-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

MWJ 1998.02.09 (February 9, 1998)
=================================

  Copyright 1998, GCSF Incorporated.  All rights reserved.


Top of the Week
---------------

* **Apple Computer** may have declared marketing war last week,
  with the exiting of retail sales in national chain stores other
  than CompUSA (stores that, by the way, haven't been very Mac-
  oriented in recent years) and a new television commercial
  attacking the Pentium II chip where it hurts - in performance.
  While the commercial brought no press coverage (can't have Apple
  looking _good_ in print, can we?), the retail changes brought
  nationwide reaction of the mostly negative kind - and, as we
  found, it's largely traceable to one article from the Associated
  Press that goes out of its way to make Apple look bad by ignoring
  the real reasons for the withdrawal. In today's MWJ_ **Analysis**,
  we tell you why this is a logical move - and how Apple has already
  done it in Australia - and point out exactly why the coverage most
  media outlets used was hopelessly unbalanced.

* **Ask the Staff** finally returns with a few selections from the
  bag containing our Repository of Unanswered Questions. In this
  installment, we explore how to get better performance out of
  CD-ROM drives that spin down at inappropriate times, clarify our
  boneheaded assertion that PostScript 3 has "transparency" support,
  and give a concrete but simplified example of how digital
  signatures (and checksums) work to answer a query about the
  validity of MWJ_'s own digital signature. If you have technical
  questions, please send them to us at <mwj@gcsf.com>. We can't
  answer all questions, but we'll do what we can.

* **Recent Intrigue** has follow-up on online sales of Alsoft HFS
  Plus utilities, more on the Reuters "news" of Apple's declining
  market share, a clarification on Phil Schiller's resume, plus
  corrections on FileMaker Pro's Windows positioning and "seek
  time." There is also bankruptcy for APS Technologies, a new home
  for Nodester and Widgetizer, a new CEO for Logitech with a
  familiar name, and the unfortunate cancellation of Cyberdog
  licensing talks. "Press Watch" finds interesting delusions from
  James Coates, and we also look at 1GHz processor prototypes and a
  bug in the 68K version of a complicated QuickDraw routine.
  Interspersed in this are our new "In Brief" references to twelve
  other business stories, five other press articles, two more pieces
  of Macintosh news and three developer briefs.

* **Product Showcase** has "In Brief" listings for over four dozen
  products in Internet, Macintosh and Developer categories. That
  leaves us free to focus our attention on Internet Config 2.0b1,
  Conflict Catcher 4.1.1 (get it free with Square One 4.0),
  QuarkXPress 4.0.1r1, Adobe Acrobat forms improvements, Adobe Type
  Manager and Adobe Type Reunion updates (perhaps), StuffIt Expander
  4.5, "Think Different" ad campaign posters, and Frontier 5.0.

* If you have feedback on "In Brief" sections, please let us know
  at <mwj@gcsf.com>. We've tried to achieve a balance of broad
  coverage within our space and time limitations. If we're not
  succeeding, we need to know - it's your feedback that's led us to
  try this format, and we're hopeful it serves your needs better.


Recent Intrigue
---------------

**Follow-Ups**

* The popular Alsoft utilities **PlusMaker** and **PlusMaximizer**
  are now available for online purchase [1]. PlusMaker rewrites your
  hard drive, in place, to be an HFS Plus volume without requiring
  reformatting; PlusMaximizer gives you the option with every disk
  format request of using 512-byte allocation blocks. (MWJ_
  1998.01.18)

  [1] <http://www.alsoftinc.com/beforeyoubuy.html>

* In late January, we took Reuters to task for a "news story"
  about Apple's declining market share, when the story (picked up by
  several media outlets) had no new information in it whatsoever. We
  used it as an example of how reporters pick up on wire service
  copy and run with it even though there's no compelling reason to
  believe it's either news or interesting (something we'll see again
  later in this issue). Just a few days later, the
  _Dallas_Morning_News_ used the Reuters report to point out how
  Apple's market share is dwindling, in fact. Reuters is the one
  that wound up with egg on its virtual face, though. When the wire
  service's management figured out how old this "news" actually was,
  they had to run a correction, pointing out that the sentence about
  information "Apple disclosed in its 10K report filed with the US
  Securities and Exchange Commission" should really have been about
  information "Apple disclosed in its 10K report _filed_in_December_
  with the US Securities and Exchange Commission." (Emphasis added.)
  Note that the original story ran on January 23, and the correction
  on January 26 - both covering something that had happened a month
  ago and which had been extensively mentioned by Reuters' own
  reporters in the interim while providing normal coverage of
  Apple's quarterly results. What other company generates wire
  service reports about month-old information that's already been
  covered? (MWJ_ 1998.01.26)

* Several industry sources inform MWJ_ that Phil Schiller,
  mentioned as a potential Next transplant at Apple Computer, is
  not. This is not Schiller's first employment at Apple, but his
  list of previous employers, to our improved knowledge, does not
  include Next. This is probably why we couldn't confirm that he
  worked there. Our apologies. (MWJ_ 1998.01.26)

* A reader cautions us that FileMaker Pro is the #2 _stand-alone_
  database for Windows, raising questions about how it relates to
  products like Microsoft Access or Lotus Approach, both of which
  are bundled with "office"-type software suites and may not be
  included in the calculations. Claris/FileMaker Inc. claims the
  stand-alone designation is more important, because it shows how
  people who purchase a database-only product vote for FileMaker and
  not for the bundled products; detractors would point out that so
  many people have the bundles, it becomes harder to claim a broad
  constituency for FileMaker. The original data [2] behind this
  claim is now about a year old, but we'll let you know if we find
  anything new. (MWJ_ 1998.02.02)

  [2] <http://www.claris.com/news/press/products/FMP3/rel-number2.html>

* "Seek time," properly defined, is the average time necessary for
  a disk drive (typically a hard disk) to find a given _track_ on
  the disk, not a given _block_ as we specified last week. A track
  contains many blocks; the exact number depends on the hard drive
  and on perhaps on the specific track on the disk. When this
  mistake was brought to our attention, we tried to minimize it by
  claiming that one spin of a hard disk typically takes something
  like 0.13ms - and since one spin is long enough to find any block
  if the track is already located, the delay is minimal. This would
  make perfect sense, too, if hard disks actually spun at 7200
  revolutions per _second_ as we calculated, and not the more
  realistic 7200 revolutions per _minute_ available in today's
  better drives. That's 120 revolutions per second, or 8.3ms per
  spin. That can add significantly to the average time necessary to
  find a block - but remember, on average, half of the blocks will
  be found in less than 4.15ms, and half in more than 4.15ms. Only
  in the worst case would you get 8.3ms per block added to the time,
  and on the 2048-block sample file we mentioned, that'd add another
  17 seconds to the transfer rate. In the absolute worst case. In
  other words, fragmentation can still be bad, so know what you're
  doing when picking smaller allocation blocks. (MWJ_ 1998.02.02)

**Business News**

* **APS Technologies** filed for Chapter 11 bankruptcy protection
  on January 26. According to the company, as reported by MacWEEK
  [3], APS is doing a fine business but has basically run out of
  cash for two reasons. First, drive-manufacturer Micropolis went
  bankrupt earlier this year, and Micropolis owes APS about US$7
  million. Second, Apple's reversal on Mac OS cloning last year hit
  APS hard, since the company had only rolled out its Motorola-based
  M*Power clones a few months earlier. APS profits are still high,
  but, as noted in Meredith Willson's _The_Music_Man_, you gotta
  have "cash for the merchandise." Bankruptcy protection is designed
  to help businesses reorganize and deal with bad debts instead of
  ceasing business operations altogether - some of the biggest Mac
  OS peripheral makers have been through the proceedings, such as
  Radius and Newer Technology, so it's not a sign of imminent
  destruction.

  [3] <http://www.macweek.com/mw_1205/sy_aps.html>

* **Nodester** and **Widgetizer**, two QuickTime VR utilities
  ostensibly from Panimation, have a new home at a company called
  Roundabout Logic [4]. Panimation didn't sell the tools - in fact,
  it didn't own them. They're developed by Duilio Proni of Econ
  Technologies, a contract developer based in Florida. Proni's
  contract work extends back at least a decade, and he has, in the
  past, withdrawn his contracted products from companies that
  weren't selling them to his satisfaction or otherwise living up to
  the contract. Knowing this, you can find significant details in
  the press release, such as "By March 1, 1998, Roundabout Logic
  will ship the latest updates of Nodester and Widgetizer on CD with
  an entirely new look to the packaging. And for the first time, the
  product line will be available from resellers worldwide
  (previously they were sold direct only)." Just for fun, our staff
  looked up the RoundAbout Logic domain name and found Proni listed
  as the administrative contact - both Panimation and RoundAbout
  Logic are based in Florida, and both servers are managed by Mac OS
  hosting company Reprahduce. The Nodester and Widgetizer product
  transition was effective February 1.

