horcel@kuhub.cc.ukans.edu (03/21/91)
In case anyone is interested...
System Overview
Q: Why is Apple talking about System 7.0?
A: Apple is discussing with developers the core technologies under
development for inclusion with the next major Macintosh System
Software release, System 7.0. Apple's goal in talking about
System 7.0 is to shorten the amount of time necessary to bring a new
generation of powerful application software to Macintosh users.
Q: Why is Apple telling the Press?
A: Apple is informing the press to explain what we're working on in
our labs, why we've chosen the paths that we have and how these
technologies will affect users and developers.
Q: What is Apple's message to users?
A: System 7.0 will extend the user's reach through an expanded set
of capabilities that utilize the same consistent, intuitive techniques
that users already know.
Q: What's in System 7.0?
A: Although the complete feature set of System 7.0 will not be announced
until later this year, the following projects will be a part of
the release:
* Virtual Memory/32-Bit Addressing
* IAC Architecture
* Outline fonts
* New Print Architecture
* Layout Manager
* Database Access Manager
* Finder 7.0
In addition, System 7.0 will integrate 32-Bit QuickDraw (introduced in April)
and the Communications Toolbox (shipping Q3).
Apple is also discussing with developers other projects currently
under development:
* Sound Manager Enhancements
* File System Enhancements
* International Text Support
* Installer 3.0
The complete feature set of System 7.0 will be announced later this year.
Q: When will System 7.0 be made available to customers?
A: Apple will announce customer availability later this year.
Q: What Macintosh computers will be able to run System 7.0?
A: System 7.0 is being designed to run on all Macintosh Plus, SE,
SE/30, II, IICX and IIX computers.
System 7.0 will require two megabytes of RAM.
68030-equipped Macintosh computers and Macintosh II computers with PMMU will
have the additional benefit of Virtual Memory.
Q: Apple says that eventually all Macintosh computers will run System 7.0.
Does that mean that all Macintosh computers will eventually be shipped
with two megabytes of RAM?
A: Not necessarily. Apple is exploring various configurations of RAM,
ROM, processor and hard disk that will support System 7.0.
Q: What does that mean?
A: The total amount of memory that a Macintosh uses for system software
is a combination of RAM and ROM. In general, when more software is put
into ROM, a Macintosh requires less RAM. In addition, for Macintosh
computers with PMMUs, the hard disk can be used to extend RAM with
virtual memory. These technologies provide for many alternative
configurations in the future.
Q: Is Apple announcing System 7.0?
A: No. Apple is making a directional statement to third-party
developers about new technologies that will be included in a
future version of Macintosh System Software.
Q: What are Apple's plans for System 7.0?
A: Apple will move the entire Macintosh product line to System 7.0.
During this transition, Apple will offer both the System 6.0
series and System 7.0 CPU configurations.
Q: How many current applications will be compatible with System 7.0?
A: Application software that conforms to the Inside Macintosh
guidelines will be compatible with System 7.0. As System 7.0 is
still in development, it is impossible to determine which
applications will be 100% compatible. When System 7.0 ships,
Apple plans to make a compatibility report publicly available.
Q: What should customers do to prepare for System 7.0?
A: There is nothing that users need to do now. In making new
purchase decisions, customers should buy the Macintosh configuration
that suits their current needs. Users and businesses that need to
make long range decisions now may want to purchases machines with two
megabytes or more of RAM today.
Q: Why will System 7.0 need two megabytes of RAM?
A: The new features of System 7.0 will require more memory than is
available in a one megabyte system to operate effectively.
Q: Where is the multitasking Macintosh Operating System?
A: The Macintosh operating system has been multitasking since the
introduction of MultiFinder. Many applications take full advantage
of MultiFinder by allowing concurrent printing, recalculating
spreadsheets, sorting databases, and downloading files.
Q: Isn't Apple now putting Macintosh users through an OS/2 experience?
A: Absolutely not. System 7.0 is an extension of today's Macintosh
System Software. Apple is designing System 7.0 to provide for
upwardly compatible applications which is a fundamental part of
the Macintosh System Software strategy. System 7.0 will allow
developers to create even more innovative Macintosh software
and hardware that extends the power of today's Macintosh.
Q: Can a Macintosh II-class machine running Apple's Virtual Memory
operate with 1 megabyte of physical RAM?
A: While this configuration may work, Apple will recommend 2
megabytes of RAM when running System 7.0.
Q: I own a Macintosh II computer with one megabyte of RAM and
I want to upgrade to System 7.0. Should I buy more RAM or an MMU?
A: RAM. The least expensive way to upgrade a one megabyte
Macintosh II to System 7.0 is to add another megabyte of RAM.
Q: Does Apple have plans to add new capabilities to the
System 6.0 series?
A: No. Users of the System 6.0 series can continue to
use their systems. The upgrade path for System 6.0 series
users is System 7.0 with more RAM.
Q: Why not?
A: Apple believes in providing consistency across our products
for our users and developers. This consistency can only be
achieved by focusing on one system software platform.
That platform is System 7.0.
Q: Does this mean that all users need to upgrade to System 7.0?
A: No, users can continue to use the System 6.0 series and their
current applications. System 7.0 and new applications software
will provide many new capabilities that many users will want.
System 7.0 offers users an opportunity to add more functionality to the
Macintosh computers they own today.
Core Technologies for System 7.0
Virtual Memory
Q: What is Virtual Memory?
A: Virtual Memory (VM) extends the user's available memory by
transparently treating the hard disk as additional RAM.
Q: Why is Virtual Memory important?
A: Virtual Memory allows users to run more applications at
once and work with larger amounts of data than they can today.
Q: Will Virtual Memory be compatible with application software?
A: Yes. Virtual Memory is backward compatible with all
applications that adhere to Inside Macintosh.
Q: Which Macintosh computers can use Virtual Memory?
A: Macintosh IIx, IIcx, and SE/30 are ready to use Virtual
Memory--no additional hardware is needed. Macintosh II can
take advantage of Virtual Memory by adding the 68851 PMMU
coprocessor onto the Macintosh II logic board (a socket is
provided on the board for this chip). This coprocessor chip
is on the Apple price list. This is the same co-processor
needed to run A/UX, Apple's version of AT&T's UNIX operating system.
Apple's 68000-based systems?Macintosh Plus and Macintosh SE?cannot
take advantage of the Virtual Memory capability of System 7.0.
Macintosh SE owners have the option of the Macintosh SE/30 logic
board upgrade to gain Virtual Memory capabilities.
Q: Why can't Macintosh Plus and Macintosh SE use Virtual Memory?
A: The 68000 microprocessor used in the Macintosh Plus and
Macintosh SE does not have the memory management hardware
necessary for Virtual Memory. This memory management
capability is one of the primary differences between the 68000 and its
successor chips.
32-Bit Addressing
Q: What is 32-Bit Addressing?
A: 32-Bit Addressing enables the Macintosh to use up to 4
gigabytes of memory. The basic software and hardware of the
Macintosh already supports the 32-Bit Addressing model.
However, Macintosh currently is limited to 8 megabytes of
memory because 32-Bit Addressing is not yet fully implemented
throughout the system.
Q: Why is 32-Bit Addressing important?
A: Macintosh users want more memory for a variety of uses.
Some just want to run more applications under MultiFinder.
Some want to use graphics software that creates multimegabyte
images. Some want to use enormous databases. And some want to
manipulate even larger word processing documents. More memory
has endless uses.
Q: So does the transition to 32-Bit Addressing mean lots of
application incompatibility?
A: No. Applications that conform with 32-Bit Addressing
guidelines ("32-Bit Clean") already run on all Macintosh
computers. These applications will immediately benefit from the
enlarged memory space with System 7.0. Applications that are
not 32-Bit Clean will continue to run under System 7.0 but will
not have the benefit of additional memory space. Apple has been
working with its third-party developers to make sure that all
application software abides by 32-Bit Cleanliness rules.
Q: What Exactly is "32-Bit Clean"?
A: Applications that follow Apple's memory manager guidelines
in Inside Macintosh are 32-Bit Clean. 32-Bit Clean applications
are both upwards and downwards compatible with 24-Bit and 32-Bit
Addressing modes. These guidelines are repeated in Technical Note #212
"The Joy of Being 32-Bit Clean."
Q: What if my existing software is not 32-Bit Clean?
A: Applications that are not 32-Bit Clean continue to run with System 7.0.
Q: Will 32-Bit Addressing become standard?
A: Yes. At some point in the future, Apple will make 32-Bit
Addressing standard on new Macintosh computers.
Q: Since much of system software is in ROM, will I need a new
ROM to get the benefits of 32-Bit Addressing?
A: Apple is researching ways of offering 32-Bit Addressing
to all Macintosh II computers. While an optional ROM upgrade
is not out of the question, other alternatives are being
considered. Apple will keep developers and customers updated
on issues relating to 32-Bit Addressing.
Q: What does 32-Bit Addressing mean for Macintosh Plus
and Macintosh SE?
A: These Macintosh computers cannot get the benefit of 32-Bit
Addressing. 32-Bit Clean and non-32-Bit Clean applications
will continue to run on these computers. Only Macintosh
computers using the 68020 or 68030 microprocessor can have
the benefit of 32-Bit Addressing.
Interapplication Communication Architecture
Q: What is Interapplication Communication Architecture?
A: Interapplication Communication Architecture (IAC) is a
framework for applications to exchange commands and data,
both locally and over networks. IAC consists of several pieces:
* Program-To-Program Communications (PPC)?a low-level tool
for exchanging data between two programs, either locally
or across networks. PPC provides a unified, consistent
programming interface for both local and network communication.
PPC will be able to deliver messages either Immediate (IPC)
or Store-and-Forward.
* Event Manager PPC?a high-level tool for applications to
exchange commands and data. Event Manager PPC presents a
simple, natural interface to applications.
* AppleEvents(TM)?an Apple-defined protocol of standard messages
that Applications can send to other applications. Examples
include "Open Document," "Print," "MoveWindow."
* Live Copy/Paste and Link Manager?Live Copy/Paste gives you
live links between documents. For example, the user can
link a spreadsheet table into a word processing document;
whenever the spreadsheet changes, the word processing document is
automatically updated.
* Clipboard Copy/Paste?a current capability of Macintosh and
is also part of IAC. Macintosh applications universally
support copy and paste between applications via the Clipboard.
Q: What does Live Copy/Paste offer the user?
A: As applications begin to offer Live Copy/Paste, users will
be able to make applications work better together and avoid
repetitive copy and paste. Users can build up libraries of
commonly used objects?like graphics or paragraphs of text?and
link them into their documents. When you need to changethe data,
you change every occurrence of that data. And, because Live
Copy/Paste works with AppleShare(R) file servers, you can
easily share data with another person. Imagine linking in the
sales numbers from each of your sales people's spreadsheets.
Your master spreadsheet is always up-to-date.
Q: Does Live Copy/Paste work with existing applications?
A: No. Applications must be revised to take advantage of
Live Copy/Paste. Apple is simplifying the task by providing
extensive user interface guidelines and toolbox support
for Live Copy/Paste.
Apple's outline fonts
Q: What are outline fonts?
A: Outline fonts are mathematical descriptions of characters.
Sharp text at any size on any device can be generated from
outline fonts. Today, the fonts in your Macintosh are called
"bitmap" fonts. These fonts are small collections of pixels
that create the text you see on the screen. With bitmap fonts
the System File can become huge and still not have all the
fonts in all the sizes you might want.
The new Apple fonts are outline fonts.
Q: What are the benefits of outline fonts?
A: Outline fonts provide sharp text at any size on any device.
This means beautiful documents on the screen for multimedia
presentations as well as on the page from any printer. Outline fonts
also simplify the customer experience by creating a single font standard for
the Macintosh computer.
Q: This is confusing. I thought my Macintosh "Style" menu
already had a function for outline fonts.
