Microsoft vs. the Rest
From Stan Liebowitz in response to The Rise of the
Microsoft Monopoly:
You ignore the fact that WordPerfect did a bad job creating a
Windows word processor and Lotus did a bad job as well. And it wasn't
the 'secret APIs' since Ami PRO had a much better word processor with
access to the same documentation. Same thing with 1-2-3, where
Borland's product was better in Windows. If WordPerfect had done as
good a job as Ami and if Lotus had done as good a job as Borland things
might have gone differently. Secret APIs might have nothing to do with
it.
Stan,
Thanks for writing - and for providing the raw data
behind your graphs. You paper goes to great lengths to compare user
ratings for the various competing programs. WordPerfect was the top
rated word processing program from 1987 through 1992, and Word for
Windows caught up with it in 1993. The huge drop in WP ratings from
1992 to 1993 can probably be attributed to it being a DOS program in a
market increasingly dominated by Windows, as WordPerfect itself hadn't
really changed.
According to your
research, it wasn't until 1994 that WordPerfect for Windows caught
up with Microsoft Word, and by that time Microsoft had over 60% of the
market.
In a similar way, Excel
displaced Lotus 1-2-3 as the highest ranked spreadsheet in 1993,
and it has been top rated ever since.
Microsoft had a huge advantage in both markets.
WordPerfect and Lotus had developed their own unique interfaces and
keystroke commands in the keyboard-driven DOS era. Word for Windows and
Excel were developed for two graphical environments, Mac and Windows,
so they were designed from the ground up to use a mouse and standard
commands. It's easier to switch between programs that have many
identical commands than between two completely different environments,
which is what WordPerfect and Excel represented.
The problem for both companies was not wanting to
alienate long-term users while porting their most important programs to
a whole new environment. Microsoft had two big advantages here: as a
Macintosh developer, it new the importance a consistent user interface,
and as the developer of Windows, Excel, and Word, it was better able to
make sure that apps and operating system worked very well together.
Thus it's no surprise that it took Lotus and
WordPerfect some years to achieve the same kind of user ratings that
Excel and Word for Windows had achieved. And by that time, the
handwriting was on the wall for 1-2-3 and WordPerfect.
Dan
Another Factor: Windows 3
From Ian Baker:
A factor which you did not mention but which I had always believed
to be a major reason for MS Office doing so well was the surprise
launch of Windows 3. MS and IBM had been working on a joint OS (OS/2?),
which promised all the advanced features Windows 2 and 3 lacked. I
thought the rest of the industry had been developing for OS/2, not
knowing that MS intended to kill it. As a result, when Windows 3 came
out only MS had a word processor and spreadsheet ready for it.
Regards,
Ian Baker
Ian,
Good memory. OS/2 was a great way for Microsoft to
benefit from IBM technology, much of which made its way into Window.
Each company had different goals: IBM wanted to use OS/2 to drive sales
of its personal computers, software, printers, and other hardware. To
that end, IBM specified that its APIs be incompatible with Windows 1.0.
And they insisted that it be optimized for the dead-end 80286 CPU, not
the forward thinking 80386. (Windows 3.0 was compatible with the 80286
- as well as the older 8086 and 8088 - but it could take full advantage
of the 80386.)
Microsoft wants an operating system they could license
to manufacturers, and in the end it was probably the bulk licensing of
Windows 3.0 to PC makers vs. OS/2 as an expensive OS not bundled with
PCs that sealed the fate of OS/2.
Dan
What About ClarisWorks?
From Mike Bohan:
Hi Dan,
I agree with the article, but was wondering why it didn't mention
ClarisWorks/AppleWorks on the Mac side? When I started with my first
Mac at home in Dec. 1993, a 660 AV
Quadra, I believe that ClarisWorks was bundled with it. About a
year later, when I switched from an IBM PS/2 Model 60 to a 6100/60 Power Mac with SoftWindows
at work, I used ClarisWorks for many years for both word processing and
spreadsheets. I only switched to Office when it became the official
standard at work.