  [4] <http://www.roundaboutlogic.com>

* **Guerrino de Luca**, former head of Claris and of Apple Product
  Marketing, has landed a gig as CEO of **Logitech**, the world's
  largest maker of computer mice and related equipment. Logitech is
  a solid company that can use de Luca's obvious leadership skills -
  but, being gadflies, we feel compelled to point out that Logitech
  is _also_ the company that got a group of senior executives to sit
  down with a "management consultant" and turn a clear and concise
  mission statement into an unfocused, buzzword-filled monstrosity
  without ever challenging the "consultant" leading them through
  this abysmal idea. The consultant turned out to be a disguised
  Scott Adams, creator of "Dilbert," and you can read (and see) the
  entire "Mission: Impertinent" story [5] online from the
  _San_Jose_Mercury_News_.

  [5] <http://spyglass1.sjmercury.com/archives/dilbert/>

* News filtered out last week that Apple Computer is so down on
  **Cyberdog** that the company won't even license the code to a
  third-party for maintenance. When Apple dropped "active
  development" on OpenDoc and Cyberdog last March (MDJ_ 1997.03.15),
  the company said they'd stay in the Mac OS for a long time and
  that they'd do what they could to help developers out. In fact,
  Apple did license the OpenDoc Framework (ODF), a developer tool to
  help programmers write OpenDoc parts, to Hutchings Software, which
  has since revised it and released several new and updated
  versions. Kantara Software has been trying to do the same thing
  with Cyberdog for nearly a year now, and while Kantara's Steven
  Roussey claims the talks were going along fine, Apple abruptly
  changed its mind recently and said they are "not interested" in
  licensing OpenDoc, Cyberdog or any similar technologies. It's one
  thing to abandon your self-proclaimed "way of the future" for
  budgetary reasons - it's quite another to actively prevent
  interested developers from advancing the software at no cost to
  you but with benefit to those customers and developers who were
  foolish enough to believe your promises. It could just be the
  normal Apple turmoil - perhaps the folks responsible for the talks
  got laid off, and their logical replacements just don't want to
  deal with the issue. Either way, it's yet another unhappy reminder
  of Apple's inept handling of its own promises.

**Business News In Brief**

* Adobe retail strategy changes to include smaller home versions
  of products
  <http://www.techweb.com/wire/story/TWB19980129S0012>

* The Association of Macintosh Trainers is re-established
  <http://www.cyphernet.com/amt>

* Cendant Software buys Syracuse Language Systems
  <http://www.cucsoftware.com/cdpress/SYRACUSE.html>

* Chip manufacturer S3 purchases Exponential Technology's patent
  portfolio
  <http://www.news.com/News/Item/0,4,18791,00.html>

* SPA adopts new principles of competition that irritate Microsoft
  <http://www.spa.org/gvmnt/tos/compprinciples.htm>

  <http://spyglass1.sjmercury.com/premium/business/docs/microsoft04.htm>
  (Subscribers only)

* US Senate probing Microsoft also
  <http://allpolitics.com/1998/01/29/ap/microsoft/>

* Power Computing dead and gone - out of business
  <http://www.nytimes.com/library/cyber/week/013098power.html>

* Sun and Motorola agree to put Java in embedded systems of all
  kinds
  <http://www.javasoft.com/pr/1998/01/pr980126.html>

* Global Village posts smaller loss than last year
  <http://www.globalvillage.com/press_release_text/earnings/Q3FY98.html>

* Macromedia reports surprise small profit not counting
  acquisition costs for Solis
  (not available on the Web)

* MetaCreations sales sluggish, revenues flat even with higher
  Windows percentages
  <http://www.metacreations.com/press/mcre_4q97_results.shtml>

* RealNetworks posts US$2.6 million loss, larger than a year ago
  <http://www.realnetworks.com/corporate/pressroom/pr/q4results.html>

**Press Watch**

* **James Coates** of the _Chicago_Tribune_ is weighing in on
  Apple's business decisions again, and those of you who've
  monitored Coates' coverage know this is likely not to be good.
  Actually, it's just interesting - in the February 1 "Binary Beat"
  column, Coates claims that Apple's decision to turn Claris into
  FileMaker Inc. and pull the other products back in-house is a
  payoff to - Microsoft. Coates believes that ClarisWorks 5.0 is the
  only Windows suite that could make serious competition for
  Microsoft Office, and so he postulates that the shifting is just a
  maneuver to hide Apple's upcoming discontinuation of the Windows
  version. At the same time, reducing emphasis on ClarisWorks for
  Mac OS could "clear the way" for Microsoft Office 98 for
  Macintosh, making back Bill Gates all of the US$150 million he
  invested in Apple last year. It might make a good screenplay, but
  like most of Coates' fanciful thoughts (including his December
  1996 assertion, never retracted or clarified, that Apple was
  "discontinuing" the Macintosh), there's no evidence to support it.

**Press Watch In Brief**

* Apple "rethinks" schedule goals for Mac OS and Rhapsody (_MacWEEK_)
  <http://www.macweek.com/mw_1206/nw_rethink.html>

* Larry Ellison leaks like a gauze catheter (NEWS.COM)
  <http://www.news.com/News/Item/Textonly/0,25,18552,00.html?pfv>

* _TechWeb_ figures out Rhapsody is now a server OS
  <http://www.techweb.com/wire/story/apple/TWB19980202S0015>

* Former PCC support engineer Chris Heatherly shares his views (MRP)
  <http://www.macresource.com/mrp/contributions/heatherly.shtml>
  Internet News In Brief

* Vendors submit proposal to track threaded conversations in HTML
  <http://www.w3.org/TR/1998/NOTE-HTMLThreading-0105>

**Macintosh News**

* **IBM** made news this week with the demonstration of a
  prototype **1GHz microprocessor**. That's 1000MHz, or three times
  as fast as today's top-of-the-line desktop microprocessors. The
  technology is expected to (in a few years) yield chips as fast as
  1.1GHz, but that's not the interesting part - the chip was made
  using a variation of the same process that creates the "Mach 5"
  PowerPC chips (the 250MHz and faster 604e chips used in Apple's
  high-end Mac OS systems today). The 1GHz chip is based on the
  PowerPC architecture, and makes Intel's showing of a 450MHz chip
  at last year's show look a little pathetic, especially since the
  PowerPC chips have been kicking Pentium butt in processor-
  intensive comparisons (see this week's _Mac_the_Knife_ column [6]
  in _MacWEEK_ for an Intel rumor that's so funny it deserves to be
  true). As usual, press outlets couldn't let the event go by
  without assuring the masses that Intel intends to make chips just
  as fast - but they didn't point out that Intel doesn't have any
  prototypes ready to show, nor did they point out that Intel's
  existing chips continue to lose ground to PowerPC chips as both
  products see increased clock speeds.

  [6] <http://www.macweek.com/mw_1206/op_knife.html>

* Programmer Ken Broomfield has reported a bug in a QuickDraw
  routine named **CopyDeepMask**. Masking is discussed briefly in
  this issue's "Ask the Staff" section. CopyDeepMask is an imaging
  routine that copies an image through a mask, but the mask can be
  more than just black or white (meaning that pixel operations
  either do or don't affect the destination) - a "deep" mask of
  24-bit color can be used, so that the final result is a
  combination of the source, destination _and_ mask pixels. It turns
  out that the 68K version of CopyDeepMask is incorrectly failing to
  remove its arguments from the processor stack, causing a crash at
  the end of the call. The problem affects all PowerPC-based systems
  plus the Quadra 660AV and Quadra 840AV with System 7.5.3 or later,
  or PCI-based Mac OS machines with System 7.5.2 or later. So why
  haven't you heard of this? Because **QuickTime 2.5** and later
  patch CopyDeepMask to work correctly - you'll only see the bug if
  you have QuickTime disabled. Apple's test configurations somehow
  missed this, so the non-patched QuickDraw doesn't include the fix
  even though the QuickTime team obviously found out about it and
  patched CopyDeepMask to solve the problem. Your solution is just
  as simple - make sure you don't disable QuickTime on those
  machines.

**Macintosh News In Brief**

* _MacWEEK_ reports on next-generation PowerPC chips
  <http://www.macweek.com/mw_1206/nw_ibmmotorola.html>

* Recent articles in Apple's Tech Info Library
  <http://til1.info.apple.com/techinfo.nsf/Changes>

**Developer News In Brief**

* Free and portable TrueType engine now available to programmers
  <http://www.physiol.med.tu-muenchen.de/~robert/freetype.html>

* Apple sets dates of May 11-15 for Worldwide Developers'
  Conference
  <http://devworld.apple.com/>

* Developer Depot offers 10% off Eddy Award-winning developer
  products in February
  <http://www.devdepot.com/SBSpecials.qry>


Product Showcase
----------------

**Internet Software**

* Internet Config 2.0b1 [7] is the first major update to the
  public-domain preference module. IC 2.0b1 lets you manage multiple
  sets of preference, and includes a Control Strip module and
  Location Manager module to facilitate switching between them. The
  programming kit has been "reworked" to "simplify" software
  development for those relying on Internet Config, plus other
  changes too trivial to mention. This is a beta, debugging version,
  and some programs will cause a visit to MacsBug with runtime
  errors that aren't really "crashes" (Internet Explorer is
  mentioned by name), so those with weak hearts shouldn't try
  installing it.