A: It does, but that is something different. The "Outline"
option in the Style menu actually traces
the character to give it an outlined appearance. It looks like
this. It is simply a graphics trick. However, the new Apple
fonts are called outline fonts because they are based on mathematical
outlines, not bitmaps. These outline fonts are also called spline
fonts or scalable fonts. If you really want to get carried away,
keep in mind that you will be able to "Outline" the outline fonts!
Q: Do Macintosh owners need to buy PostScript fonts anymore?
A: Macintosh owners may want to buy PostScript fonts that
are not yet available in Apple's format. It is important to
remember that today's PostScript fonts, like all of the existing
Macintosh font technology, will still operate normally in the future.
For example, PostScript fonts and bitmaps will remain popular on
1MB Macintosh computers like the Plus and SE. PostScript fonts
may also remain useful in multivendor environments. We expect some
vendors will continue to offer their typefaces in PostScript format
and add the Apple format version of that typeface. However,
the Apple font format will be all most Macintosh owners really need.
Q: How many fonts will be available in the Apple format?
A: Hundreds of fonts will be available within a few months of
first shipment, and thousands soon after. It is impossible to
answer this question precisely. There are two main reasons for
this. First, since Apple's format was designed to be very flexible,
many vendors will be able to automate the conversion of their
existing library to the Apple format. Second, the open format
is available to anyone, so even small players will be
able to create new typefaces. Apple does not have to get
involved in licensing or support. Since there are hundreds
of specialized fonts now available in bitmap formats, these are
all candidates for conversion to outline.
Q: Will Apple be providing fonts in the new format? If so, how many?
A: Apple does not intend to be in the font business, so we will
offer a core set of fonts and then stop. This promotes a healthy
aftermarket for type vendors. The Apple core set will consist
roughly of the fonts Apple ships today with Macintosh computers and
LaserWriter printers, plus a small number of additions. The final
list will be announced later.
Q: Who really needs this technology? After all, LaserWriter NT
and NTX users already enjoy scalable type. Why put it into the Macintosh?
A: Today, the benefits of outline fonts are available from
Apple only through these two LaserWriter models. Now, outline
fonts will enhance the screen display, the ImageWriter II,
the AppleFax Modem, the ImageWriter LQ and the LaserWriter IISC. A
wide range of third-party output devices will also use these
fonts for best possible text quality.
Q: Does this mean that future Apple printers will not support PostScript?
A: No. Keep in mind that the existing Apple printer line
consists of both "intelligent" and "passive" printers. Where we
put the processing power is generally a price/performance decision.
Consequently, future Apple printers will support the new Apple font
format in a variety of ways. Apple is committed to maintaining
excellent system support for PostScript printing. However, our
policy is to not comment specifically on hardware products
under development.
Q: Does this mean that Apple won't be using Display PostScript?
A: Yes. But this should come as no surprise. Apple announced
over a year ago that we will be improving the internal software
of the Macintosh instead of adopting an outside language. This
removes limits from what Apple can do in software while maintaining
excellent backward compatibility. This new font format, like
32-Bit QuickDraw, demonstrates both of these benefits. At the same
time, we are committed to maintaining an excellent interface
to PostScript printers.
New Print Architecture
Q: What is New Print Architecture?
A: The New Print Architecture is designed to extend the printing
capabilities of Macintosh.
Q: What are the advantages of the the New Print Architecture?
A: There are three advantages to the New Print Architecture:
1. New features. Background printing on all printers, increased
performance, support for outline fonts, color/gray scale support,
elimination of document reformatting, and an enhanced user interface
will extend the lead the Macintosh has in printing.
2. A wide variety of new printing devices. Where in the past
it has taken years to support new printers on the Macintosh,
with the New Print Architecture it takes only a few months.
We expect to have more well integrated printers available on Macintosh than
any other computer.
3. Compatible expansion for the future. Expandability is
designed into the new print architecture. With the New Print
Architecture we expect to be able to transparently offer
new features to both the user and application.
Q: How does the New Print Architecture compare to printing in
Presentation Manager?
A: So far there are very few drivers for Presentation Manager.
With Presentation Manager, Microsoft is writing application
independent drivers for the first time. Apple has utilized
four years of experience to develop a new print architecture
that utilizes outline fonts, the Line Layout Manger, 32-Bit QuickDraw,
and other system utilities. With the New Print Architecture
the Macintosh will remain the benchmark printing platform.
Q: Is it true that all of the current printer drivers will
be incompatible with System 7.0?
A: Yes. Apple's New Print Architecture is designed to
make the creation of printer drivers easy.
When System 7.0 ships, Apple will have new printer
drivers to support all Apple output devices.
Q: Who will write replacement drivers for these devices?
A: Apple will work closely with third-party developers to
help in the creation of new printer drivers built around Apple's
New Print Architecture.
Line Layout Manager
Q: What is the Layout Manager?
A: The Layout Manager allows applications to display typographical
quality text.
Q: What are the benefits of using the Layout Manager?
A: Using the Layout Manager, applications can display sophisticated
formats like kerning, ligatures and justification for any text.
For international text systems, like Japanese or Arabic, the Layout
Manager has additional support for composed characters.
Database Access Manager
Q: What is the Database Access Manager?
A: The Database Access Manager is the Macintosh System interface
that allows applications to transparently connect to remote
databases on host computers.
Q: What benefits does this Database Access Manager give to developers?
A: The main benefit is that applications like spreadsheets, desktop
publishing, or graphics programs can now directly access host data
in a standard way regardless of the host computer and database.
Q: How does Apple's approach compare to IBM's OS/2 Extended Edition
or Microsoft's SQL Server products?
A: The Apple Data Access Manager provides standard access to
remote host databases. This is where the bulk of computerized
data is found. In contrast, the IBM product is only a local
database that resides on a single user's machine. The Microsoft
product is a local area network database requiring a dedicated
computer. Both the IBM and Microsoft database extensions are
optional. The Data Access Manager is a standard part of
Macintosh System Software.
Q: What databases does the Database Access Manager support ?
A: ORACLE, Sybase, Ingres, Informix, RDB, Vax-RMS and IBM systems.
Many other databases will be supported in the future.
Finder 7.0
Q: What's new about Finder 7.0?
A: Finder 7.0 improves the Macintosh user interface in three
important ways. First, Finder 7.0 will integrate system functions
that previously had different user interfaces into one consistent,
intuitive interface. Second, we are building in new powerful
features like a quick-find facility, document stationery
templates, aliases that will allow users to organize their
files in multiple ways, and others. Third, Finder 7.0 will
be extensible providing for the integration of new capabilities
like electronic mail and backup in the future.
Q: Will desk accessories continue to run with Finder 7.0?
A: Yes they will. In addition, because applications can now be
installed in the Apple menu like desk accessories, developers
will be able to provide users with better desk accessories. These
new desk accessories will have all the power of applications with
the instant-access features of the original desk accessories.
Q: What's the relationship of Finder to MultiFinder?
A: MultiFinder is a set of operating system capabilities
that give the Macintosh the capability to run multiple applications
concurrently (multitasking). The Finder is the system utility
software that gives Macintosh users control over their desktop.
The Finder is what you use whenever you launch (double-click)
an application, drag a file onto your hard disk, move folders
between windows, etc.
Q: How does the Finder compare to Presentation Manager or Windows?
A: Neither PM or Windows has a Finder. With these systems,
the user sees a graphic display but does not get the intuitive,
direct control over system functions that the Macintosh provides.
For example, in the Macintosh, a user can copy a file from one
disk to another by merely dragging it. In Windows or Presentation
Manager, file copy requires the user to type cryptic file names
into a dialog box and then the system does the copy. This
forces users to remember file names exactly and to remember
arcane name formatting restrictions.
Q: I have a large number of files on high-capacity hard disks.
Will the Finder 7.0 do anything to help manage files better?
A: Finder 7.0 takes advantage of a new system feature called
the Desktop Manager which can handle many more files more quickly.
In addition, the quick-find facility will allow users to access
files more quickly by automatically finding the folder a file
is stored in, opening it on the desktop, and highlighting the
file that the user seeks.
System Software Explorations
Sound Manager Enhancements
Q: What are the improved audio capabilities?
A: The audio improvements represent new functionality in the Sound
Manager including:
* a real-time sequencer
* multiple channels of simultaneous sound
* audio compression/expansion
* integration of MIDI management tools
Q: Why are these improvements so important?
A: The sound enhancements provide the foundation for more and
better audio in current applications as well as a whole new range
of applications with integrated audio capabilities.
File System Enhancements
Q: What's new in the Macintosh File System for system release 7.0?
A: Five enhancements?FileIDs, Catalog Search, Desktop Manager,
File System Manager and B*tree Manager?will make the Macintosh
work smarter for users.
Q: Why are the File System Enhancements important?
A: As applications take advantage of System 7.0 features,
customers will have greater ability to organize their hard
disks and manage those drives more effectively. Applications will be
able to locate documents much more quickly and under a wide range
of search criteria.
Q: How does the Desktop Manager improve performance of
large disks?
A: Currently, desktop information (file icons and comments)
is stored in an invisible Desktop file. Because of the current
implementation, there is a limit of approximately 2,000
entries in the desktop file and, more importantly, performance
becomes sluggish long before the maximum number of entries
is reached. The new Desktop implementation
removes this size restriction and greatly improves
performance in all cases.
Installer 3.0
Q: What is the "one button Installer"?
A: The "one button Installer" is actually version 3.0 of
Apple's installation program. Installer 3.0 offers "one button"
solution to installing system software on Macintosh personal
computers. Installer 3.0 also offers complete control of the
installation process to those users who want to customize their
installation.
MultiFinder
Q: Is MultiFinder a multitasking operating system?
A: Yes. MultiFinder shares the CPU's time among a number of
applications so that a customer can work on a word processing
document while downloading a file or recalculating a spreadsheet.
In technical terms, multitasking is the ability to perform a
number of tasks concurrently. MultiFinder uses a cooperative
scheduling algorithm to run several applications concurrently.
Q: Will there continue to be a distinction between MultiFinder
and single Finder?
A: No. In System 7.0, MultiFinder will always be turned on.
Q: Why will MultiFinder always be on in System 7.0?
A: Many parts of System 7.0 depend on the functionality of
MultiFinder. As a result, MultiFinder will always be turned on.
Q: What is pre-emptive scheduling?
A: Pre-emptive scheduling is a method of allocating CPU time
among several applications that involves temporarily interrupting
each application in turn when that application has used
up its available time.
Q: Why doesn't MultiFinder offer pre-emptive scheduling?
A: Apple choose to focus on other features that we feel
are more important. Apple is looking at offering pre-emptive
scheduling in future releases of Macintosh System Software.
HyperCard
Q: Will HyperCard support System 7.0 features?
A: Future releases of HyperCard will support System 7.0. While
some features are transparently supported, others will necessitate
additional development. For instance, HyperCard will need to
be extended to take advantage of the high-level SQL calls
included in System 7.0. Likewise, support for other features
in the Live Copy/Paste will mean adding additional code. Other
features, like resolution-independent graphics and
Apple's outline fonts, are transparent to HyperCard and
will need no additional work.
Macintosh Communication Toolbox
Q: What is the Communications Toolbox?
A: The Communications Toolbox is a powerful facility that
gives the Macintosh a fundamental capability to communicate
with remote computers, providing users and applications with
consistent and extensible access to terminal emulation, data
connection, and file transfer functions.
Q: Why has Apple developed the Communications Toolbox?
A: Apple is extending the consistency and modularity that
characterize the user-interface Toolbox to the communications
environment. With the Macintosh Communications Toolbox,
Macintosh sets a new standard in empowering users and developers
to take advantage of communications.
Q: When will it be available?
A: The Macintosh Communications Toolbox will be released
to developers during the third quarter of 1989. The
Communications Toolbox will become standard system software
when released as part of System 7.0.