I work in a Windows shop and needed to easily exchange documents.
(I've been hiding out from our IT Dept. for many years. I'm a physicist
and know my way around computers, so I tell them that I need my Mac for
scientific purposes. They tell me that if I don't call them for support
or connect to their network, they'll leave me alone.) I still use
AppleWorks as a painting and drawing application and would be perfectly
happy to use it for word processing and spreadsheets, if it was
updated. Unfortunately, because of Office's dominance, I haven't tried
iWork yet. I've gotten attached to tabbed spreadsheets in Excel and
AppleWorks doesn't support it.
P.S. - My first personal computer in 1982, was a DEC Rainbow
100A, with dual CPUs (Z80 & 8088), 400 KB dual floppies, and
triple boot (CP/M, CP/M-86 and MS-DOS with VT-220 terminal mode). My
software standards from that time were Wordstar, Multiplan and dBASE
II. I even saw a DEC engineer running Unix on one, at a local DEC user
meeting at the time. Now that was a computer! I was never too happy
with my IBM PC and only got that old DEC, warm and fuzzy feeling back,
when I switched to the Mac.
Regards,
Mike Bohan
Mike,
Thanks for writing. I remember the Rainbow 100, a
machine that ran CP/M well but wasn't exactly IBM compatible on the DOS
side. One of many better ideas - alongside the Heath/Zenith Z-100, TI
Professional PC, etc. - that got run over by the IBM standard.
I didn't cover ClarisWorks in the article because it's
one area where Apple ran over Microsoft. From version 1.0, ClarisWorks
made Microsoft Works obsolete - and then Apple ported it over to
Windows. I still use and love AppleWorks. I've tried Pages and Numbers,
and I think they would suit me perfectly, but I just can't see spending
money on them when AppleWorks is still doing the job.
Dan
Who Owns WordPerfect?
From Rick Mansfield:
Dan, you mention in your newest post that Novell owns what is left
of WordPerfect, but actually WP is now owned by Corel.
Rick Mansfield
Rick,
Novell bought WordPerfect in June 1994 and sold it to
Corel in January 1996. However, Novell retained ownership of the
WordPerfect Office technology and was the publisher of WordPerfect when
it was killed off by Word for Windows. That's why Novell, not Corel, is
suing Microsoft.
Dan
Shafted by Gizmos2go.com
From Jeffrey Porrello:
Dear Dan:
I found a link on your site <gizmos2go.com> which you stated
had the lowest price on Mac OS X 10.0. I went ahead and
ordered OS X 10.0 from them, and they sent me update 10.0.3. They
have no phone number. They do not respond to emails and they make up
for their "low price" by charging high shipping costs.
I went on line to read some reviews about Gismos2go, and I only wish
I had read them before. Those who had problems have had the worst, most
distressing problems possible, and the Better Business Bureau rates
them as "F" "a good reason not to buy online"!!! They have changed
their business name many times!
They do not refund shipping for any reason, even when it is their
error, and only give store credit with an abysmal 7 day return
policy.
Please remove their reference or at least warn people.
I have been taken.
Sincerely,
Jeffrey L. Porrello
Jeffrey,
I'm sorry to hear that, and I've removed both links to
Gizmos2go from our website. It's hard enough finding dealers who even
list old versions of the Mac OS, and I made the foolish assumption that
what they listed was what they were selling. The listing says nothing
about it being an upgrade.
Had I investigated further, the company's "no refunds"
policy would have raised a red flag. Gizmos2go will never send money
back, only exchange defective merchandise or give you store credit. If
you paid by credit card or debit card, I urge you to go to your bank
and reverse the charge. If you sent payment through the mail, go to
your local post office and file a mail fraud complaint.
Dan
Is a 2003 PowerBook G4 Worth More than a 2005 iBook
G4?