  [7] <ftp://ftp.share.com/internet-configuration/>

**Internet Software In Brief**

* Netscape Navigator 4.04.1 128-bit secure version
  Adds stronger security for US and Canadian residents only
 <http://home.netscape.com/download/client_download.html?communicator4.04>

* dataComet 4.50
  US$20 shareware Telnet and terminal emulation client with
  AppleScript support
  <http://www.databeast.com/>

* NetBots 2.5.1
  Shareware automation tool for monitoring Internet servers or
  clients
  <http://www.printerport.com/klephacks/netbots.html>

* NetObjects Fusion 3.0
  Announced new version of HTML authoring program featuring
  multiple and external editors
  <http://www.netobjects.com>

* RumourMill 1.1.1
  Minor bug fixes in US$35 shareware NNTP news server from Peter N
  Lewis and Jim Calvin
  <http://www.stairways.com/rumormill/index.html>

* Tabasoft SendGraphic 1.0
  US$20 shareware utility that builds graphics for your Web sites
  on the fly
  <http://tabasoft.ancitel.it/TABASOFT/SendGraphic.html>

* URL Manager Pro 2.0v4
  User-driven improvements to shareware bookmark manager
  <http://www.url-manager.com/index.html>

* Web Crossing 2.0.2
  High-end Web-based chat and discussion board manager with new bug
  fixes
  <http://webcrossing.com/>

* Vicom Internet Gateway 4.5
  Adds filter features, AppleScript support to gateway giving
  multiple user access through one Net link
  <http://www.vicomtech.com/vig.main.html>

**Macintosh Products**

* Conflict Catcher 4.1.1 [8], a free update from Casady & Greene,
  adds the ability to fix the "damaged" Open Transport resource
  files installed by Mac OS 8.1. They're not all that "damaged" -
  the resource fork of each file is simply longer than the resource
  map says it should be, and ResEdit can easily fix the problem as
  well. Still, it's a nice touch, along with updated links for
  DVD-ROM additions, and a fix so that holding down Command-Period
  (to skip straight to the Finder) will no longer bypass required
  extensions like Appearance or Text Encoding Converter.

  [8] <http://www.casadyg.com/support/updates/getcc4up.html>

* By the way, Binary Software will _give_ you Conflict Catcher 4
  free, just for buying **Square One 4.0** at the normal price.
  Square One is a program launcher and more (version 3.0 was
  reviewed in MDJ_ 1996.12.11). You apparently don't get to keep
  Conflict Catcher 4 if you take advantage of Square One's money-
  back guarantee, but we're not totally clear on that. More details
  are available online [9].

  [9] <http://www.binarysoft.com/order/cc4free.html>

* **QuarkXPress 4.0.1r1** fixes bugs with Bezier items, pictures,
  printing, boxes, style sheets, and text flow. More information is
  available in an enclosed "read me" file that wasn't made available
  to us. This is the first update to XPress 4.0 since its release
  last fall.

* **Acrobat Forms Author 3.5 Plug-In** is the promised Acrobat
  add-on to make forms creation in PDF files much more robust. Form
  creators can use JavaScript for action on client systems (such as
  validation or hiding fields), exporting of forms and data, a new
  layout grid, submission of embedded PDF graphics to Web servers,
  and other new features that let PDF-based forms replace HTML-based
  forms in a lot of cases. Those using Acrobat Exchange (the
  commercial product) can create the new-style forms with the
  Exchange plug-in [10]; those using the free Adobe Acrobat Reader
  can fill in the new-style forms with the free Reader plug-in [11].

  [10] <http://www.adobe.com/supportservice/custsupport/LIBRARY/4da6.htm>
  [11] <http://www.adobe.com/supportservice/custsupport/LIBRARY/4daa.htm>

* According to MacFixIt [12], **Adobe Type Manager Deluxe 4.0.3**
  and **Adobe Type Reunion Deluxe 2.0.3** are now available from
  Adobe - but not online. The former will cost you US$19.95; the
  latter US$9. According to Ted Landau's readers, Adobe claims the
  only changes are new support for Japanese and Portuguese script
  systems, with a single bug fixed that eliminates the need for
  fonts to be marked "read-only" to be opened on a network. Other
  readers claim ATM 4.0.3 improves legibility of anti-aliased fonts
  against colored backgrounds plus other bug fixes. To order, call
  (800) 682-3623 and press option 2, or bug your local Adobe office.
  It's hard to imagine why Adobe would make an update, claim it's
  only for non-US users, and then offer it only through a toll-free
  number limited to the US and Canada.

  [12] <http://www.macfixit.com/>

* StuffIt Expander 4.5 [13] is again available. New features
  include the ability to mount ShrinkWrap disk images and to decrypt
  Private File-encrypted files, making Private File a much more
  useful program. Since the Expander Enhancer isn't ready for this
  release yet, those who have the older version installed may wish
  to wait for an update to avoid conflicts with different versions
  of the StuffIt Engine.

  [13] <http://www.aladdinsys.com/expander/expander2.html>

* If you like the **Think Different** ad campaign, Apple is
  offering a set of five posters based on the print ads for US$6.75
  plus shipping and handling to those who are interested. Supplies
  of the first printing have run out, but Apple said they would take
  backorder requests and fulfill them when printing was completed.
  The five posters feature Gandhi, Pablo Picasso, Thomas Edison,
  Alfred Hitchcock and Amelia Earhart. US and Canadian residents may
  call (800) 825-245 and ask for product #L02761A. Information for
  international residents wasn't available, but we _think_ you can
  E-mail Starting Line [14] and ask for details. Don't quote us on
  that until you can verify it; we're trying to do so now.

  [14] <s.line.order@apple.com>

* Frontier 5 [15] is now available from UserLand Software. There
  are all kinds of bug fixes, new features and enhancements in the
  free scripting language (the first OSA language released, back in
  1991), but the real news is that Frontier 5 is cross-platform and
  works under Windows as well. Mac-specific enhancements include
  better Web site management features, support for file-based
  objects, a cross-platform file format (based on fat Web pages,
  something that always crashed our browsers) and a nifty new way to
  view the object database as an outline instead of just as tables.
  Full change notes are available online [16]. Frontier 5 remains
  free, but UserLand owner Dave Winer reserves the right to make
  commercial products based on it.

  [15] <http://www.scripting.com/frontier5/>
  [16] <http://www.scripting.com/frontier5/changeNotes/appChanges.html>

**Macintosh Products In Brief**

* AdobePS Watermark Plug-in
  Adds watermarks (background images) to Adobe's LaserWriter driver
  <http://www.adobe.com/prodindex/printerdrivers/watermrk.html>

* Apple Memory Guide, February 1998
  Lists memory configurations and expansion instructions for all
  Macs and printers
  <http://swupdates.info.apple.com/cgi-bin/lister.pl?Apple.Support.Area/Apple.Software.Updates/US/Macintosh/Utilities/>

* EtherPeek 3.5 for Macintosh
  Summary statistics, better capturing in US$995 network monitoring
  tool (upgrades US$250)
  <http://www.aggroup.com/news/epmac35.htmy>

* Apple Network Assistant 3.0.1 Updater
  Fixes problem that copied the wrong System Folder during a Copy
  Hard Disk command
  <http://swupdates.info.apple.com/cgi-bin/lister.pl?Apple.Support.Area/Apple.Software.Updates/US/Macintosh/Networking-Communications/ANAT/Network_Assistant/>

* _Programming_4th_Dimension:_The_Ultimate_Guide_
  New book and accompanying CD-ROM on 4D programming from experts
  at Foresight Technology
  <http://www.foresight.com/products/program4d/new.html>

* FMPro Tuner 1.0
  Extension gives FileMaker Pro more CPU time when in background,
  makes networking faster
  <http://www.proteron.com/fmprotuner>

* Font Box 2.6
  Usability enhancements in font management and repair utility
  <https://insidertwo.theinside.com>

* form*Z 2.9.5.3
  Various updates for various versions of the high-end 3D rendering
  program
  <http://www.formz.com/support/updates.html>

* Fujifilm MX-700 Digital Camera
  Pocket-sized camera with 1280 X 1024 pixel resolution for 1.5
  million square pixels
  <http://biz.yahoo.com/bw/980203/fujifilm_1.html>

* Mac OS Extended Calc 1.0
  Glenn Austin's nice calculator for HFS and HFS+ disk space
  requirements, US$15 shareware
  <http://www.austin.dyn.ml.org/software/>

* Extensis Intellihance 3.0.2
  Artificially intelligent photo retoucher now works in Adobe
  PhotoDeluxe
  <http://www.extensis.com/products/Intellihance/>

* Extensis Mask Pro for US$100 off - only US$199
 High-end Photoshop plug-in forms complex masks quickly, but demo
  crashes our system
  <http://www3.digitalriver.com/maskpro/2051>

* BoxTop Software Photoshop Filters
  Bug fixes for PhotoGIF (Lite, Filter, Regular), ImageVice,
  ColorSafe, ProJPEG
  <http://www.boxtopsoft.com/download.html>