Q: How will users get the Communications Toolbox?
A: Apple is encouraging the third-party developers who
incorporate the Communications Toolbox into their applications
to bundle the Communications Toolbox with their application.
32-Bit QuickDraw and LaserWriter 6.0
Q: What is 32-Bit QuickDraw?
A: QuickDraw is the graphics system software, given away
in every Macintosh, that is responsible for putting objects,
icons, text, and pictures on the Macintosh display. On
68000-based machines, it supports 8 colors. Until recently,
on 68020/030 Macintosh computers, QuickDraw supported up to
256 colors. Today, extensions to QuickDraw, called "32-Bit
QuickDraw," allow QuickDraw to work with the entire range of
visible color, over 16 million colors. There is no longer
any color limitation on color Macintosh computers.
Q: How will the product be distributed?
A: Developers can license 32-Bit QuickDraw and System 6.0.3
from Apple for shipment with their products. In addition,
32-Bit QuickDraw will be distributed to all dealers, user
groups and bulletin boards typically receiving Apple System
Software. 32-Bit QuickDraw will be incorporated into System 7.0.
Q: What markets would want 32-Bit QuickDraw?
A: 32-Bit QuickDraw is especially useful in markets
demanding high-quality color. In publishing and video,
full color is useful for showing realistic images from natural
sources. For presentations, it is helpful for producing
the continuous tone "ramps" from one color to another that
are used in slides. Finally, 24-bits of color make continuous
data easier to visualize for many scientific applications.
As an enabling technology, image visualization can be expected
to open many other new markets.
Q: What are 16-bit, 24-bit and 32-bit color?
A: 16 bits of color can produce very life-like images, 24
bits per pixel is known as "full color" because with 16
million colors available, the eye loses its ability to
distinguish between color incrementally. The additional 8
bits of color that differentiate 24-bit color from 32-bit
color are usually used to store non-color information about
the pixel; for example, one of the bits could be used for
"transparency" information to allow a level of the
background to "show through" the color of a pixel. This is
known as an "alpha" byte.
Q: What is LaserWriter 6.0?
A: LaserWriter 6.0 is a new release of Apple's LaserWriter
driver. Nearly all Macintosh applications use Apple's
graphics system software, QuickDraw, to draw on and off the
screen. The LaserWriter driver translates QuickDraw
instructions into PostScript commands, allowing PostScript
printers (like Apple's LaserWriter printers) to reproduce
what the user sees on the screen at high resolution.
Q: What's new about LaserWriter 6.0? How is it different
from the LaserWriter 5.2 driver that now ships with
LaserWriter printers and System Software?
A: Color printing. LaserWriter 6.0 adds the capability
to translate color QuickDraw images into color PostScript commands.
Any application that supports color QuickDraw now also
supports color printing on color PostScript printers.
Previously, unless an application sent color PostScript directly
to the printer, color printing was not possible on these
printers.
Halftone printing. Users of monochrome PostScript printers
benefit as well. Color images are halftoned by the printer.
Halftoning is a technique that produces dot clusters
of varying size that are perceived as different shades of gray.
The resulting print is much more faithful to the original
image than a high-contrast print composed only of solid black
and white regions.
Faster text printing. The font query mechanism has been
improved substantially in LaserWriter 6.0. It takes less
time for the printer to report its available fonts to the
Macintosh. The result is reduced overall time-to-print,
especially for users who have large font library hard disks
connected to their printer.
32-Bit QuickDraw printing. LaserWriter 6.0 supports output
of images created using 32-Bit QuickDraw. A print of a
32-bit image will show smoother color transitions; in
general, rendering will be more accurate and realistic
than an 8-bit image print.
Extensible menu for page-size choices. The Page Setup
dialog of LaserWriter 6.0 includes the page size choices
US Letter, US Legal, A4 Letter, and B5 Letter. It replaces
the Tabloid choice of previous drivers with an "Other" button.
Clicking this button causes a pop-up menu to appear,
offering the page sizesTabloid, No. 10 Envelope, and
A3. Additional page sizes can be added to this menu by
installing the proper resource. Thus, printer vendors can
ship a driver with their product that includes a page size
specially created for that device. Current color printers
have smaller printable areas than the LaserWriter, and thus
some parts of full-page images are lost when printed on these
devices. Users can now avoid this by selecting a page
size appropriate for their printers.
Q: Will all applications work with LaserWriter 6.0?
A: Apple's testing indicates that most applications will
work fine with LaserWriter 6.0. Most applications use
QuickDraw for printing as well as for screen imaging; these
applications rarely have problems with LaserWriter 6.0.
Other applications do their own conversion of a screen
image to a PostScript page description, and send this
PostScript directly to the printer (bypassing most of the
LaserWriter driver). Some of these applications will not
print as expected with LaserWriter 6.0. There are several
possible effects:
1. Output of a color image is in black and white,
even on a color printer.Many applications that send
their own PostScript to the printer do not send any of the
PostScript required for color printing. It is difficult
for an application to determine whether the printer
is color or not. The options are A) ask the user, or
B) assume a black and white printer. Most applications
do the latter.
2. No output.
A few applications that send their own PostScript rely
on certain variables in the Laser Prep code that is
a part of the LaserWriter driver. Apple has discouraged
this practice, but not with 100% success. The Laser
Prep code has changed in LaserWriter 6.0. Applications
that assume that certain variables are defined will
generate PostScript errors when the user tries to print;
nothing will be printed. The work-around for this is to use
LaserWriter 5.2 until the developer revises the application.
3. Other problems when printing.
Some problems may occur when printing using
"Color/Grayscale" mode, but not with "Black & White" mode.
This is because a few applications assume they will be printing
in black and white. They try to write directly to data
structures that changed when the color capability was
added to the driver. The work around for this is to use
"Black & White" mode when printing until the developer
revises the application.
Q: In the past, new LaserWriter drivers were incompatible
with older drivers. Is this still the case?
A: Yes. LaserWriter 6.0 is not compatible with LaserWriter 5.2.
LaserWriter "wars" can be avoided by ensuring that all users
on a network who share printers have the same version
of the driver installed.
Q: Should every user change to LaserWriter 6.0?
A: No. Those users who meet one of the following
criteria, should use LaserWriter 6.0:
1. Use a color Macintosh (IICX, II, IIX, or SE/30) and
print documents containing color (or grayscale)
2. Use a printer with an attached font library disk
(i.e. have several hundred fonts available)
3. Share a printer, via a network, with any other user
who uses LaserWriter 6.0
Q: How do I get LaserWriter 6.0?
A: LaserWriter 6.0 will be part of Apple's color disk that
will also include 32-Bit QuickDraw. This disk will be
distributed to all Apple authorized dealers. The driver
will also be distributed to electronic bulletin boards,
user groups, APDA, VAR reps, Apple System Engineers, and
reps for National and University Accounts.
LaserWriter 6.0 will be available for licensing to vendors
of color PostScript printers and other third-party developers.
Q: Will LaserWriter 6.0 be included with system software
or LaserWriter II printers?
A: No. LaserWriter 5.2 will continue to ship with both
system software and LaserWriter II printers. When a new
Macintosh is added to an existing network whose users have
LaserWriter 5.2, it will be fully compatible. The network
will need to update to LaserWriter 6.0 only if one or more
users desire its color and font-handling features.
>>> Got it from a friend. Hope it will help someonerad@genco.bungi.com (Bob Daniel) (03/26/91)
In article <1991Mar21.095051.29206@kuhub.cc.ukans.edu> horcel@kuhub.cc.ukans.edu writes: > >Q: Where is the multitasking Macintosh Operating System? >A: The Macintosh operating system has been multitasking since the > introduction of MultiFinder. Many applications take full advantage > of MultiFinder by allowing concurrent printing, recalculating > spreadsheets, sorting databases, and downloading files. This makes me squirm a bit that Apple says MF is multitasking. At best, MF is multiapplication. Try bringing up a dialog and watch all background processes come to a hault. Try pulling down a menu bar (even under Sys7) and watch everything stop. During a download, pull up a dialog from another application and watch the download come to a hault and eventually cancel out. While an application has a progress dialog up, try putting it in the background and pull up something else. Alot of this has been improved in Sys7 with "semi-modal" dialogs but does Apple really think the current MF is really Multitasking? OK, so it's "cooperative- multitasking", but is MF really all that cooperative? Only if the developer makes it friendly? I don't think so, even if the developer uses modal dialogs, he/she has no control of allowing background tasking to occur (without doing some acrobatic tricks). Even under Sys7, he/she would have to convert ALL modal dialogs to "semi-modal" dialogs to allow background *tasks* to continue. Then there is menus. Even under Sys7, pulling down a menu will hault all background tasks. I like MF alot and I like Sys7 a whole lot but it bothers me that Apple would consider MF to be multitasking (even if you consider it to be, it isn't very cooperative). Making a statement like that makes it appear that they don't need to improve the multitaskiness of the Mac. Maybe it's just that I'm spoiled with UNIX, but that statement does make me squirm. I hope Sys8 will address the issues I've mentioned. I feel it should not be up to the developer to be responsible for the "multitaskiness" and that it should be controlled from the operating system. Sys8 could compete with OS/2 and UNIX if it was capable of preemptive multitasking. I'd like to hear unflammable comments... moof moof, Bob Daniel rad@genco.uucp
wilkins@jarthur.Claremont.EDU (Mark Wilkins) (03/26/91)
In article <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >This makes me squirm a bit that Apple says MF is multitasking. At best, MF >is multiapplication. Try bringing up a dialog and watch all background >processes come to a hault. Background processes keep going when a dialog is up, except in some very rare cases where the dialog doesn't check for user input. Otherwise they proceed as usual. > Try pulling down a menu bar (even under Sys7) >and watch everything stop. Very true. This is a difficult problem to fix, I suspect, but we shouldn't give up hope that someone at Apple is working on a solution. >During a download, pull up a dialog from another >application and watch the download come to a hault and eventually cancel out. Again, dialogs usually don't interrupt anything at all. I experimented extensively with this just now to make sure. >While an application has a progress dialog up, try putting it in the >background and pull up something else. This is a problem under System 6 but application developers have the opportunity to deal with it in System 7 and most do. >I like MF alot and I like Sys7 a whole lot but it bothers me that Apple >would consider MF to be multitasking (even if you consider it to be, it >isn't very cooperative). Making a statement like that makes it appear >that they don't need to improve the multitaskiness of the Mac. I think it's really not good to assume that just because Apple calls the Mac multitasking (which it really is -- I've run Mathematica calculations, downloads, and a Go game with my computer all at once, on occasion) that Apple isn't working on improving it. The flaws in the current multitasking environment are well-known, freely admitted, and are almost certainly being worked on. It's just a difficult problem to solve without breaking everything all at once. I believe that the current strategy is to break things little by little so that it doesn't hurt as much. :-) >Maybe it's just that I'm spoiled with UNIX, but that statement does make >me squirm. I hope Sys8 will address the issues I've mentioned. I feel >it should not be up to the developer to be responsible for the >"multitaskiness" and that it should be controlled from the operating >system. Sys8 could compete with OS/2 and UNIX if it was capable of >preemptive multitasking. The Macintosh, just like a UNIX system, requires the developer to code in certain ways in order to make a program run in a friendly way under a multitasking environment. With the Mac, you need to make certain system calls regularly. With UNIX you need to watch your resource usage in certain ways. When you fail to follow either rule, your results will be about the same in either environment. I don't claim it's easy to program a Mac application in a MultiFinder friendly way, although it's not all that hard. However, commercial software authors are motivated to do what it takes to make a Mac program MultiFinder friendly, and so most programs out there are. If you're a programmer, then you may feel far more strongly than the typical advanced Mac owner, but you shouldn't mistake your concerns for those of the whole community, most of which is far more interested in backward compatibility than an immediate jump to a preemptive multitasking system with no regard to how much existing software breaks. >I'd like to hear unflammable comments... I hope I didn't seem harsh. In any case, you're not alone in your views and I'm sure the engineers at Apple know this. >moof moof, >Bob Daniel >rad@genco.uucp -- Mark Wilkins -- ******* "Freedom is a road seldom traveled by the multitude!" ********** *-----------------------------------------------------------------------------* * Mark R. Wilkins wilkins@jarthur.claremont.edu {uunet}!jarthur!wilkins * ****** MARK.WILKINS on AppleLink ****** MWilkins on America Online ******
nerm@Apple.COM (Dean Yu) (03/26/91)
In article <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >In article <1991Mar21.095051.29206@kuhub.cc.ukans.edu> horcel@kuhub.cc.ukans.edu writes: >> >>Q: Where is the multitasking Macintosh Operating System? >>A: The Macintosh operating system has been multitasking since the >> introduction of MultiFinder. Many applications take full advantage >> of MultiFinder by allowing concurrent printing, recalculating >> spreadsheets, sorting databases, and downloading files. > >This makes me squirm a bit that Apple says MF is multitasking. At best, MF >is multiapplication. Try bringing up a dialog and watch all background >processes come to a hault. Try pulling down a menu bar (even under Sys7) >and watch everything stop. During a download, pull up a dialog from another >application and watch the download come to a hault and eventually cancel out. >While an application has a progress dialog up, try putting it in the >background and pull up something else. > >Alot of this has been improved in Sys7 with "semi-modal" dialogs but does Apple >really think the current MF is really Multitasking? OK, so it's "cooperative- >multitasking", but is MF really all that cooperative? Only if the developer >makes it friendly? I don't think so, even if the developer uses modal dialogs, >he/she has no control of allowing background tasking to occur (without doing >some acrobatic tricks). Even under Sys7, he/she would have to convert ALL >modal dialogs to "semi-modal" dialogs to allow background *tasks* to continue. >Then there is menus. Even under Sys7, pulling down a menu will hault all >background tasks. > I really should know better than to open my mouth on this subject, but... Background applications do not "come to a halt" when you have a modal dialog up in the foreground application. They will continue to receive time. Yes, things do stop when a menu is held down. Usually, that's only for a short period of time. You can always write an INIT that installs a procedure in MenuHook which calls _EventAvail or _GetOSEvent if you really want. That would give up time to background applications. (That was Dean the Skanky INIT writer talking, not Dean the Apple Employee.) >I like MF alot and I like Sys7 a whole lot but it bothers me that Apple >would consider MF to be multitasking (even if you consider it to be, it >isn't very cooperative). Making a statement like that makes it appear >that they don't need to improve the multitaskiness of the Mac. > I think that's a misinterpretation. I don't think that saying that we put MultiFinder into the System is equivalent to "That's as multitasking as we're going to get." Also, we of the net are pretty much all power users to some extent. We've all had MultiFinder since 1.0, so we don't think of it as a new feature. Do keep in mind that there are people out in the boonies who don't have access to the Internet that may still possibly be running Finder, and System 7.0 will definitely be a boon to them. -- Dean Yu Blue Meanie, Negative Ethnic Role Model, & Underslept. Apple Computer, Inc. My opinions and so on and so forth...