From Brian Troisi:
At macinthebox.com, there is a
15" 1.5 GHz PowerBook G4
(512 MB RAM, 80 GB HD, and DVD burner) for $750. But at applepalace.com,
they have a mid 2005
iBook G4 1.33 GHz (640 MB RAM, 40 GB HD and Combo drive) for $625.
What do you think the better deal is?
I wouldn't mind having a bigger screen. Do you think it is worth the
extra $125 for the PowerBook? Do you know of any specific issues
related to this PowerBook that would make me not want to buy it? Such
as the laptop being really hot or logic board issues? I know that in
previous messages you recommended an iBook over a PowerBook for
ruggedness and price, but now the price difference isn't that big, and
the PowerBook is really powerful and has a bigger screen.
Thank you very much for all of your help! I really appreciate it!!!
:-)
Brian,
The first place to look is the MacInTouch iBook and
PowerBook Reliability Report from Jan. 2006. The 1.5 GHz 15"
PowerBook is probably the Sept. 2003 model, which has a 35% repair rate
over 27 months on the market - but 70% of those were during the first
year. By comparison, the 12" 1.33 GHz iBook G4 has a much lower repair
rate at just 14% - but that was over just 7 months on the market.
Another factor is what needs repair. For the PowerBook, it was often
very expensive things like the logic board and display. For the iBook,
it was most frequently the optical drive, which can be replaced
inexpensively with a third-party drive.
Both models support Core Image, which is a plus, but
the PowerBook has twice as much video memory, supports up to 2 GB
of system memory (vs. 1.125 GB in the iBook), has a larger display (as
1280 x 854, it has 39% more pixels than the 1024 x 768 screen in the
iBook) that's widescreen, and has a PC Card slot. And it has a bigger
hard drive and a DVD burner.
In general, I'd say the PowerBook is easily worth the
20% difference in price based on features. Reliability is a bit more
nebulous, as the iBook was still current when MacInTouch took its
survey. I'd guess that any logic board or screen issues that might have
afflicted the PowerBook would already be taken care of, so I'd still
give the edge to the PowerBook.
Dan
The End of PowerPC Software Development
From John Muir:
Hi Dan,
I just read this and immediately thought Low End Mac should hear
about it: When do we stop
making 'em fat?
Brent Simmons is a well known Mac developer, as you may know, behind
the platform's leading RSS news reader NetNewsWire.
He's basically advocating cutting PowerPC out of the picture sooner
rather than later.
I've already seen the Omni figures he links to, but I think the
browser stats below represent the wider picture among Mac users:
Mac OS X and Safari gain more market share
He argues that people with newer hardware are more likely to be
customers, so Omni's very strongly Intel figures indicate where
developers should target their apps. I'm not so sure about that, but I
knew you'd want to know!
John
John,
Thanks for writing and sharing these links. According
to the second article, Intel-based Macs went from 32.8% of Macs on the
Internet in Feb. 2007 to 56.9% on Jan. 2008. Macs overall went from
6.38% to 7.57% of Internet user share, and PowerPC Macs are on the
descent. PowerPC Mac user share on the Net has declined by 24% over the
same period.
If we extrapolate this data, Intel-based Macs will
account for well over 6% of all Internet users in January 2009 and
PowerPC Macs will probably drop to around 2.5%. Intel Macs may go even
higher, since for a good part of 2006 Apple was still selling PowerPC
Macs, so 7-8% isn't unattainable. Between them, both types of Macs
could account for 10% of Net usage within a year - and Intel Macs would
account for 75% of them.
Simmons isn't suggesting that developers abandon
PowerPC Macs yet, although he anticipates that could be the case within
a year. He also says that if he were developing a new app, he'd
probably go Leopard only.
At Low End Mac, we're fans of people who keep
supporting old stuff - like iCab, whose
version 2.9.9a (finalized May 2006) supports 680x0 Macs and System
7.5.x, and 3.0.5 (Jan. 2008) supports all PowerPC Macs running Mac OS
8.5 and newer. Even the current build of iCab 4.0 reaches back quite a
ways, supporting Mac OS X 10.3.9 and later.