* QuickTime 3.0b16
  Latest pre-release of QuickTime 3.0, now expected to be released
  at Macworld Tokyo on 1998.02.16. Expires 1998.02.19.
  <http://swupdates.info.apple.com/cgi-bin/lister.pl?Apple.Support.Area/Apple.Software.Updates/US/Macintosh/Unsupported/>

* LightningDraw/WEB for US$79
  Lari Software offering 20% discount on QuickTime 3 package until
  QuickTime 3.0 is released
  <http://www.larisoftware.com>

* MovieWorks 4.0.2
  Multimedia authoring tool works with QuickTime 3.0, on sale for
  US$50 through March 15
  <http://www.movieworks.com>

* February Specials from Nisus Software
  Discounts on bundles including Nisus Writer 5.1.1 and an older
  (3.47) Nisus Compact
  <http://www.forecommerce.com/nisus_shop/cgi/show/nisus_shop/specials.inv>

* QUED/M 3.0.3
  Unspecified changes to macro-powered text editor that predates
  Nisus Writer
  <http://www.nisus-soft.com/download.html>

* Symantec Antivirus for Macintosh Intercept 4.5.3
 Patch to fix SAM Intercept for compatibility with Microsoft
  Office 98 - required for Office
  <ftp://ftp.symantec.com/public/english_us_canada/products/symantec_antivirus_macintosh/ver4.5/updates/samint453patch.sea.hqx>

* TechTool updates
  New versions of TechTool, TechTool Pro (not 2.0 yet) and DNA
  files for Mac OS 8.1
  <http://www.micromat.com/micromat/software.html>

* Jeremy's CSM 1.8.1
  Bug fixes to Jeremy Kezer's US$10 shareware Control Strip
  modules, now including Caps Lock status
  <http://members.aol.com/jbkezer/csm.html>

* Data Hammer 1.5
  US$35 PowerPC-only data summarizing tool now extracts bulleted
  list of topics in analyzed text
  <http://www.glu.com/DataHammer.html>

* WATCHIT 1.5
  Upgrade to automatic time-management program no longer requires
  FileMaker Pro
  <http://www.hi-resolution.com/>

* Mac OS Installer Helper 1.0b7
  Helps reconstruct System Folder after clean installation; fixes
  some bugs and recognizes more files
  <http://www.QuickTimeFAQ.org/software/>

* PGPdisk 1.0.1 Updater
  Small bug fix in absolutely secure disk encryption program for
  HFS Plus volumes now online
  (Note: Forms for verification don't seem to work with Microsoft
  Internet Explorer)
  <http://www.pgp.com/products/PGPdisk.cgi#update>

* Neatnik 2.0.2
  New release to clarify limitations in US$10 shareware for
  cleaning up Finder windows and views
  <http://users.aol.com/KarlBunker/Neatnik.html>

* VirusScan 3.0 for Macintosh
  Adds new features for scanning files as downloaded, plus more new
  stuff - on sale for less than US$30
  <http://www.mcafeemall.com/mall/mcafee/vsmacxfact.html>

* Zippier Zip 5.0.3
  Shareware removes the dialog box when Iomega drivers are loaded
  after startup; works with Iomega Guest 5.0.3
  <http://www.montalcini.com>

* Disk-o-Matic 1.0
  Automatically opens specified items when disks are mounted or
  inserted
  <http://www.montalcini.com>

* Jon's Commands 2.0.3
  Useful AppleScript extensions have bug fixes and clipboard
  enhancements
  <http://hyperarchive.lcs.mit.edu/HyperArchive/Archive/dev/osa/jons-commands-203.hqx>

* Quicken 98 Release 3 Updater
  Fixes more bugs in Intuit software, including generation of
  proper tax files for MacInTax
  <http://www.intuit.com/support/quicken/index/ndxm_8_updates.html>

* Metro 4.0
  Integrated MIDI and digital audio plus effects for under US$200
  from Cakewalk
  <http://www.cakewalk.com/Press/metrofnl.htm>

* Cakewalk In Concert
  Keyboard accompaniment program follows a player's adjustments in
  tempo and style
  <http://www.cakewalk.com/Press/icfnl.htm>

* YOU DON'T KNOW JACK Volume 1 and Volume 2 updaters
  Fixes problems with the game programs that may show up on some
  machines
  <http://www.berkeleysystems.com/lite/freestuff/updt.html>

* Bumbler Bee-Luxe 1.0.1
  Dadgum Games' bee-attack game has better 603-based performance
  and some easier swarms
  <http://www.dadgum.com/>

**Developer Software In Brief**

* BuildSim 2.0
  High-end toolkit for rapid production of simulations and other
  analysis tools
  <http://www.tangentsys.com/products/web_buildsim/bs_page1.html>

* c-tree Plus V6.7A
  US$895 database access manager with improved performance and
  transaction history
  <http://www.faircom.com>

* MRJ SDK 2.0 Final
  <http://applejava.apple.com/text/download.html>

* MRJ SDK 2.0.1 Pre-release
  No available information on what's different - was available
  before 2.0 Final
  <http://applejava.apple.com/text/prerelease.html>

* Java Foundation Classes 0.7 Pre-release
  "Swing" classes promoted by Sun and Netscape as universal Java
  utility classes
  <http://applejava.apple.com/text/prerelease.html>

* Everything CD for Mac OS Scripting
  US$70 CD-ROM with OSAX, source code, shareware, freeware, tools,
  archives and other scripting goodies
  <http://www.isoproductions.com/cd/macscripting/>

* SuperCard 3.0.3
  Fixes a problem with the Mac OS 8 Finder
  <ftp://ftp.allegiant.com/SuperCard/Updates/SC_303_Updater.hqx>

* WASTE 1.3
  Latest release of Marco Piovanelli's WorldScript-Aware Styled
  Text Engine
  <http://www.boingo.com/waste/>


Apple Declares War?
-------------------

**Company Redefines Retail Experience, Finally Differentiates
  PowerPC Chips**

  After last fall's introduction of the "Think Different" ad
  campaign (MWJ_ 1997.09.29), Apple promised that the initial 60-
  second television advertisement would be just the first in a
  series of aggressive new marketing moves for the company,
  including a new campaign to distinguish Apple's products from
  those of competitors. Many of us have been wondering what happened
  to those promises, after a Christmas season filled with nothing
  more than Apple-branded ads that only appeared most strongly in
  major metropolitan areas. In the rest of the US, and probably the
  world, Apple's image continued to be one of old, disconnected demo
  machines in large electronics stores, coupled with blind
  acceptance that the Pentium II processor (hawked in ads every time
  you turned around) was the only way to get a computer with power.
  Apple continued to let competitors define the playing field, as
  it's done for so many years.

  Last week may, just may, have marked the beginning of the end of
  that problem. In a couple of bold moves with the potential for
  huge backlash, Apple Computer withdrew its products from US
  nationwide resellers other than CompUSA, and introduced a new
  television advertisement touting the superiority of the PowerPC G3
  chips by showing the Pentium II chip on the back of a _snail_.

  Not surprisingly, press reaction has been mostly negative or
  mostly silent. Before we discuss why that's misleading at best,
  let's look at what actually happened.

**Redefining Retail Expectations**

  Rumors have been circulating that changes were afoot in Apple's
  retail channels for a few weeks now, with reports from those
  working at major chains like Best Buy and Computer City indicating
  that Apple products were being pulled from the shelves, or that
  salespeople were being told that the Macintosh lines were being
  dropped. Best Buy, based in Eden Prairie, Minnesota, was first to
  confirm the news late in January, announcing that by a "mutual
  decision," Apple hardware would no longer be sold at the company's
  284 US stores, although Macintosh software would still be on the
  shelves. The company claims that home office products account for
  40% of sales, but Apple's brand brought in less than half of one
  percent of that figure, making Apple's total share of Best Buy's
  revenue something like a fifth of a percent. In other words, one
  out of every US$500 Best Buy received came from Apple-branded
  hardware sales.

  If this makes you pause, you're not alone. Most Best Buy stores
  carry thousands of products from dozens of manufacturers, and it's
  unlikely that more than a handful of companies can beat that kind
  of revenue level for the stores. Manufacturers like Sony, Compaq,
  Mitsubishi and others that make big-ticket items (desktop
  computer, big-screen TVs, high-end stereos) logically have the
  best shot, but you would be stunned to hear that Best Buy was
  dropping a manufacturer like Uniden because the company's
  inexpensive cordless phones didn't account for more than US$1 of
  every US$500 in sales. (That's just an example; Best Buy has not
  commented on numbers for other manufacturers.)

  A more likely scenario is that Best Buy wisely committed a pre-
  emptive strike. Remember last summer, when Apple decided to put
  its advertising account up for review because BBDO West's "The
  Power to Be Your Best" slogan had run out of gas? An advertising
  agency "review" is more or less the firing of an ad agency -
  instead of sticking with the current agency, Apple decided to see
  proposals from several invited agencies, including BBDO West, and
  see which one was the best. Since BBDO's ongoing campaign had
  prompted the review, BBDO realistically saw little chance in
  winning such a review. Rather than face the humiliation of losing
  the account, BBDO turned on Apple and publicly withdrew from the
  review, blaming Apple's shifting management and inconsistent
  messages for their inability to produce compelling advertising.
  (MDJ_ 1997.06.27) And, unsurprisingly, much of the nation's press
  reported it as "BBDO fires Apple" instead of the other way around.