dorner@pequod.cso.uiuc.edu (Steve Dorner) (03/26/91)
In article <11391@jarthur.Claremont.EDU> wilkins@jarthur.Claremont.EDU (Mark Wilkins) writes: >>While an application has a progress dialog up, try putting it in the >>background and pull up something else. > > This is a problem under System 6 but application developers have the >opportunity to deal with it in System 7 and most do. Some application writers deal with it under System 6. It's not that hard. What I do under system 6 won't work under System 7, but I'll worry about that when I actually get System 7 to play with. > The Macintosh, just like a UNIX system, requires the developer to code in >certain ways in order to make a program run in a friendly way under a >multitasking environment. With the Mac, you need to make certain system >calls regularly. With UNIX you need to watch your resource usage in certain >ways. Uh, name one way you have to watch resource usage under UNIX that wouldn't ALSO apply to the Macintosh. Multifinder is an amazing hack, but no more. It just doesn't have the basic tools to deal with a true Multitasking environment. 1. Cooperative multitasking means a single poorly-behaved application can ruin performance for every other app. 2. No inter-process memory protection means a single poorly-behaved application can ruin everything for every other app. 3. No processing priorities mean it's impossible for the user to decide which tasks are important and should get the CPU. [BEFORE 12,000 ninnies post that the frontmost app has defacto high-priority, please tell me how to write the following function: GiveCPUTimeToCrucialBackgroundAppsButDont- YieldToHorrendousResourcePigsLikeTheMPWLinker(); I really need that function for when my application is frontmost.] For the things most users do, Multifinder is probably adequate; most users don't really NEED multitasking, just "multi-availability", and Multifinder is alright for that. Most users don't use buggy software, either, so 1 and 2 are less of a problem. Multifinder is hell for developers, though, because a) we'd like to compile in the background, and it just isn't practical and b) debugging apps without protection can be very frustrating. (I can't believe it; I just posted a cooperative/preemptive multitasking note. Next thing you know, I'll get involved in a land war in Asia...) -- Steve Dorner, U of Illinois Computing Services Office Internet: s-dorner@uiuc.edu UUCP: uunet!uiucuxc!uiuc.edu!s-dorner
delaney@xn.ll.mit.edu (John R. Delaney) (03/27/91)
This damed "true multitasking" religious debate comes up every few months and most of us reading this group are plain and simply bored with it. We have heard ALL the arguments many time before. Please either hold your piece or put the word "multitasking" in your Subject line so the majority of us can ignore such postings. Thanks in advance, John the atheist
EHYOUNK@MTUS5.BITNET (03/27/91)
I completly agree with the statement that Mac OS should have prementive
multitasking.. Multifinder is a joke, they is no way to download from a modem
while running another app. Here are some other problems.
1. There is no way to set a task's inportance, each other the backgroud app's
get equal time, except for the foreground app, witch can steel all of the cpu
time if it wants too.
2. What about all those nice coprossers the fx has. When will we see a system
that truely supports them!
3. When will we see the new print architechure, line-layout manager, and apple
scripting.
I used to think the apple was a little behind in hardware, and ahead of
the game in system software. Now I think the opposite might be true. I sure
hope that apple gets these into system 8, and I hope system 8 doesn't take
2 years to get here. My next computer purchase may depend on apple coming out
with system 8 on time.
Ed Younk
Michigan Tech Universitytagreen@bronze.ucs.indiana.edu (Todd A. Green) (03/27/91)
In article <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >In article <1991Mar21.095051.29206@kuhub.cc.ukans.edu> horcel@kuhub.cc.ukans.edu writes: >> >>Q: Where is the multitasking Macintosh Operating System? >>A: The Macintosh operating system has been multitasking since the >> introduction of MultiFinder. Many applications take full advantage >> of MultiFinder by allowing concurrent printing, recalculating >> spreadsheets, sorting databases, and downloading files. > >This makes me squirm a bit that Apple says MF is multitasking. At best, MF >is multiapplication. Try bringing up a dialog and watch all background >processes come to a hault. Try pulling down a menu bar (even under Sys7) >and watch everything stop. During a download, pull up a dialog from another >application and watch the download come to a hault and eventually cancel out. >While an application has a progress dialog up, try putting it in the >background and pull up something else. >[edited] >he/she has no control of allowing background tasking to occur (without doing >some acrobatic tricks). Even under Sys7, he/she would have to convert ALL >modal dialogs to "semi-modal" dialogs to allow background *tasks* to continue. >Then there is menus. Even under Sys7, pulling down a menu will hault all >background tasks. Actually all you have to do (or one of the things that you can do) is add a filter proc to your dialog code and give time to other processes. But your right about the Menus, not to mention disk access should have been placed in the "background" eons ago. (IMO). > [more editing] >Maybe it's just that I'm spoiled with UNIX, but that statement does make >me squirm. I hope Sys8 will address the issues I've mentioned. I feel >it should not be up to the developer to be responsible for the >"multitaskiness" and that it should be controlled from the operating >system. Sys8 could compete with OS/2 and UNIX if it was capable of >preemptive multitasking. Yeah, I'm wondering how virtual memory, IAC, file sharing and whatnot are all going to work "smoothly" with out preemptive multitasking. I guess we'll all soon find out. (Unfortunately) >I'd like to hear unflammable comments... > >moof moof, >Bob Daniel >rad@genco.uucp Todd =============================================================================== Todd A. Green "<_CyberWolf_>" --> Pascal <- InterNet: Unix Systems Administration --> Unix <--- tagreen@bronze.ucs.indiana.edu Macintosh Systems Administration --> VMS <---- tagreen@rose.ucs.indiana.edu WCC:136.04 Phone:(812) 855-0949 --> C <------ BitNet: "This is the end, my only --> Mac <---- tagreen@iubacs.BITNET friend, the end" --> NeXT <--- NeXT Mail: -Jim Morrison --> SunOS <-- tagreen@lothario.ucs.indiana.edu ===============================================================================
peirce@outpost.UUCP (Michael Peirce) (03/27/91)
In article <1991Mar26.153602.276@ux1.cso.uiuc.edu>, dorner@pequod.cso.uiuc.edu (Steve Dorner) writes: > > Multifinder is hell for developers, though, because a) we'd like to compile > in the background, and it just isn't practical and b) debugging apps without > protection can be very frustrating. I wouldn't exactly call it hell... I often compile in the background. I'm not sure what planet your from, but it sure seems to work for me. In fact, I'm writing this note with uAccess (and it's downloading some new News over the serial like too, yikes!) as I'm compiling a big MacApp based program in MPW. Sure feels like I'm getting two things done at once. Guess not though... There are solutions for the debugging apps without protection problem you mention too. My debugger of choice is Jasik's The Debugger and it has this handly little feature called - memory protection. Jeez, I though I couldn't do that. Harumph! -- michael And just wait until you get a look at ToolServer. Not only will it let you do background compiles and edit at the same time, you can ask that unused Mac down the hall to crunch some big build for you and have it politely let you know when it's done... (Sorry for the above flames, but I just get so tired of people telling me what I can't do with my Mac - when I'm already doing it...) -- Michael Peirce -- outpost!peirce@claris.com -- Peirce Software -- Suite 301, 719 Hibiscus Place -- Macintosh Programming -- San Jose, California 95117 -- & Consulting -- (408) 244-6554, AppleLink: PEIRCE
brendan@cs.uq.oz.au (Brendan Mahony) (03/27/91)
In <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >This makes me squirm a bit that Apple says MF is multitasking. At best, MF >is multiapplication. Try bringing up a dialog and watch all background >processes come to a hault. Try pulling down a menu bar (even under Sys7) >and watch everything stop. Surprisingly enough I get similar behaviour from a Sun 3/60 running Unix under Suntools. Pop up a menu and watch all your screen output freeze in midstride. Multitasking on the Mac gives the user control over what is using the CPU. This is important since the machine is only just providing a reasonable responsiveness as it is. Obviously if more cycles existed it would be nice to have a pre-emptive system, but I would rather have a Macintosh. -- Brendan Mahony | brendan@batserver.cs.uq.oz Department of Computer Science | heretic: someone who disgrees with you University of Queensland | about something neither of you knows Australia | anything about.
