythical Man Month. It covers the
development of the original disk operating sys-
tem for the IBM 360 mainframe and explains
everything that was done wrong. It wasn’t the
development team’s fault; this was the first
Project of such complexity ever attempted;
there was much to be leamed.—DJD
Lower cost llgs?
| find it strange that anyone should be taken
in by the story that Apple Computer cannot pro-
duce Apple li equipment at a price the market
will bear. The Apple II used to have its own line
of equipment.
What's left? The RGB monitor, the processor
itself, and the 5.25 drive. That's all, folks. The
Apple Il Video Overlay Card sells for a song
compared to the Mac video overlay card.
The story I heard was that Apple Computer
can produce a low cost Mac for less than the
IIgs and sell the Mac for greater profit. There is
the rub; the Mac has a larger market price. All
this proves is that Apple Computer has not used
their best talent to reduce the production cost
of the Ilgs. They have let the IIgs production go
sour on them.
And Apple’s answer to not producing a lower
cost IIgs is, of course...shove Macs, expensive
Macs, at the schools. Pushing a Mac into the
school systems is the same as selling an IBM
clone. The decision is the same for the school.
The llgs can be booted over a network. A
Mac on a network must have its own boot
device. Does Apple Computer point out this IIgs
advantage? No.
The ligs could handlé 4 megabytes of memo-
ty from day 1. How many years did it take
before the Mac got its first megabyte? Does
Apple Computer talk about the ligs advantage?
No.
The Apple II line was successfully using hard
Vol. 6, No. 7
drives before the Mac had a hard drive avail-
able. Does Apple Computer mention this in
their advertising? No.
The most successful integrated desktop pro-
gram in the world runs on an Apple IL. Will the
Mac ever have this success? No.
Sigh, when will we ever learn?
Ken Kashmarek
Eldridge, lowa
Your letter actually encompasses two
approaches, lowering the price of the Ilgs, or
raising it’s perceived value. Most of us who
love the Apple Il probably feel it’s the latter
approach Apple has most obviously avoided.
At Apple Fiesta, Barney Stone indicated that he
believed that the Apple II would be a visible
part of Apple’s 1990 fall promotions.
Another question of interest is whether
Apple included marketing expenses in the cost
of its Ilgs systems. Apple has been roundly crit-
icized for the inefficacy of it’s 1989 fall promo-
tion for its lower-cost systems, and that weight
lies heavily around the neck of the ligs (the
only Apple II system included in the promo-
tion).
We've found it strange to note the number
of Macintosh articles regarding the price of Mac
systems, including Steven Levy's article “Just
Say Low" in the July 1990 MacWorld. It con-
tained a couple of insightful comments from
Apple's Brodie Keast: that despite Apple’s
experiment in cutting the price of the Mac Plus
last winter (so that it could be offered near or
below the $1000 mark) sales did not rise
remarkably, and that there were significant dif-
ferences in the manufacturing costs for a more
powerful model (the SE/30). It makes us won-
der how much a “low-cost Mac” will actually
affect the market. —-DJD
Vote with your wallet
I feel we can be on the brink of another
Apple II revival with the reports in A2-Central
and John Sculley’s letter in The Apple Iigs
Buyer's Guide. This represents the most encour-
agement in years.
I also agree with Paul Statt’s view that to
make the Apple Il a success, it is up to us to
purchase new hardware and software and to
extend those subscriptions to our favorite publi-
cations. If there’s a market, then there will be
products produced to meet this demand. So,
put your money where your heart is.
Lawrence Perry
Asheboro, N.C.
Apple can help the Apple II market by more
effective advertising to increase interest and
confidence in Apple II systems. Nobody can
buy a system that they don’t know about. No
one (except hard-core enthusiasts) will buy a
system if the company doesn’t make its sup-
port for the system obvious.
If Apple keeps that end of the bargain, then
the viability of the market depends on the
Apple II consumers. Paul Statt made the point
in his InCider editorial that ultimately we vote
with our wallets.
Incidentally, don't overlook Mac companies
in this regard. Many Mac peripherals will work
on the Ilgs, and it doesn't hurt to scan a Mac
magazine for hard disk prices (for example)
and make a few calls.—DJD