  It seems that Best Buy made a similar choice, but without the same
  rancor. Knowing that Apple was changing the way retail purchasing
  works, the retailer very wisely chose to trump Apple's card,
  answering press queries by confirming the end of the hardware
  sales relationship. The "mutual decision" language and the early
  announcement made it look like Best Buy was dumping Apple, instead
  of the other way around. We believe the latter is much more
  likely, as a closer look at Apple's actual decision shows.

**What Did Apple Do?** -- On February 2, Apple announced that it
  would "focus" its national retail efforts around CompUSA and the
  well-publicized "Apple store-within-a-store" concept that had paid
  off so well in the Christmas quarter. Although Apple has lately
  had problems understanding that "focus" means "a narrowing of
  attention on a specific area, usually to the exclusion of other
  areas," they seemed to get it this time. Unlike previous
  announcements where Apple executives maintain the company is
  "focusing" on every product segment imaginable (MWJ_ 1997.10.20),
  a "focus" on CompUSA means Apple is discontinuing the relationship
  with other national retailers. In addition to leaving Best Buy,
  Apple is withdrawing hardware sales from Computer City, Circuit
  City, OfficeMax, and Sears.

  Those who study the "big picture" will recognize this move. It's
  exactly what Apple Computer Australia did last year (MDJ_
  1997.05.15), withdrawing from national department stores to focus
  on regional dealers and AppleCentre stores (MDJ_ 1996.08.19).
  Apple Australia made the move because their products were getting
  hammered in the stores - not necessarily on sales, but on image.
  Demonstration machines were outdated, disconnected or both; Apple
  Australia's then-general-manager said that superstore salespeople
  were making "disparaging remarks. Those machines that were
  actually up and running tended to feature the exciting Finder,
  giving those looking for features little chance of finding them
  unless they already knew where to look. The salespeople often
  received "spiffs," or evangelical payments, to sell machines of
  competing manufacturers - usually those running Windows where the
  retailer or the manufacturer could count on lucrative technical
  support contracts or warranty repair work for simple problems that
  nonetheless seem complex under Windows. Combine this with the
  now-inescapable catchphrases "Windows is just as good as a
  Macintosh" and "Apple is doomed," and you've got a recipe for
  exactly the kind of disaster Apple Australia was experiencing.

  When customers enter the stores and come out with a negative
  impression of your products based on anything other than the
  products themselves, you've got a problem. There are three ways to
  fix it. You can do the sales for the store with massive national
  advertising of your product's features, but that can't overcome
  broken products in the store and salespeople who mislead the
  customers. You can work with the store to reeducate and train the
  salespeople on the products, but that's work - work the employees
  don't have to do to sell competing products which may carry spiffs
  (spiffs alone won't work if the staff knows beans about the
  product). Or you can get your products the hell out of that store
  - making customers work a little harder to find them, but being
  much more sure they'll hear the truth when and if they do arrive.

  Apple Australia chose the third option, and much like happened in
  the US, they were characterized as defensive and retreating for
  doing so. Unlike in the US, however, Apple Australia's executives
  were much more open about the abysmal state of retail sales in the
  chain stores, and about pointing out that fixing it seemed
  unlikely given current management priorities at said stores. And
  also unlike the US, most of the bad-mouthing came from -
  executives of those same stores, who characterized Apple's moves
  as circling the wagons in an attempt to shift the spotlight away
  from their inability to adequately sell the products they'd
  originally fought hard to get into their stores.

  In the US, the words are much more circumspect, but you probably
  already know the story's the same. We've made trips through
  several local stores belonging to the retail outlets that will no
  longer carry Apple products, and without naming names, we have to
  say this is a good move. Systems were 12 to 18 months old, if they
  were even present (one local "superstore" doesn't carry the Apple
  products at all, making you wonder how the parent company would
  expect to get significant revenue from them). Those that were
  present were usually disconnected. The printer selection was
  pitiful to non-existent. Software was either absent or next to the
  PC clones. Salespeople had no idea about the Macintosh line, but
  were happy to sell you one if you were sure it was what you
  wanted, even though everyone else uses Windows, which is (of
  course) cheaper. And yet these same stores have the gall to act
  disappointed when reluctantly reporting that the only people
  buying Macintosh products are those who come in knowing they want
  one. No one who's undecided ever has a chance in this kind of
  situation.

  Apple has recognized the intolerable nature of this situation for
  some time, but has only recently taken concrete action to fix it.
  The first step was in November, with the creation of the CompUSA
  store-within-a-store agreement, one that is now complete (the
  Apple areas are supposedly in place in all 148 US superstores).
  CompUSA had been one of the worst offenders in terms of crucifying
  Apple's image - reports throughout 1996 and early 1997 were almost
  uniformly bad concerning the chain's treatment of Apple products.
  CompUSA executives were quoted in the press more than once (for
  example, MDJ_ 1996.10.15) saying sales of Apple products were down
  more than 50%. It's illogical to think that CompUSA, which has
  been unconcerned about Apple's sales, would talk to the computer
  maker in an attempt to improve relations; it's more than logical t
  o think that Apple, which has publicly and repeatedly talked about
  improving the retail experience, would approach CompUSA with a
  request to fix the problems or stop selling the products
  altogether.

  CompUSA wanted to fix the relationship, and here's the kicker: it
  worked. In the stores where the store-within-a-store areas were
  completed in November, sales of Macintosh products rose from 3% to
  14% of CompUSA's total revenue. Think about that, because it's
  huge - a 367% increase in _one_month_! (The percentage difference
  is only valid if sales revenue from one month to the next was
  flat, but available information strongly indicates CompUSA's
  revenue also increased during the same period, meaning the actual
  dollar amount may have been even higher than a 367% increase.)
  Some of it was undoubtedly due to new Power Macintosh G3 products,
  but similar sales increases were not noted in other retail outlets
  where G3 machines were also available. One has to attribute a
  large share of the credit to a new retail environment that
  honestly represents Apple products.

  CompUSA, however, has always been on the forefront of these kinds
  of moves. CompUSA was the first national chain to win approval for
  carrying Apple-branded merchandise in the early years of this
  decade, and has since grown to be the nation's largest computer
  retailing chain. Although details were not disclosed, it's fair to
  _guess_ that Apple paid part of the costs for the store
  conversions, and so far it's clearly been worth it. It's also fair
  to _deduce_ (note that no party has confirmed this) that Apple
  wanted to make similar improvements with other major retail
  chains, and those chains didn't want to make the effort. Rather
  than leave things in an intolerable state, Apple picked up its
  ball and went home, leaving CompUSA as the sole national retail
  outlet for its hardware.

  Note that this doesn't mean CompUSA stores are the only retail
  outlets where you can find Macintosh machines. The announcement
  has nothing to do with _regional_ chains such as Fry's or
  ComputerWare, which continue to present Apple products honestly,
  to the best of our knowledge. It does mean that you can reasonably
  expect Apple to take similar steps with regional chains that
  aren't honestly representing the hardware in their stores. The
  deal also doesn't affect the thousands of "authorized Apple
  dealers" in one-store or one-city shops. The denial of hardware
  only affects national merchants other than CompUSA, and they're
  listed above. Even Best Buy was diplomatic enough to leave the
  door open - the chain noted that it really focuses on consumer
  products, and Apple is focused on education and creative
  professionals. That leaves open the possibility that when and if
  Apple's renewed "focus" on consumer markets materializes, Best Buy
  might like to be back in the game. Until then, the chain is
  passing on the kinds of changes and training that CompUSA has
  made, and Apple is apparently saying that is now the minimum level
  of commitment to carry Macintosh products.

  This is mostly good news for Apple. With the advent of the Apple
  Store online, any customer with a computer can research the newest
  products and build one to individualized specifications, and then
  order the machines from either Apple or from CompUSA over the Web.
  And in most cities throughout the country, other, smaller dealers
  will be happy to lose the competition from the major chains that
  were not only stealing occasional sales but gutting demand by
  bad-mouthing Apple products. There will now be places where it's
  harder to find an Apple product than it was before - the Columbia,
  South Carolina newspaper _The_State_ noted that there are no
  CompUSA stores anywhere within the state (of South Carolina, not
  within the newspaper). However, the newspaper pointed out that
  Sears, Best Buy, Circuit City and OfficeMax will continue to
  service and support Macintosh computers bought in their local
  stores. The Sheryl Jean article also pointed out that mail-order
  and catalog sales are unaffected by the changes.

  The real problem for Apple in all of this is simple: customers who
  answer TV ads and choose these electronics superstores for a
  first-time computer purchase will no longer even know Apple is an
  option, unless they do their homework first and consciously
  investigate CompUSA or regional Apple dealers. Apple has to
  counteract this problem by creating more national demand,
  something they took on later in the week. But first, a small
  detour.