Aron_Fingers_Nelson@cup.portal.com (03/27/91)
I know this doesn't make you feel any better but for us musicians in the Mac world, using MIDI manager allows multiple applications to continue in the background even when menus are pulled down. When applications are written taking advantage of MIDI manager and multifinder "friendly" techniques, at least the simulation of multi-tasking occurs. An example: I can record a piece of music in my sequencer (which is in the background) while sending patches to my synthesizers, with another program all while switching in and out of my word processor. aron_nelson@cup.portal.com
gwills@maths.tcd.ie (Graham Wills) (03/27/91)
In article <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >This makes me squirm a bit that Apple says MF is multitasking. At best, MF >is multiapplication. Reading uninformed peoples' comments who know nothing, but pontificate freely makes me squirm >Try bringing up a dialog and watch all background >processes come to a hault. No they don't. Only with badly written programs. All MacApp-using programs don't do this, for example. >During a download, pull up a dialog from another >application and watch the download come to a hault and eventually cancel out. I have done this many times. It just ain't so for well written programs. Don't criticise a good OS. criticise bad programmers >While an application has a progress dialog up, try putting it in the >background and pull up something else. Why? Is it supposed to turn me on or something? All that happens is that I keep typing my Word file and the progress bar keeps on filling until it's full. It then uses the notification manager to tell me it's finished (just in case I obscured the window). >but is MF really all that cooperative? Only if the developer >makes it friendly? I don't think so, even if the developer uses modal dialogs, >he/she has no control of allowing background tasking to occur (without doing >some acrobatic tricks). If developpers write bad code, is that Apple's fault ? A modal dialog is exactly what it says it is. Modal. It stops everything. If you don't want to stop everything, DON'T USE MODAL DIALOGS. I mean it's not like they're particulary wonderful or necessary! >Then there is menus. Even under Sys7, pulling down a menu will hault all >background tasks. > This would be a problem for people with major disabilities, as if you need to take 5-10 seconds to choose a menu item, that could cause problems, but for most people? What's the problem with a 0.5 s pause? It's just the same as if another program was using the time. >Sys8 could compete with OS/2 and UNIX if it was capable of >preemptive multitasking. Picture the scene: A dealer's room i a major city... Customer: "I'm very impressed. The graphiocs are excellent, the range of packages are very good. It should take far less time to train our staff to use than UNIX. The printer facilities are so much better and I really liked the voice-mail system. There's just one thing.. I noticed that when the spread- sheet was recalculating and I was typing, the clock was still working in the background as well? Was that *pre-emptive* multitasking or not?" Dealer: "Wee, no, it was co-operative" Customer: "Well in that case I'll just have to buy a UNIX box" Yawnn. In the real world people buy computers to DO things, not to BE things It's only compu-nerds and techies who worry about such subtelties. What percentage of company purchasers do you think would worry at all about the diffrerence between co-op and pre-em? It's only the developpers who get a rougher time (unless you use MacApp, of course) > >I'd like to hear unflammable comments... > There you go.
freek@fwi.uva.nl (Freek Wiedijk) (03/28/91)
The mac is not multitasking iff it is showing a watch cursor. Question: Did you ever see a watch cursor on the mac. If so: that mac was not multitasking at that moment. BTW: did you know that the watch cursor contains a HUGE bug: either it indicates nine o'clock (which it isn't), or it spins like crazy. Why can't it show the proper time? Freek "the Pistol Major" Wiedijk E-mail: freek@fwi.uva.nl #P:+/ = #+/P?*+/ = i<<*+/P?*+/ = +/i<<**P?*+/ = +/(i<<*P?)*+/ = +/+/(i<<*P?)**
baumgart@esquire.dpw.com (Steve Baumgarten) (03/28/91)
In article <91085.161852EHYOUNK@MTUS5.BITNET> EHYOUNK@MTUS5.BITNET writes:
I completly agree with the statement that Mac OS should have
prementive multitasking.. Multifinder is a joke, they is no way to
download from a modem while running another app. Here are some
other problems.
It would be nice if people at least *tried* things before posting. Of
course you can download and work on something else at the same time.
I do it all the time. I also have a Gifwatcher window open (it
displays a GIF file as it's being downloaded), and sometimes I even
run MacCompress in the background to decompress a bunch of text files.
All works as expected.
Would you mind telling us which programs you're talking about? All
the commercial comm programs work in the background; ZTerm (shareware)
does; I can't think of any others right now. Probably there's one out
there. If it doesn't work in the background, it's broken, and I
wouldn't use it.
1. There is no way to set a task's inportance, each other the
backgroud app's get equal time, except for the foreground app,
witch can steel all of the cpu time if it wants too.
I like this. It means my word processor doesn't slow down just
because I'm running a compile or uncompressing a bunch of files. On
my Sun 4 at work (with 20 megs of memory), I usually run my compiles
on another machine, since my emacs session becomes very unresponsive
when anything major is going on in the background.
2. What about all those nice coprossers the fx has. When will we
see a system that truely supports them!
The ADB and serial coprocessors are already supported. Which were you
referring to? I hope you're not talking about multi-processing in
general -- there are few systems in the world that support
multi-processing well, and none are personal computers like the Mac.
3. When will we see the new print architechure, line-layout
manager, and apple scripting.
Good question. I suspect it's in the works, though it's a shame it
won't all be in System 7 as originally promised. But there are so
many good things in System 7 that I can understand why Apple wanted to
get it out the door and into people's hands as soon as possible.
I had heard that DDE in Windows 3.1 was going to work over networks;
now I hear that that's planned for a future release. Apple's
publish/subscribe manager works locally or across a net, whether or
not all the applications are running. When will we see that from
Microsoft?
--
Steve Baumgarten | "New York... when civilization falls apart,
Davis Polk & Wardwell | remember, we were way ahead of you."
baumgart@esquire.dpw.com |
cmcl2!esquire!baumgart | - David Lettermanminich@unx2.ucc.okstate.edu (Robert Minich) (03/28/91)
by EHYOUNK@MTUS5.BITNET: | I completly [sic] agree with the statement that Mac OS should have | prementive [sic] multitasking.. Multifinder is a joke, they is no | way to download from a modem while running another app. Here are some | other problems. I download things in the background all the time. It works very nicely, thank you. Downloads, however, aren't very cpu intensive, so this is one of the easiest things to do in the background. What software are you using??? Multifinder isn't perfect. Far from it. For instance, why can't applications be switched out during IO? I have some documentation to Switcher where many of the same limitations exist, but at least they had the purpose of keeping separate apps that never expected to run with less than the entire machine under its control. | 1. There is no way to set a task's inportance [sic], each other the | background app's get equal time, except for the foreground app, witch can ^^^^^ | steel all of the cpu time if it wants too. Cute. And somewhat true. Any app that gets CPU time can steel it, though. Think about it: who's going to stop you besides irate users? | 2. What about all those nice coprossers [sic] the fx has. When will we | see a system that truly supports them! I believe that would be A/UX. It is a cruel joke, indeed, to even mention the IOPs as a sign of Apple prowess when you can't use them without a big, expensive OS upgrade. It reminds me of arguments in IBM PC land about all that power available... for DOS... <chuckle-snort> | 3. When will we see the new print architecture, line-layout manager, and | apple scripting. The first two were what made Sys 7 sound great to me. Finally laying the old printing method, designed for a machine that was pushed very hard just to get something on an ImageWriter (RAM constraints), seemed like a logical idea. Unfortunately printing will apparently continue to be a weaker spot of the Mac than it should. System wide scripting should probably wait until apps get a hold of IAC and stability sets in. How about Sys 7.3? Or will we see 7.0.7 followed by 8.0.0? Thought for the day: release System 6.1.0 with some of Sys 7's functionality but not all the extra weight. -------------------------------------------------------------------------------+ | I used to think the apple was a little behind in hardware, and ahead | of the game in system software. Now I think the opposite might be true. | I sure hope that apple gets these into system 8, and I hope system 8 | doesn't take 2 years to get here. My next computer purchase may depend | on apple coming out with system 8 on time. | | Ed Younk | Michigan Tech University The Mac's system software is still very good. We may complain about cooperative multitasking and unprotected memory, but those are not items that are easily plopped on top of a lot of history and fiendishly wonderful hacks to keep older software going. (Witness the MultiFinder special case for putting MickeySoft Excel in the first meg of memory.) Protected and virtual memory don't work with the 68000 machines (I still think a 68010 would be a wonderful compromise.) Ed didn't specify exactly "game" Apple was ahead of, so I can only say that I don't of any platform that is ahead in everything. UNIX is both ahead of and behind Apple. Windows is sort of around the neighborhood. UNIX workstations are way ahead in low level OS design. Big IBM iron is just plain confusing -- a mix of ancient history with new hacks. Part of what gets me is that Apple can release basically the same machine as the True (tm) classic Mac (the 128K) and call it something wonderfully new. I sort of wish Apple would take the NeXT style and sell Macs at prices that makes people ask why they'd pay _more_ for "equivalent" PC hardware. A lot more market share with a better class of machines than the Classic would cause a lot less pain than ramming your head against hardware limitations in the future. I imagine software developers wouldn't mind a bit, either. -- |_ /| | Robert Minich | |\'o.O' | Oklahoma State University| "I'm not discouraging others from using |=(___)= | minich@d.cs.okstate.edu | their power of the pen, but mine will | U | - "Ackphtth" | continue to do the crossword." M. Ho
jcav@quads.uchicago.edu (john cavallino) (03/28/91)
In article <BAUMGART.91Mar27172632@info7.esquire.dpw.com> baumgart@esquire.dpw.com (Steve Baumgarten) writes: >I had heard that DDE in Windows 3.1 was going to work over networks; >now I hear that that's planned for a future release. Apple's >publish/subscribe manager works locally or across a net, whether or >not all the applications are running. When will we see that from >Microsoft? Didn't Apple also bump networkable IAC to a future release? (Not trying to flame, just curious) -- John Cavallino | EMail: jcav@midway.uchicago.edu University of Chicago Hospitals | USMail: 5841 S. Maryland Ave, Box 145 Office of Facilities Management | Chicago, IL 60637 "Opinions, my boy. Just opinions" | Telephone: 312-702-6900
dwal@ellis.uchicago.edu (David Walton) (03/28/91)
In article <1991Mar27.234212.20128@midway.uchicago.edu> jcav@quads.uchicago.edu (john cavallino) writes: >In article <BAUMGART.91Mar27172632@info7.esquire.dpw.com> baumgart@esquire.dpw.com (Steve Baumgarten) writes: >>I had heard that DDE in Windows 3.1 was going to work over networks; >>now I hear that that's planned for a future release. Apple's >>publish/subscribe manager works locally or across a net, whether or >>not all the applications are running. When will we see that from >>Microsoft? > >Didn't Apple also bump networkable IAC to a future release? (Not trying to >flame, just curious) No, not as far as I know. IAC works transparently over a network, unless the programs that used the Edition Manager and the PPC Toolbox in the beta four release were canned demos...:-). ...not that I'm ruling this out, mind you...:-) >John Cavallino | EMail: jcav@midway.uchicago.edu Hi, John! -- David Walton Internet: dwal@midway.uchicago.edu University of Chicago { Any opinions found herein are mine, not } Computing Organizations { those of my employers (or anybody else). }
peirce@outpost.UUCP (Michael Peirce) (03/28/91)
In article <91085.161852EHYOUNK@MTUS5.BITNET>, EHYOUNK@MTUS5.BITNET writes: > > I completly agree with the statement that Mac OS should have prementive > multitasking.. Multifinder is a joke, they is no way to download from a modem > while running another app. Here are some other problems. What comm program are you using? All the ones I've used allow this without any trouble at all. Even those who don't like MultiFinder say things like "besides downloading, what's it good for..." If your comm program doesn't do this switch to one that does, there are many to choose from. > 2. What about all those nice coprossers the fx has. When will we see a system > that truely supports them! Those coprocessors are used. The serial port coprocessor, and the ADB bus coprocessor, even the graphics coprocessor if you have an 8-24 GC card. They have't got the DMA going yet, though, that's true. (A/UX does though!) > 3. When will we see the new print architechure, line-layout manager, and apple > scripting. This falls into the category of promising too much early on. Once they were well into the system 7 development they noticed that it might be 1993 before they finish the whole thing. So they pruned some of the features and focused on what they thought most important (and most under control, software development-wise!). I expect we'll see these features in a follow on system 7 (7.0.X or 7.1?). > I used to think the apple was a little behind in hardware, and ahead of > the game in system software. Now I think the opposite might be true. I sure > hope that apple gets these into system 8, and I hope system 8 doesn't take > 2 years to get here. My next computer purchase may depend on apple coming out > with system 8 on time. Considering the size and complexity of system 7, they're actually doing a pretty good job. Any LARGE software effort is extremely hard to manage effectively. Personally, I'm pleased to see them focus on key features and delivery it, then move on to the rest afterward. They're doing a VERY nice job of involving developers with System 7 early enough that there should be fix'em versions of almost everything that breaks released concurrently with system 7 or close to it. -- michael -- Michael Peirce -- outpost!peirce@claris.com -- Peirce Software -- Suite 301, 719 Hibiscus Place -- Macintosh Programming -- San Jose, California 95117 -- & Consulting -- (408) 244-6554, AppleLink: PEIRCE
bskendig@set.Princeton.EDU (Brian Kendig) (03/28/91)
In article <1991Mar27.163059.27061@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes: >The mac is not multitasking iff it is showing a watch cursor. >Question: Did you ever see a watch cursor on the mac. >If so: that mac was not multitasking at that moment. Oh, puh-LEAZE. You can have a loop in your program that does some tight calculations. You can set the cursor to be a stopwatch. Then when you call either "GetNextEvent(); SystemTask();" or "WaitNextEvent();" in your loop, the Mac operating system will be given a slice of time to do things like update windows (such as the alarm clock), do some work in other programs (such as unstuffing with Stuffit in the background), process AppleTalk messages, and so on. This argument over multitasking is getting really ludicrous, now. Do you want to open up MacWrite II, SuperPaint, HyperCard, and a few other applications, then have your work in Microsoft Excel be slowed to a crawl as your Macintosh takes time out every few milliseconds to make sure MacWrite II hasn't suddenly developed a need to do something? Preemptive multitasking means that the more applications you have open, the slower your machine will go; the benefit, of course, is that you can have things go on in the background. The way the Macintosh is designed, you can only really work in one application at a time (which isn't a problem, because very few people if any can type a report and enter figures into a spreadsheet simultaneously). Wouldn't you rather have the system devote all its time to the program you're working in right now? Okay, so there are a few times when it woud be nice to have processes run in the background: when you're downloading files, for instance, or unstuffing lots of archives, or so on. It's not impossible -- it's already been done! And System 7.0 will bring it even further by letting you work in other applications while the Finder performs long tasks such as copying files, by introducing semimodal dialogs. >BTW: did you know that the watch cursor contains a HUGE bug: >either it indicates nine o'clock (which it isn't), or it spins like crazy. >Why can't it show the proper time? I figure you forgot to put a smiley in there. << Brian >> | Brian S. Kendig \ Macintosh | Engineering, | bskendig | | Computer Engineering |\ Thought | USS Enterprise | @phoenix.Princeton.EDU | Princeton University |_\ Police | -= NCC-1701-D =- | @PUCC.BITNET | "It's not that I don't HAVE the work to *do* -- I don't DO the work I *have*."