From a dollars and cents standpoint, it doesn't make
sense to develop for old Macs, as the number in use will continue to
decline. You have to be passionate to develop for old, discontinued
hardware, and we applaud those who continue to do so.
Yes, the day will come when most developers abandon
PowerPC, but I don't expect to see that happen in a big way until Mac
OS X 10.6 comes out in another year or so.
Dan
Dan,
It's self-evident that Mac developers will follow the user base and
target PowerPC less and less. It's just the timing which seems quite
stark. Clearly if Mac OS X itself drops PowerPC, that's the moment
when developers will too in great number. Leopard still supports a
great many PPC models quite nicely for the moment though - the G3 and
867 MHz debate notwithstanding - and I think Simmons may be floating
his opinion just to try the water. Certainly, as someone learning their
way through Cocoa and Xcode myself just now, Leopard's technologies are
what I'm concentrating on. Perhaps turning down PowerPC is part of the
same equation. I just don't think it's wise until OS X does
too.
If you think that 10.6 is likely to be Intel only - and that Apple
are going to bring it out next year or so - I'd consider that even more
of an incentive to promote Intel Macs over PowerPCs. If a machine like
the Power Mac G4 or G5 is doomed to "obsolescence" - i.e. no longer
supported by the current operating system and most new software - that
must surely play a role in the balance.
I'm hoping that the PowerPC continues to see full support for a
while yet. Intriguingly, the iPhone SDK tools are all Universal
Binaries and run just fine on my PowerBook. But the PPC platform's fate
is naturally to be left behind now that Apple do not sell them. It's
all a matter of time, and judgment.
John
John,
I anticipate Apple will be resuming the 18-24 month
cycle for new versions of Mac OS X. Tiger was an anomaly; Leopard was
delayed about 6 months to provide more programmers for the iPhone. So
anywhere from May through November 2009 is what I'd predict for
10.6.
Intel only? I don't think so, as Apple was still
selling PowerPC product well into 2006 - notably the top-end Power Mac
G5. The bar will be raised again, but I would be surprised and very
disappointed if the next version abandons PowerPC. I suspect we'll see
that with 10.7 sometimes in 2011.
Dan
Mac Ethernet Problem
From Geoff Phillips:
Good news! I got the Power Mac in, and it is in fact a dual 450 as
reported in the auction. The guy currently has a few other apple
machines and displays up there for good prices, and I can vouch for his
good packing and fast shipping. So I got a dual 450/512/20/dvd/clean
install of 10.4 for $150 with shipping. Not bad!
My one issue is that it takes the thing forever to turn on the
ethernet. When it first boots, the computer is completely usable, minus
the ethernet connection for about 5 minutes. Then it kicks on and works
without a problem. Seems to work fine coming out of sleep as well. Any
ideas? If that's all that's wrong with a $150 computer I'm happy, as
PCI ethernet cards are cheap enough.
Thanks for your help and for running this spectacular website!
Geoff Phillips
Geoff,
Glad to hear it! Macs can be terribly unreliable on
10/100 ethernet networks using the built-in ethernet port when there
are a lot of Macs on the network, some 10Base-T and some 100Base-T -
and the more
Macs on the network, the worse the problem becomes. When I worked at
Baker Book House, we ended up putting third-party ethernet cards in all
of our PCI Macs because of this problem. It might solve your problem,
and PCI ethernet cards are cheap these days.
Dan
Text in italic type was updated 2007.03.27 to clarify
that the problem requires a good number of Macs and the presence of
both 10Base-T and 100Base-T Macs on the network.
Dan Knight has been publishing Low
End Mac since April 1997. Mailbag columns come from email responses to his Mac Musings, Mac Daniel, Online Tech Journal, and other columns on the site.