**Press Reaction Infuriating** -- If you heard about Apple's
  mostly-positive change in retail focus on last Monday's news, you
  probably didn't hear about a mostly-positive move. Our media spies
  found negative coverage in local newspapers, on cable newscasts,
  and even on ABC Radio News at 3 PM EST on the day of announcement.
  Even though Apple's VP of sales Mitch Mandich explicitly said
  (even in the press release), "This does not represent a retreat
  from retail, but instead a redefinition of what the retail buying
  experience will be for our customers," hundreds of news outlets
  said exactly the opposite.

  You can mostly thank the Associated Press for that. In an unnamed
  story that moved Monday morning, the wire service reported as
  follows: "The move is just the latest pullback by Apple, whose
  Macs are progressively losing share of a home computer market it
  once dominated. Computer buyers have instead been buying less
  expensive systems running on Microsoft's Windows operating
  system."

  Note the multiple problems in this paragraph. "Just" the latest
  pullback indicates that Apple has been retreating from sales
  before and will continue to do so, when there is no evidence to
  support this. (There are continued reports on Ric Ford's
  MacInTouch [17] about Apple not granting new dealer licenses at
  this point in time, but if this is what AP meant, they should have
  said so.) The assertion that Macs once "dominated" the home
  computer market are completely unfounded - anyone with an eye to
  Apple history knows that the Macintosh has never held a majority
  share of the home computer market (or the overall computer
  market), and even Jim Carlton says that Apple never held more than
  16% market share at any point in its history. Yet the claim that
  Apple used to be better makes a decline seem more sharp, even when
  evidence doesn't support it. And, as always, references to Windows
  systems assert that they cost less without examining overall cost
  of ownership or relative system power, areas where Apple is very
  competitive.

  [17] <http://www.macintouch.com>

  Later in the day, AP released a similarly-structured but
  differently-written report by David E. Kalish, a business writer
  for the wire. Kalish got a few more details out, but didn't try
  any harder than the uncredited earlier reporter (probably also
  Kalish, deduced from some identical sentences) to present a
  balanced point of view. He noted that Apple's stock "plunged more
  than 3 percent" without mentioning this was a drop of around 62
  cents per share, well within normal trading parameters. Kalish
  quoted market research from Computer Intelligence, an industry
  research firm we've mentioned more than once for taking data about
  certain markets and unrealistically extrapolating it to imply
  things about Apple's business as a whole - extrapolations that are
  nearly always wrong and yet never retracted by the company or the
  press (for example, MDJ_ 1996.10.28, 1997.03.06, 1997.07.16).
  Kalish quotes CI analyst Dave Tremblay as saying the pullback is
  "an admission of lack of success in selling PCs in those
  channels," and also bemoans how the lack of visibility on shelves
  will hurt Apple's recovery. Kalish doesn't bother to mention that
  shelves full of outdated or unplugged machines weren't helping. He
  mentions Apple's 3% stock price drop but doesn't quantify the 367%
  rise in Apple sales at CompUSA - he only mentions that sales "have
  grown."

  Kalish quotes an analyst at Dallas-based Channel Marketing thusly:
  "If I was an Apple owners, I would be very disappointed right now
  there wasn't a place (in my area) I could go to get Mac products.
  That doesn't give me any choice as a consumer." Neither Kalish nor
  the analyst pointed out that Apple products are still available in
  some 3,500 stores nationwide - and the assertion of "nowhere
  local" to go is especially weird for Channel Marketing, since
  CompUSA is based in Dallas. In other words, AP's article played up
  all the negative points and consciously left out any positives.
  That's not what we expect from the wire service.
  _The_Wall_Street_Journal_ or the _New_York_Times_, yes. AP? No.

  And it didn't have to be that way. Reuters' coverage of the
  announcement, which came hours later, was fairly well balanced,
  including the increase in sales at CompUSA (but only the 3% to 13%
  comparison, not the delta percentage we mentioned earlier). The
  _San_Jose_Mercury_News_ article [18] in Tuesday's edition was
  downright positive, with statements that accurately portray
  Apple's position. Compare some of these to the AP report: "The
  decision focuses Apple's retail strategy on stores that
  aggressively sell its products." And "Many industry watchers agree
  that it was a move Apple needed to counter lagging sales. As Apple
  lost market share during its upheavals, the company didn't get
  aggressive support from the larger chain stores, says Tim Bajarin,
  president of Creative Strategies. 'Apple had to make a decision of
  this nature to get the stores that had the Mac to aggressively
  sell them,' Bajarin says." Kalish also quoted Bajarin, one of the
  few analysts that truly understands the Mac OS market, but Kalish
  only quoted him in terms of the negative risks to Apple from the
  decision. Bajarin's assertion that these were probably "good
  calculated risks" at this point seems lame in Kalish's context of
  doom and gloom. Kalish has pulled this kind of trick before, using
  Newton's spin-off in May as an excuse to write an "Apple is
  doomed" article (MDJ_ 1997.05.27). That one went so far out of its
  way to cast aspersions that Kalish's assertions didn't even agree
  with the numbers quoted within the article. A _Seattle_Times_
  article on Saturday was similarly upbeat on Apple's move.

  [18] <http://www.sjmercury.com/business/apple/docs/store020398.htm>

  Unfortunately, the AP story was first on the wire last Monday, and
  that's what most media outlets went with. AP's story didn't
  mention the poor retail experience for Apple products at these
  chains at all, and only the _Mercury_News_ touched on it lightly.
  If any of these people, including Computer Intelligence analyst
  Dave Tremblay, had actually been _in_ a randomly-selected major
  chain store trying to get honest answers about Mac OS products,
  they'd have seen the problem immediately. Of course, that assumes
  they'd know misleading information about Apple's products when
  they heard it, and that's a big assumption for an analyst or
  mainstream press reporter anymore, sad to say.

**Defining the Products**

  So Apple is moving to fewer outlets but with a stronger presence
  in the ones left. When the presence in the other outlets is that
  bad, you really have little choice. But now the company has
  another problem - how to get consumers to actively investigate the
  Macintosh option, given that it may not be present in their choice
  of national chain stores?

  Part of the solution is to create compelling products, which Apple
  is slowly accomplishing. The other part is to make sure people
  know _why_ Apple's products are better than the competition's. For
  years, Apple has focused on the Mac OS as superior to Windows as
  the key differentiator, and it's failed miserably. Those who use
  both operating systems know the Mac OS is still easier to use,
  generates more productivity, and has lower overall costs of
  ownership. However, popular culture has accepted that Windows is
  "just as good," and years of Apple marketing have failed to even
  nudge this notion from its favored spot in the collective psyche
  of first-time computer buyers.

  Starting Wednesday night, Apple went after the other half of the
  duopoly - Intel. Apple premiered a new 30-second television
  commercial that night on ABC, and announced the new spot only at 5
  PM EST, after most analysts and financial types had left for the
  day. Some who saw the announcement expected a familiar commercial
  - after all, at the November 10 roll-out of the Power Macintosh G3
  systems, Steve Jobs showed a 15-second television commercial of
  Muhammad Ali (in his prime) boxing and weaving in front of a
  camera.

  This was not what Apple aired.

  Instead, viewers were treated to a large and not-very-photogenic
  snail, slowly moving across the screen - carrying a Pentium II
  processor on its back (it was nearly bigger than the snail). The
  announcer (a new voice for Apple commercials) told the audience
  that some people think the Pentium II is the fastest processor on
  Earth, but that's not quite true: the chip inside every Power
  Macintosh G3 machine is "up to twice as fast."

  This is a slight exaggeration, but not by much. Apple has a
  summary of the claims online [19], and they show that when testing
  CPU performance (only) with the BYTEmark benchmark tests conducted
  by _BYTE_ magazine, the 266MHz PowerPC 750 chip has up to double
  the integer performance of the Pentium II/300MHz chip. Apple's
  product marketing conducted tests on the new Pentium II/333MHz
  model, and found the Power Macintosh G3 chip still beats it,
  hands-down.

  [19] <http://www.apple.com/hotnews/features/bytemark.html>

  Of course, as we've noted before, CPU performance isn't the best
  gauge of system performance. We've pointed out how Apple's
  demonstrations invariable use memory-intensive tasks to avoid
  bottlenecks imposed by the current, highly-limited Mac OS
  input/output systems like the File Manager. Yet it's no more
  deceptive than Intel commercials, which boldly claim the Pentium
  II processor is what you need (implying it's the only choice) for
  tasks like photo processing, 3D games, and DVD video. Intel has
  been defining the market, and Apple has decided to play at that
  game. It's a risky move, but a necessary one - Apple _must_
  convince customers that their products have significant
  advantages. Contrasting against Windows has failed to do this, and
  Apple and Microsoft have now made up and are nice to each other
  anyway. The next logical thing to try is attacking Intel.

  It could backfire - Intel's processor margins and virtual monopoly
  on PC innards has given the company a huge marketing budget.
  However, the areas where Wintel PCs exceed Mac OS expectations are
  due to the operating system, not the hardware. Straightforward
  refutation of Apple's ads would require Intel to start championing
  Windows instead of its own chips, and that's unlikely. Plus, IBM's
  announcement this week of an experimental 1000MHz PowerPC
  processor (not expected to be commercially available for a couple
  of years) doesn't give Intel much room to brag.