dswt@stl.stc.co.uk (D.S.W.Tansley) (03/28/91)
In the referenced article wilkins@jarthur.Claremont.EDU (Mark Wilkins) writes: >In article <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >>This makes me squirm a bit that Apple says MF is multitasking. At best, MF >>is multiapplication. Try bringing up a dialog and watch all background >>processes come to a hault. > > Background processes keep going when a dialog is up, except in some very >rare cases where the dialog doesn't check for user input. Otherwise they >proceed as usual. <Loads more criticisms countered by intelligent & graceful responses> >>I'd like to hear unflammable comments... > > I hope I didn't seem harsh. In any case, you're not alone in your views >and I'm sure the engineers at Apple know this. > >>Bob Daniel > >-- Mark Wilkins Mark, you're a bloody hero mate! That's the best & sanest response I've seen to an anti-cooperative multitasking philistine yet! :-) One more point to Bob: I'm a Mac user, not a Mac programmer, and when I press the mouse button in the menu bar, I expect the computer to respond to my command as fast as possible -- I'm running the computer, not the computer me. With current h/w, we don't have the MIPS to handle the beautiful responsiveness I get from key parts of the interface like this. (But yes, it could be a lot more responsive still in others!) Other, non-cooperative multitasking systems I've used (Sun3, 4, Windows 3) are porridge in comparison *in this respect*. One day, processors will be fast enough. Until then, Cooperative multitasking is a good design decision for the class of *users* the Mac is designed for, IMHO. =========================================================================== Stewart Tansley | STC Technology Ltd | 'Be cool, or be | London Rd, Harlow, CM17 9NA, UK | cast out...' dswt@stl.stc.co.uk | +44 279 429531 x2763 | Subdivisions, Rush =========================================================================== 'You know how that rabbit feels - going under your spinning wheels...' ===========================================================================
peirce@outpost.UUCP (Michael Peirce) (03/28/91)
In article <1991Mar27.163059.27061@fwi.uva.nl>, freek@fwi.uva.nl (Freek Wiedijk) writes: > > The mac is not multitasking iff it is showing a watch cursor. > > Question: Did you ever see a watch cursor on the mac. > If so: that mac was not multitasking at that moment. Not necessarily true. I've writen programs (used by many people out there) that displays a watch cursor and still calls WaitNextEvent or system task. Background applications are still running and receiving time. In general, there is no reason why a watch cursor should mean no multitasking, although I'm sure there are some applications that work that way. (of course there are those who claim that the Mac isn't ever MultiTasking...) -- michael -- Michael Peirce -- outpost!peirce@claris.com -- Peirce Software -- Suite 301, 719 Hibiscus Place -- Macintosh Programming -- San Jose, California 95117 -- & Consulting -- (408) 244-6554, AppleLink: PEIRCE
cs483106@umbc5.umbc.edu (cs483106) (03/29/91)
In article <91085.161852EHYOUNK@MTUS5.BITNET> EHYOUNK@MTUS5.BITNET writes: >I completly agree with the statement that Mac OS should have prementive >multitasking.. Multifinder is a joke, they is no way to download from a modem >while running another app. Here are some other problems. What do you mean? I do it all the time. Run my compiler or MacWrite or something in the foreground while Zterm is going to it in the background. works great for me, anyway... >1. There is no way to set a task's inportance, each other the backgroud app's >get equal time, except for the foreground app, witch can steel all of the cpu >time if it wants too. true enough. But would you make a lot of use of a program that did this to you? I sure done. Programs written like that definitely don't get used much on my machine. Besides, if you get AU/X(I know, it's expensive), you get true multi-tasking in the bargin. >2. What about all those nice coprossers the fx has. When will we see a system >that truely supports them! >3. When will we see the new print architechure, line-layout manager, and apple >scripting. > I used to think the apple was a little behind in hardware, and ahead of >the game in system software. Now I think the opposite might be true. I sure >hope that apple gets these into system 8, and I hope system 8 doesn't take >2 years to get here. My next computer purchase may depend on apple coming out >with system 8 on time. >Ed Younk >Michigan Tech University
freek@fwi.uva.nl (Freek Wiedijk) (03/29/91)
bskendig@set.Princeton.EDU (Brian Kendig) writes: >In article <1991Mar27.163059.27061@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes: >>The mac is not multitasking iff it is showing a watch cursor. >>Question: Did you ever see a watch cursor on the mac. >>If so: that mac was not multitasking at that moment. > >Oh, puh-LEAZE. > >You can have a loop in your program that does some tight calculations. >You can set the cursor to be a stopwatch. Then when you call either >"GetNextEvent(); SystemTask();" or "WaitNextEvent();" in your loop, >the Mac operating system will be given a slice of time to do things >like update windows (such as the alarm clock), do some work in other >programs (such as unstuffing with Stuffit in the background), process >AppleTalk messages, and so on. If you're thinking as a computer scientist you're right: the process scheduler is still active while the cursor is a watch (though I think that this won't be the case in REAL applications, which probably won't call Get/WaitNextEvent during lenghty operations). However, thinking as a USER, you're wrong. If I see a watch cursor I can not switch contexts (i.e. multitask); at least I would expect this not to be possible. I mean: under UNIX I can always say ^Z. But on the Mac, no. Now, don't get me wrong: I prefer the Mac to UNIX. >>BTW: did you know that the watch cursor contains a HUGE bug: >>either it indicates nine o'clock (which it isn't), or it spins like crazy. >>Why can't it show the proper time? > >I figure you forgot to put a smiley in there. No. Did you never see my INIT "Watch It!"? Freek "the Pistol Major" Wiedijk E-mail: freek@fwi.uva.nl #P:+/ = #+/P?*+/ = i<<*+/P?*+/ = +/i<<**P?*+/ = +/(i<<*P?)*+/ = +/+/(i<<*P?)**
EHYOUNK@MTUS5.BITNET (03/29/91)
I hate to waste bandwith with this post, but so many people have e-mailed me
with replies, I think it is jusified.
I guess the word "impossible" was a bad choise of words. Well over
15 people have e-mailed me saying that it is possible to download with a modem
in the background useing multifinder. Some, people have been as blunt as to
tell me that I don't know what I am talking about.
What I meant in my previous post was that it isn't impossible to download
in the background, it is just impractical. If all the people that sent me
e-mail are still in the dark, yet me explain.
When downloading in the backgroud, you even had these problems:
1. When you pull down a menu in the foreground app, does the modem transfer
come to a complete halt?
2. If you are doing a task in the foreground, "or in the background for that
matter", that requires alot of disk access, does your transfer come to a stop
or at least slow to a crawl?
3. Have you even had the remote computer timeout on you or abort the transfer
because of too many retries?
4. Have you ever had your effective transfer cut to about 1/2 of the normal
rate because your running the task in the background? "This one is a real
bummer if you are calling long distance!"
About a year and 1/2 ago, I purchased 4meg of ram for my CX. I said to
myself: Oh goodie, I can tranfer files, and do my other work at the same time.
After I lossed countless hours of download time, I gave up.
I agree that for most people multifinder is good enough. But multifinder
will not be enough (at least not enough for me), when I buy my next machine.
So to summerize, IN MY OPINION, doing a tranfer in the background on the
mac is impactical. I am sorry for all the confussion, that I caused. I will
more careful in the future!
Edward H. Younk
Computer Engineering Student
Michigan Tech Universitybaumgart@esquire.dpw.com (Steve Baumgarten) (03/29/91)
In article <1991Mar27.163059.27061@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes:
The mac is not multitasking iff it is showing a watch cursor.
Question: Did you ever see a watch cursor on the mac.
If so: that mac was not multitasking at that moment.
Again, it would be nice if people who post statements like these would
at least try them out. The watch cursor indicates that the foreground
application is doing something, and not ready to accept any more user
input at the moment. It doesn't mean anything else.
When GIFConverter opens and displays GIF files, it displays an
animated juggling cursor. That cursor means the same thing as the
watch cursor -- basically, you can't do anything in GIFConverter right
then, but you could switch to another application and do some other
work. Of course, the fact that it's doing very CPU- and
graphics-intensive work doesn't mean that your download stops, or that
anything else in the background stops.
That's because Multifinder allows the Mac to multitask. Whether or
not it's "real" multitasking, or "official" multitasking, or
"preemptive" multitasking, it *does* allow you to have several
programs all working simultaneously. Sounds like multitasking to me.
BTW: did you know that the watch cursor contains a HUGE bug:
either it indicates nine o'clock (which it isn't), or it spins like
crazy. Why can't it show the proper time?
Look at it this way: The watch cursor shows the correct time twice a
day.