  If Apple will start following through on this new marketing, we
  could have a war on our hands as the company _finally_ begins to
  tell people why a Macintosh is a better choice than a standard PC.
  They've lost the "we're better than Windows" battle, but the "our
  machines are more powerful" option may have some life in it. If
  Apple aggressively pursues this line - and by "aggressively," we
  mean to the point that new television commercials are no longer
  newsworthy just for appearing - then some analysts and other PC
  market watchers might be surprised by the truth about Apple's
  products and capabilities. It'd be nice to see products judged by
  their capabilities instead of their bland ability to merge in with
  the competition.


Ask the Staff
-------------

**A Dip Into the Bag Containing our Repository of Unanswered
  Questions (the RUQ-Sack)**

  As sanity in news returns to the Mac OS market for a while, MWJ_
  returns to technical content, warming back up with some questions
  lobbed to our staff through the standard E-mail address [20]. If
  you have technical questions about what's going on with your
  system, drop us a line. We can't promise responses or publication
  for all queries, but we'll do what we can.

  [20] <mwj@gcsf.com>

**You Spin Me Right Round**

Q What controls the amount of time that the CD-ROM drive keeps
  spinning after the CD itself mounts? Currently this spin-down
  delay is about ten seconds on my machine. Is this configurable
  under OS 8, or must I obtain some sort of CD-ROM utility? A delay
  of about a minute would often make more sense.

A First, a primer for the uninitiated. CD-ROM drives can turn
  physical patterns (in this case, microscopic pits on a CD-ROM disc
  that either reflect a laser's light or do not) into digital data
  only by knowing precisely how to interpret those patterns. To get
  a regular groove going here, the drive has to spin the disc at a
  standard angular velocity - a set number of revolutions per
  second. When the disc spins at the right rate, the laser in the
  CD-ROM drive can measure the distance between one of the
  microscopic pits and the next pit in terms of microseconds. If the
  disk spins too quickly or too slowly, the same number of
  microseconds will leave the laser over the wrong spot on the disc,
  causing errors. It's the same way with all drives - those folks
  who had other personal computers in the late 1970s or early 1980s
  may remember that 5.25" floppy disk drives occasionally had to
  have their speed adjusted, or else the drive wouldn't be able to
  read disks that were written on a drive with the correct speed.

  No kind of motor can move a disc from no velocity to proper
  velocity instantly, so the drive does something called _spin-up_,
  a period of a few seconds so the motor can get the disc spinning
  at the proper, stable speed. Hard drives spin up as well, as
  PowerBook users know all too well - when a disk is spinning, the
  motor is keeping it rotating at the right speed. Hard disks stay
  in movement nearly all the time, unless an energy saving mechanism
  stops them. When they stop, depending on the mechanism, they
  usually slow down first to avoid wear and tear on the drive parts.
  This is, not surprisingly, called a _spin-down_ cycle.

  CD-ROM drives aren't like hard drives - when they're not actively
  using a disc, they let the motor spin down and let the disc come
  to an angular halt. If you're using a program that regularly needs
  to read small amounts of data from a CD-ROM drive (such as Riven,
  perhaps), you may get frustrated at repeated spin-down/spin-up
  cycles, each adding a few seconds onto a file reading transaction
  that probably wasn't very speedy to start with. It makes sense to
  want to leave the drive spinning to avoid the problem.

  Unfortunately, it doesn't work that way. Perhaps it's because of
  mechanical difficulties in keeping a freely-loaded disc spinning;
  perhaps its because the spinning takes more power than in a hard
  drive; perhaps it's because they've always done it that way. But
  for whatever reason, CD-ROM drives spin down after a read
  operation is done. There's a short delay - if another request to
  read data arrives within the next few seconds, it's easier on the
  drive to just keep things going. So, just in case, the drive stays
  spinning at the appropriate speed for a few seconds after each
  read request is completed. This, as we understand it, is the delay
  you want to extend to as much as a minute.

  As far as we know, you can't do this. Sometimes specifications for
  drive operations are kept with  the drive itself, so that driver
  software can change them for better performance in individual
  cases. However, we're unaware of any CD-ROM driver package that
  lets you play with these parameters the way FWB's Hard Disk
  Toolkit does with hard drive parameters stored in this way. So
  even if a drive did have controls for spin-down time, you couldn't
  access them. But we doubt that such controls exist - we can't
  remember seeing a hard drive with control over this level of
  mechanical detail.

  So what can you do? The most common solution is to off-load CD-ROM
  reading tasks to the hard drive where possible. Casa Blanca Works
  no longer actively sells their SpeedyCD control panel to the
  public (it's part of the DriveCD [21] package of CD-ROM drivers
  that ships with many non-Apple CD-ROM drives), but if you can find
  a way to purchase it, you might consider it. SpeedyCD reads all
  the directories on a CD-ROM and caches them in a file on your hard
  disk. That means that only reading from a file requires CD-ROM
  access; getting information or cataloging the disk (or even a
  "Find File" operation) hits only your hard disk. It's great for
  multiple CD-ROM devices (like those multi-disc changers). Another
  popular alternative that's more supported is FWB's CD-ROM Toolkit.
  The older 2.5 version we're familiar with didn't do as good a job
  in caching all of a disc's directories as did SpeedyCD, but new
  version 3.0 [22] promises super-fine control over your discs,
  letting you create different settings of what information to cache
  for each separate disc and have them loaded on-the-fly. You also
  get benchmark programs to help you make such tuning. CD-ROM
  Toolkit also works with the multiple-disc devices, but the
  "Jukebox Edition" [23] required for this costs extra ($99 for the
  Jukebox Edition compared to $49 for the regular version; $29 to
  upgrade to the regular CD-ROM Toolkit 3.0.1 compared to $39 to
  upgrade to Jukebox 2.1). We have no idea if SpeedyCD works with
  Mac OS 8.1 or not; CD-ROM Toolkit 3.0.1 does as long as you don't
  try to store the cache files on an HFS+ volume. Happy caching!

  [21] <http://www.proline.com/drivecd.html>
  [22] <http://www.fwb.com/cs/cdt/main.html>
  [23] <http://www.fwb.com/cs/cdt/cdtjuke.html>

**Transparently False**

Q In the Post-Macworld Expo Product Extravaganza (MWJ_
  1998.01.18), you mentioned that FreeHand 8 is getting "rave
  reviews" for transparency support. You also noted that
  transparency hasn't been present in a vector-based program before
  PostScript Level 3 because PostScript doesn't support it. So does
  PostScript Level 3 actually have alpha channel support? I have
  other vector-based design programs that can include some level of
  transparency, but it's lost when exporting to an EPS file, and I'm
  hopeful that PostScript Level 3 will fix this. But I heard
  transparency _wasn't_ in PostScript Level 3 after all. Now I'm
  confused.

A You're not the only one. First, although everyone is bound to
  say "PostScript Level 3" because Adobe spent years drilling the
  phrases "Level 1" and "Level 2" into our heads, the actual and
  correct product name is "Adobe PostScript 3."  Just so you know,
  but forgive us if we slip up and refer to "Level 3" from time to
  time.

  We've already slipped up once - PostScript 3 does _not_ have
  transparency support. PostScript is an "ink-on-paper" imaging
  model - when an object is "drawn" onto a page (or a screen, using
  Display PostScript), it's imagined to be drawn in a perfect ink
  that perfectly overlays whatever was already there. The page is
  typically assumed to be white, so to erase something, you draw in
  "white ink" on top of black ink, and that erases it. Of course, a
  PostScript printer doesn't really have all these colors of ink -
  it's just that changing the color of a given pixel always
  completely replaces the old color with the new one, just as a
  perfect ink would overlay an older ink and exhibit none of the
  previous color.

  This pretty much eliminates the hope of managing any transfer
  modes (MDJ_ 1996.11.12), a drawing concept where the existing
  contents of the destination pixel are combined somehow with the
  source pixel to produce the final result. For example, if you add
  the source pixel value to the destination pixel value and you draw
  a purely-red pixel on top of a purely-green pixel (in RGB color,
  of course), you get a purely-yellow pixel. (Try this in the color
  picker if you don't believe it - add 100% red, 100% green and 0%
  blue.)

  But this kind of transfer mode capability (which is really much
  more powerful than this) is exactly what PostScript would need to
  treat objects transparently. Right now, all such effects are done
  on the computer with QuickDraw, and sent to printers as giant
  bitmaps. QuickDraw GX has even more extensive transfer modes, but
  QuickDraw GX printing is kaput and the PostScript translation
  library doesn't handle all transfer modes in vector form, as we
  noted a couple of weeks ago (MWJ_ 1998.01.26).

  So where did we get the idea that PostScript 3 supports
  transparency? Well, we muddled some of our concepts a bit. One
  thing PostScript 3 does support, finally, is a way to mask bitmap
  imaging operations with a bitmap mask. In PostScript Level 2 and
  earlier, just like in QuickDraw, imaging programs can use a "mask"
  to keep a drawing operation from affecting all the pixels in an
  otherwise-changed area - think of a mask like a cardboard stencil
  letter you use while spray-painting, so the paint goes on the
  cardboard and leaves a nice, letter-shaped, paint-free area on the
  sprayed surface. Masks are important to graphic artists, but
  PostScript has only had the capability to use masks that are
  described as vector images. If the vectors don't calculate to
  match exactly the pixels the artist wants, the mask is useless.
  PostScript 3 adds the ability to use a bitmap mask so the imaging
  operation affects exactly the desired pixels.