--
Steve Baumgarten | "New York... when civilization falls apart,
Davis Polk & Wardwell | remember, we were way ahead of you."
baumgart@esquire.dpw.com |
cmcl2!esquire!baumgart | - David Lettermanbaumgart@esquire.dpw.com (Steve Baumgarten) (03/29/91)
In article <1991Mar27.225800.16664@unx2.ucc.okstate.edu> minich@unx2.ucc.okstate.edu (Robert Minich) writes:
The Mac's system software is still very good. We may complain about
cooperative multitasking and unprotected memory, but those are not
items that are easily plopped on top of a lot of history [...]
Also, they don't always buy you as much as you think. I've crashed my
Sun a number of times -- don't know how -- and more often, I've
crashed my window system, which is pretty much the same thing, since
you have to start everything all over again.
My Sun has all sorts of whiz-bang memory protection, but it doesn't
always seem to help as much as people think it should.
These things are nice frills, but they're hardly necessary for people
to get work done. (Unless those people are programmers, in which case
the harder it is to crash a machine the better. But most people
aren't programmers.)
--
Steve Baumgarten | "New York... when civilization falls apart,
Davis Polk & Wardwell | remember, we were way ahead of you."
baumgart@esquire.dpw.com |
cmcl2!esquire!baumgart | - David Lettermanlsr@Apple.COM (Larry Rosenstein) (03/29/91)
In article <1991Mar27.163059.27061@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes: >The mac is not multitasking iff it is showing a watch cursor. Strictly speaking, that's not true. The watch cursor indicates that the application isn't accepting input, but that doesn't mean it isn't yielding the CPU. The 7.0 Finder displays the watch cursor while copying files, yet other tasks continue to run, and you can switch to other applications. -- Larry Rosenstein, Object Specialist Apple Computer, Inc. 20525 Mariani Ave, MS 3-PK Cupertino, CA 95014 AppleLink:Rosenstein1 domain:lsr@Apple.COM UUCP:{sun,voder,nsc,decwrl}!apple!lsr
EHYOUNK@MTUS5.BITNET (03/29/91)
Hm, this is starting to get old......
I completly agree with the statement that Mac OS should have
prementive multitasking.. Multifinder is a joke, they is no way to
download from a modem while running another app. Here are some
other problems.
>It would be nice if people at least *tried* things before posting. Of
>course you can download and work on something else at the same
>time.
>I do it all the time. I also have a Gifwatcher window open (it
>displays a GIF file as it's being downloaded), and sometimes I even
>run MacCompress in the background to decompress a bunch of text
>files.
>All works as expected.
.....See my previous post on this subject.
1. There is no way to set a task's inportance, each other the
backgroud app's get equal time, except for the foreground app,
witch can steel all of the cpu time if it wants too.
>I like this. It means my word processor doesn't slow down just
>because I'm running a compile or uncompressing a bunch of files. On
>my Sun 4 at work (with 20 megs of memory), I usually run my
>compiles
>on another machine, since my emacs session becomes very
>unresponsive
>when anything major is going on in the background.
You complety miss my point. I never sugested that true
multitasking is going to be as fast as running a single proccess! At
least Unix gives you a choise in setting which task is getting the
most cpu time. In multifinder, a simple "bad app" can bring
multifinder to a standstill. I could write a little app that, if you ran
it in the foreground would bring your modem transfers to a dead
stop!
>The ADB and serial coprocessors are already supported. Which were
>you
>referring to? I hope you're not talking about multi-processing in
>general -- there are few systems in the world that support
>multi-processing well, and none are personal computers like the
>Mac.
How about Scsi DMA?
Oh BTW, I hope that the rumors that the '040 mac's will not support
scsi DMA are complete false.
Ed Younkjcav@quads.uchicago.edu (john cavallino) (03/29/91)
In article <1991Mar27.234212.20128@midway.uchicago.edu> jcav@quads.uchicago.edu (john cavallino) writes: >In article <BAUMGART.91Mar27172632@info7.esquire.dpw.com> baumgart@esquire.dpw.com (Steve Baumgarten) writes: >>I had heard that DDE in Windows 3.1 was going to work over networks; >>now I hear that that's planned for a future release. Apple's >>publish/subscribe manager works locally or across a net, whether or >>not all the applications are running. When will we see that from >>Microsoft? > >Didn't Apple also bump networkable IAC to a future release? (Not trying to >flame, just curious) I have been variously informed (including mail from Larry Rosenstein at Apple) that AppleEvents are fully network-transparent, as is the underlying protocol. :-) Sorry about the error. -- John Cavallino | EMail: jcav@midway.uchicago.edu University of Chicago Hospitals | USMail: 5841 S. Maryland Ave, Box 145 Office of Facilities Management | Chicago, IL 60637 "Opinions, my boy. Just opinions" | Telephone: 312-702-6900
jstewart@rodan.acs.syr.edu (Ace Stewart) (03/29/91)
>Would you mind telling us which programs you're talking about? All >the commercial comm programs work in the background; ZTerm (shareware) >does; I can't think of any others right now. Probably there's one out >there. If it doesn't work in the background, it's broken, and I >wouldn't use it. I may not be using the current version, but my version of Versaterm Pro does not allow background downloading. If that is the case, use Zterm...it does exactly what you need anyway. --Ace -- Ace Stewart | Affiliation: Eastman Kodak Company, Rochester, New York jstewart@rodan.acs.syr.edu jstewart@sunrise.bitnet jstewart@mothra.cns.syr.edu jstewart@sunspot.cns.syr.edu ace@suvm.bitnet rsjns@suvm.bitnet
umh@vax5.cit.cornell.edu (03/29/91)
In article <91085.161852EHYOUNK@MTUS5.BITNET>, EHYOUNK@MTUS5.BITNET writes: > I completly agree with the statement that Mac OS should have prementive > multitasking.. Multifinder is a joke, they is no way to download from a modem > while running another app. Here are some other problems. This is not *quite* true. You can do it if you're very careful *about* what you do. I once downloaded in background, crashed an app in foreground, went to Macsbug and issued an es, and the downloading did no hiccup- blew my mind. But in general it is a real pain having to walk on eggshells while your download goes ahead. Maynard Handley
dorner@pequod.cso.uiuc.edu (Steve Dorner) (03/29/91)
In article <1991Mar28.163655.6496@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes: >that this won't be the case in REAL applications, which probably won't >call Get/WaitNextEvent during lenghty operations). My application does; I like to think it's "REAL". I even allow switches practically all the time. -- Steve Dorner, U of Illinois Computing Services Office Internet: s-dorner@uiuc.edu UUCP: uunet!uiucuxc!uiuc.edu!s-dorner
baumgart@esquire.dpw.com (Steve Baumgarten) (03/30/91)
In article <1991Mar29.013650.21345@rodan.acs.syr.edu> jstewart@rodan.acs.syr.edu (Ace Stewart) writes:
I may not be using the current version, but my version of Versaterm
Pro does not allow background downloading. If that is the case, use
Zterm...it does exactly what you need anyway.
You're using a very old version then, since for as far back as I can
remember, Versaterm worked splendidly in the background.
--
Steve Baumgarten | "New York... when civilization falls apart,
Davis Polk & Wardwell | remember, we were way ahead of you."
baumgart@esquire.dpw.com |
cmcl2!esquire!baumgart | - David Lettermanirwin@aylmer.uchicago.edu (Mark Irwin) (03/30/91)
In article <1991Mar29.013650.21345@rodan.acs.syr.edu> jstewart@rodan.acs.syr.edu (Ace Stewart) writes: > >>Would you mind telling us which programs you're talking about? All >>the commercial comm programs work in the background; ZTerm (shareware) >>does; I can't think of any others right now. Probably there's one out >>there. If it doesn't work in the background, it's broken, and I >>wouldn't use it. > > >I may not be using the current version, but my version of Versaterm >Pro does not allow background downloading. If that is the case, use >Zterm...it does exactly what you need anyway. > >--Ace >-- > Ace Stewart | Affiliation: Eastman Kodak Company, Rochester, New York >jstewart@rodan.acs.syr.edu jstewart@sunrise.bitnet jstewart@mothra.cns.syr.edu > jstewart@sunspot.cns.syr.edu ace@suvm.bitnet rsjns@suvm.bitnet Ace definitely isn't using the current version of VersaTerm Pro. Version 3.1 definitely supports background file transfers and I know some versions before this did as well. I only wish that it would support multiple windows like MacLayers. Does anybody know if version 3.5 is out yet. It appears that it is but I haven't received an upgrade notice yet. Mark Mark Irwin Dept of Statistics, Univ of Chicago irwin@galton.uchicago.edu mark@stat.ubc.ca
klingspo@mozart.cs.colostate.edu (Steve Klingsporn) (03/31/91)
In article <593@genco.bungi.com> rad@genco.bungi.com (Bob Daniel) writes: >Try bringing up a dialog and watch all background >processes come to a hault. Try pulling down a menu bar (even under Sys7) >and watch everything stop. During a download, pull up a dialog from another >application and watch the download come to a hault and eventually cancel out. >While an application has a progress dialog up, try putting it in the >background and pull up something else. When you bring up a dialog, if things are programmed correctly, other background applications DO NOT come to a halt, but rather run as usual. This includes any programs that are downloading, drawing to the screen, what have you. When you pull down the menuBar, app's in the background could conceivably draw over the menuBar. It's possible to write your own menu handling routines, and keep processes running -- even different threads in your own program. I've downloaded MANY things before with dialogs on-screen. What kind of applications are you running? Up-to-date ones? I believe there is something called "GetNextEvent" or "WaitNextEvent" that you should call from your application to give other applications the time they deserve. You'll want to note that people who say "The Amiga is a true multitasking system" also suffer from the fact that when a menu is down (or even the mouse button, sometimes), things seemingly come to a halt. >Alot of this has been improved in Sys7 with "semi-modal" dialogs but does Apple >really think the current MF is really Multitasking? OK, so it's "cooperative- >multitasking", but is MF really all that cooperative? Only if the developer >makes it friendly? I don't think so, even if the developer uses modal dialogs, >he/she has no control of allowing background tasking to occur (without doing >some acrobatic tricks). Even under Sys7, he/she would have to convert ALL >modal dialogs to "semi-modal" dialogs to allow background *tasks* to continue. >Then there is menus. Even under Sys7, pulling down a menu will hault all >background tasks. Big deal. Write your own menu code. I've yet to see anything in unix/X-Windows that is all that impressive. Nor, do I think NeXT has that impressie of software, so why are you complaining? >I like MF alot and I like Sys7 a whole lot but it bothers me that Apple >would consider MF to be multitasking (even if you consider it to be, it >isn't very cooperative). Making a statement like that makes it appear >that they don't need to improve the multitaskiness of the Mac. > The first Macintosh applications weren't designed to multitask. It would take a gross System overhaul to impliment "true" multitasking on the Macintosh. Besides, how many things can YOU do at once? The Macintosh does a very good job of handling multiple processes. I love it. There really isn't a problem, as I see it, with the MacOS' Process Manager (7.0). >Maybe it's just that I'm spoiled with UNIX, but that statement does make >me squirm. I hope Sys8 will address the issues I've mentioned. I feel >it should not be up to the developer to be responsible for the >"multitaskiness" and that it should be controlled from the operating >system. Sys8 could compete with OS/2 and UNIX if it was capable of >preemptive multitasking. > OS/2 is crap. Unix is primitive. I guess it's one of the better command-oriented operating systems I've seen, yet it's quite unfriendly. Using a Macintosh is a learning experience & seems quite easy to figure out, no matter which application(s) you are running. Unix isn't. In fact, without documentation, the only real way to find out commands is to do 'help <cmd> or 'man <cmd> and see what commands are cross-referenced in the end. Plus, Man kinda sucks. Overall, Macintosh is the clear-cut advantage. A/UX has "True" Multitasking. It can be done on a Mac, bu t I don't think it's all that necessary. >I'd like to hear unflammable comments... <ignition> >moof moof, >Bob Daniel >rad@genco.uucp Moof!!!!!