  The LaserWriter 8.5.1 driver (or the AdobePS 8.5.1 driver), as
  mentioned in MWJ_ 1997.12.01, takes advantage of this on
  PostScript 3 printers. QuickDraw also allows the use of bitmaps as
  masks in bitmap drawing operations, but that hasn't done the
  LaserWriter driver much good. The masking bitmap can be turned
  into an almost-completely-accurate vector-based path for use as a
  mask within the printer, but PostScript Level 1 limits such paths
  to 1500 segments, and that's not enough for some masks. PostScript
  Level 2 can do better, but will also fail in low-memory
  situations. So the LaserWriter driver has always ignored the mask
  parameter, drawing things as if the mask was empty. It's the
  addition of this new bitmap mask capability in PostScript 3 that
  raises the probability the mask will succeed, so now the
  LaserWriter driver uses it where possible. This can help
  applications achieve transparency effects, but is not transparency
  in itself. Our apologies for the confusion.

**On the Dotted Line, Please**

Q I'm feeling rather obtuse regarding public key cryptography:
  I re-read MDJ_'s _MacCyclopedia_ article on the subject (MDJ_
  1997.04.03), but I'm still missing something. How exactly does
  having GCSF's public key and software like PGP Freeware allow me
  to authenticate a digitally-signed issue of MWJ_? For example, how
  does PGP know if someone altered the E-mail message in transit,
  then copied the encrypted signature to the altered file?

A We'd hate to be responsible for an acute case of obtuseness.
  Let's get into some details.

  There are two concepts to note in trying to understand digital
  signatures.  The first one is that for the purposes of this
  discussion, public keys and private keys are interchangeable. 
  That is to say, if a message is encrypted with one key, the other
  one will unlock it.  You are not _required_ to use your public key
  to encrypt and your private key to decrypt, but you only want one
  of the two keys in the public view.  But remember that either key
  can be used to lock, and the other one to unlock.  You just can't
  encrypt _and_ decrypt with only one key of the pair.

  The second concept to understand is the idea of a _checksum_.  All
  computer data is stored as bits and bytes, which can be
  interpreted as numbers.  It doesn't matter if a given set of bytes
  is supposed to be interpreted as a GIF image of your Aunt Judy's
  wreck of a car - you _can_ interpret it as raw numerical data. 
  Normally you wouldn't want to do this, but it can be quite
  valuable.  A checksum is a way of condensing a large set of data
  into a smaller number.  The point isn't compression, because the
  number can't be turned back into the original data.  The point is
  verification - if the same data produces the same checksum every
  time, you know it hasn't changed since the first checksum
  calculation.

  Let's take a simple phrase and compute an absurdly simple one-byte
  checksum.  We'll add up the ASCII values of the letters in the
  sentence "AppleScript is user-friendly!" like so:

  A     =  65
  p     = 112
  p     = 112
  l     = 108
  e     = 101
  S     =  83
  c     =  99
  r     = 114
  i     = 105
  p     = 112
  t     = 116
  <spc> =  32
  i     = 105
  s     = 115
  <spc> =  32
  u     = 117
  s     = 115
  e     = 101
  r     = 114
  -     =  45
  f     = 102
  r     = 114
  i     = 105
  e     = 101
  n     = 110
  d     = 100
  l     = 108
  y     = 121
  !     =  33
  ___________
  Sum:   2797


  Since a byte can't hold a value larger than 255, we'll divide the
  large total by that number and use the remainder, giving us 247. 
  There's no way to turn that "247" back into the message we got,
  because most of the information in the message was discarded. 
  However, if the message changed to "AppleScript isn't user-
  friendly!", look at what would happen:

  A     =  65
  p     = 112
  p     = 112
  l     = 108
  e     = 101
  S     =  83
  c     =  99
  r     = 114
  i     = 105
  p     = 112
  t     = 116
  <spc> =  32
  i     = 105
  s     = 115
  n     = 110
  '     =  39
  t     = 116
  <spc> =  32
  u     = 117
  s     = 115
  e     = 101
  r     = 114
  -     =  45
  f     = 102
  r     = 114
  i     = 105
  e     = 101
  n     = 110
  d     = 100
  l     = 108
  y     = 121
  !     =  33
  ___________
  Sum:   3062


  Using the same remainder concept, we find that 255 times 12 is
  3060, so the remainder this time is two.  And behold - the
  checksum is different for a different message.  This is a very
  simple example - if someone knew this checksum algorithm, they
  could add enough extra characters (perhaps gibberish) so that the
  value of added letters was exactly 256, making the checksum
  verify.  You'd probably ignore a few garbage characters in a
  message, thinking it was just a typo.

  _Good_ checksum algorithms take extra steps.  They shift the bits
  in the checksum around between bytes, or they include some other
  form of multiplication, and they almost always use a larger
  checksum value.  A one-byte value maxes out at 255, but a four-
  byte value maxes out at just over 4.2 billion - that's a lot of
  characters to add or change to make things work.  In a good
  checksum algorithm, the results of the checksum will ideally be
  uniformly distributed across the range of all possible checksums,
  so a one-byte checksum has a one-in-256 chance of randomly
  matching any other message, and a four-byte checksum has a one-in-
  4.2 billion chance of randomly matching any other message.

  The good checksum algorithms are used in products like StuffIt -
  each archive has a checksum in it, computed when it was stuffed
  and recomputed when it's unstuffed.  If the two computations don't
  match, StuffIt tells you there's a problem with the archive.  It
  doesn't know _what_ the problem is - all it knows is that the
  checksum isn't right.  The data could be fine and the checksum
  corrupted, for all it knows.  But compression algorithms leave
  little room for error, and the checksum is an important way to
  know if something has gone wrong in transit.

  You may now see how these two concepts make a digital signature. 
  When it's time to sign a file or a message or another form of
  data, PGP uses strong algorithms to compute a good checksum for
  the message - one that's incredibly unlikely to be accidentally
  computed by any message that's even remotely-similar.  The
  checksum is then added to the message, with some other information
  such as when the message was signed and probably the length of the
  message.  Then the appended checksum and other information are
  encrypted with your _private_ PGP key.  That's not how it usually
  works: normally messages are encrypted with the recipient's public
  key, so only the recipient can decrypt them with his or her
  private key.  In this case, however, the only key that can decrypt
  the message is your _public_ key, which all concerned parties
  presumably have.

  When the signed message, like this issue of MWJ_, arrives in your
  mailbox, PGP sees the digital signature block and decrypts it
  using the GCSF Public PGP Key [24].  That gives PGP the checksum,
  which it then verifies by recalculating the message's checksum. 
  If the two match, you're certain that no one tampered with the
  message en-route.

  [24] <http://www.gcsf.com/pages/gcsf/gcsf_keys.html>

  Why?  Only you can create your digital signature because the
  checksum is encrypted with your private key.  Anyone can decrypt a
  signature and see the checksum if they like, and maybe they can
  compute a new checksum for it - but they can't encrypt it with
  your private key, because only you have it (presuming you take
  reasonable physical security precautions).  If the encrypted
  information also includes the length of the message, an attacker
  would have a _very_ hard time computing a different message with
  the same length _and_ checksum, which is what he'd need to leave
  the signature intact but change the message.  That keeps the
  chances of the signature being forged almost infinitesimally
  small, and that's good enough for government work.


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

  MWJ_, The Weekly Journal for Serious Macintosh[tm] Users, is
  published by GCSF, Incorporated.

  Publisher:  Matt Deatherage

  Please address _all_ correspondence to <mwj@gcsf.com>.  Due to
  circumstances beyond our control, mail at this address is read
  infrequently and responses may not be quick.  Phone lines are open
  between 10 AM and 6 PM Central Standard (US) Time, Monday through
  Friday, and Fax machines are available at any hour, as is
  voicemail.

  This file is formatted as setext.  For more information, send
  email to <setext@tidbits.com>.  A file will be returned shortly.

  This file is digitally signed using PGP technology to verify the
  integrity of the transmission.  Our DH/DSS corporate PGP key may
  be obtained at
  <http://www.gcsf.com/pages/gcsf/gcsf_keys.html#Anchor-GCSF_DSSKey>.

  Copyright (c) 1998 GCSF, Incorporated.  All rights reserved.

  All trademarks are the property of their respective holders and
  owners.

  GCSF, Incorporated.
  P.O. Box 1021
  El Reno, OK  73036-1021
  (405) 262-1399
  Fax:  (405) 262-1560
  <mwj@gcsf.com>

-----BEGIN PGP SIGNATURE-----
Version: PGP for Personal Privacy 5.5.3

iQA/AwUBNN8/87H4QMSEyVHCEQItnACfZN6ELAIYR5vZWZ+aG+gne842XdQAoIMP
NdM9faXepVgbii+TIsZ2mcQS
=xaq1
-----END PGP SIGNATURE-----