Jim.Spencer@p510.f22.n282.z1.fidonet.org (Jim Spencer) (03/31/91)
Steve Dorner writes in a message to All
In article <1991Mar28.163655.6496@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes: >that this won't be the case in REAL applications, which probably won't >call Get/WaitNextEvent during lenghty operations).
SD> My application does; I like to think it's "REAL". I even allow
SD> switches practically all the time.
Yea, it strikes me as pretty unfriendly for a programmer to start a long process without checking every so often to see if the user wants to bail out and this check is all that is necessary to give to time to other tasks.
jjwcmp@isc.rit.edu (Jeff Wasilko) (04/02/91)
With all this talk about multitasking (it seems like a never ending subject), I thought about the places where I could really use the ability to switch from one task to another: 1. Printing While spoolers do help, when you're printing a 40 page 11x17 book with lots of scanned images, it can still take a while to generate the postscript. 2. Saving Wouldn't it be nice to switch to another program while your 1.5 megabyte file is being saved to your fileserver? Recently, I was was doing batch search and replace with Nisus. The search and replace was prety fast, but it took quite a while to save all 150 files afterward... 3. Copying files Well, system 7 will take care of this... 4. Starting an App If Apple could look at where most users spend most of the time waiting (and unable to switch), they could concentrate on adding some sort of switching ability there... Jeff -- | RIT VAX/VMS Systems: | Jeff Wasilko | RIT Ultrix Systems: | |BITNET: jjwcmp@ritvax +----------------------+ INET:jjwcmp@ultb.isc.rit.edu| |INTERNET: jjwcmp@ritvax.rit.edu |____UUCP:jjwcmp@ultb.UUCP____| |Ask me about the Desktop Publishing Mailing list -- All platforms welcome. |
n67786@cc.tut.fi (Tero Nieminen) (04/02/91)
In article <1991Mar28.163655.6496@fwi.uva.nl> freek@fwi.uva.nl (Freek Wiedijk) writes:
If you're thinking as a computer scientist you're right: the process
scheduler is still active while the cursor is a watch (though I think
that this won't be the case in REAL applications, which probably won't
call Get/WaitNextEvent during lenghty operations).
IF there were such a thing as a process scheduler thing would be a lot
easier, but unfortunately there isn't one. Applications just get called
and they are expected to return in resonable time not to make too much
slag.
However, thinking as a USER, you're wrong. If I see a watch cursor
I can not switch contexts (i.e. multitask); at least I would expect
this not to be possible.
Believe me, it can and has been done and it works. (done it myself for
one :) ).
I mean: under UNIX I can always say ^Z. But on the Mac, no. Now,
don't get me wrong: I prefer the Mac to UNIX.
You should be able to switch by clicking the small aplication icon at
the reight end ot the menubar. (But what do you do if it,s not
visible cause the screen is too small and there are too many menus).
--
Tero Nieminen Tampere University of Technology
n67786@cc.tut.fi Tampere, Finland, EuropeAdam.Frix@p18.f20.n226.z1.FIDONET.ORG (Adam Frix) (04/04/91)
freek@fwi.uva.nl (Freek Wiedijk) writes:
FW> BTW: did you know that the watch cursor contains a HUGE bug:
FW> either it indicates nine o'clock (which it isn't), or it spins
FW> like crazy. Why can't it show the proper time?
That's funny. My watch cursor doesn't show that bug at all. Mine shows a quarter to twelve.
--Adam--
--
Adam Frix via cmhGate - Net 226 fido<=>uucp gateway Col, OH
UUCP: ...!osu-cis!n8emr!cmhgate!20.18!Adam.Frix
INET: Adam.Frix@p18.f20.n226.z1.FIDONET.ORGdtiberio@eeserv1.ic.sunysb.edu (David Tiberio) (04/13/91)
In article <1991Apr2.024426.28729@isc.rit.edu> jjwcmp@isc.rit.edu (Jeff Wasilko) writes: > >With all this talk about multitasking (it seems like a never ending >subject), I thought about the places where I could really use the >ability to switch from one task to another: > If you really want to learn a lot about multitasking, you should try to find someone who has a true multitasking computer, annd then compare it to thge multifinder or system 7. Windows 3.0 for the IBM is nnot that great at multitasking, but GEOS for the IBM does a good job. If you want to see some machines that have multitasking perfected, check out an Amiga or a UNIX workstation. It is hard to understand multitasking just by looking at the multifinbder. In fact there are many limitations on the Mac that need to be overcome before true multitasking. For example, open up a terminal program, such as ZTerm. Then start sending a big file. Next, open a DA. The terminal will stop sending. In fact, even by selecting a menu or the title bar of a window you will freeze the terminal sending process. This is not good. Another major problem is memory. Mac programs assume that most of the system memory will be free. When it is multitasking, it will want a big chunk of memory oor it won't work. It may also damage the memory of another program. On an Amiga or a UNIX system, this will never happen. A 10k application will only use about 10k. Therefor, if you have 1 meg, you can run that application 100 times! I anxiously await to see System 7. I only hope it will run on my computer...otherwise I will have to stay with 6.05. > >Jeff >-- -- David Tiberio SUNY Stony Brook 2-3481 AMIGA DDD-MEN Tomas Arce Any students from SUNY Oswego? Please let me know! :) Un ragazzo di Casalbordino, Italia.
phaedrus@milton.u.washington.edu (Mark Phaedrus) (04/13/91)
In article <1991Apr13.014000.29394@sbcs.sunysb.edu> dtiberio@eeserv1.ic.sunysb.edu (David Tiberio) writes: > If you really want to learn a lot about multitasking, you should try to >find someone who has a true multitasking computer, annd then compare it to >thge multifinder or system 7. Oh, great, here goes the "What is true multitasking?" flamewar again. I'm not even going to touch this one... > For example, open up a terminal program, such as ZTerm. Then start sending >a big file. Next, open a DA. The terminal will stop sending. In fact, even >by selecting a menu or the title bar of a window you will freeze the terminal >sending process. This is not good. Given any computer system anywhere, I'll wager any amount of money that I can point out a lousy program that runs on it. :) Just because one program has trouble coping with MultiFinder does not mean that there's something terminally wrong with MultiFinder. To cite a counterexample to your example, I've been downloading programs from this mainframe to my Mac all night long, and "pipelining" the process of assembling them. That is, having White Knight downloading files from the mainframe using ZMODEM (at 2400 baud, at 95+% efficiency), using the Vantage DA to preserve interesting stuff from the headers and paste the occasional multi-part file together, using StuffIt or Compactor to unbinhex and unpack the files, and then running them to see what they do. Several times, I had White Knight downloading, StuffIt and Compactor each working on a file while I tried out a new stack in HyperCard 2, and I never had a single problem with the downloads. This certainly seems like multitasking to me... -- Internet: phaedrus@u.washington.edu (University of Washington, Seattle) The views expressed here are not those of this station or its management. "If you can keep your head while those about you are losing theirs,
oleary@ux.acs.umn.edu (Doc O'Leary) (04/14/91)
In article <1991Apr13.014000.29394@sbcs.sunysb.edu> dtiberio@eeserv1.ic.sunysb.edu (David Tiberio) writes: > For example, open up a terminal program, such as ZTerm. Then start sending >a big file. Next, open a DA. The terminal will stop sending. In fact, even >by selecting a menu or the title bar of a window you will freeze the terminal >sending process. This is not good. I just downloaded the new TrueType fonts with ZTerm in the background without a hitch. I've done background downloading in System 6.0.3 - 7.0b4 without any problems. Are you using the current version? Something tells me Dave doesn't even own (perhaps never used) a Mac. Not a flame, but I wish that kind of misinformation didn't get posted. While any experienced Mac user will know it to be false, there are some new-comers or potential new-comers that will believe it, wasting their time waiting for a download to complete and having, perhaps spreading, a misconception of what a Macintosh can and can't do. --------- Doc ********************** Signature Block : Version 2.4 ********************* * | "Please put litter in its place" * * "Was it love, or was it the idea | ---McDonald's packaging * * of being in love?" -- PF | Wouldn't that be on the ground? * * (BTW, which one *is* Pink?) | * * | --->oleary@ux.acs.umn.edu<--- * ****************** Copyright (c) 1991 by Doc O'Leary ********************
rcook@grumpy.helios.nd.edu (04/15/91)
In article <1991Apr13.014000.29394@sbcs.sunysb.edu>, dtiberio@eeserv1.ic.sunysb.edu (David Tiberio) writes: |> In article <1991Apr2.024426.28729@isc.rit.edu> jjwcmp@isc.rit.edu (Jeff Wasilko) writes: |> > |> Another major problem is memory. Mac programs assume that most of the |> system memory will be free. When it is multitasking, it will want a big |> chunk of memory oor it won't work. It may also damage the memory of another |> program. On an Amiga or a UNIX system, this will never happen. A 10k |> application will only use about 10k. Therefor, if you have 1 meg, you can |> run that application 100 times! |> On a true multitasking system like OS/2 and unix, a 10K program running 100 times should take up 10K, plus some overhead for each program to hold registers and temporary data. Assuming 1K overhead for each program (not a bad assumption for a 10K program) then you could run this program 999 times on a 1000K system. Robert Kelley Cook U. Of Notre Dame '91 I'm still open to job suggestions . . .
kblackne@mcs.drexel.edu (Ken Blackney) (04/15/91)
In article <1991Apr15.055431.7509@news.nd.edu> rcook@grumpy.helios.nd.edu () writes: > >On a true multitasking system like OS/2 and unix, a 10K program running >100 times should take up 10K, plus some overhead for each program to hold >registers and temporary data. Assuming 1K overhead for each program (not a >bad assumption for a 10K program) then you could run this program 999 times on >a 1000K system. > >Robert Kelley Cook >U. Of Notre Dame '91 > I'm still open to job suggestions . . . With all due respect, the issue you are discussing is a memory manager issue, _not_ a multitasking issue. Windows 3.0 uses a cooperative multitasking sys- tem much like the Mac and it offers "code sharing" as you describe. I would guess that the reason the Mac does not do this now is to maintain some sort of backward compatibility with older apps. Ken Blackney Computing Resource Group Office of Computing Services Drexel University
dweisman@umiami.ir.miami.edu (Ordinary Man) (04/15/91)
In article <1991Apr13.014000.29394@sbcs.sunysb.edu>, dtiberio@eeserv1.ic.sunysb.edu (David Tiberio) writes: > For example, open up a terminal program, such as ZTerm. Then start sending > a big file. Next, open a DA. The terminal will stop sending. In fact, even > by selecting a menu or the title bar of a window you will freeze the terminal > sending process. This is not good. OK, you get a few points for this but not full credit :). It *is* true that pulling AND HOLDING down a menu while downloading will pause your send/receieve *BUT* if the modem started sending a 1K block before you pulled down the menu it will continue to send WHILE YOU HOLD IT DOWN. It will then stop when it finishes the block and you're stilling holding the menu down. Since most people don't need to hold a menu down for any length of time, and the chances of pulling one down right before the next block is sent is usually rare but does happen, you can download a file in the background and work on somthing else without slowing down the download in the least bit. I've done this...I know. This also applies to ANY action that freezes background processing like dragging a window. Let go of the window/menu, and the download starts right up again. > I anxiously await to see System 7. I only hope it will run on my > computer...otherwise I will have to stay with 6.05. Well, if you have 2 Megs and a Plus or higher and a hard drive it will run, and you will like it, really, you will. :-D Dan -- /-------------------------------------------------------------------------\ | Dan Weisman - University of Miami - Florida | || || || || | |--------------------------------------------------| || || ||\ /|| | | INTERNET -----> dweisman@umiami.IR.Miami.edu | || || || | || | | BITNET -----> dweisman@umiami | ||||||| || | || | |-------------------------------------------------------------------------| | "The more I get to see, the less I understand..." - Triumph | \_________________________________________________________________________/