In article ,
"Pete C." wrote:
Joseph Gwinn wrote:
snipped
On that, you are fundamentally wrong. They are indeed both computers,
and regardless of the UI on top, any even remotely useable OS operates
on the same fundamentals. If you understand one OS, you understand
essentially any other, it is only the UI that really differs.
It's true that down deep they must all do the same thing, but this level
is quite remote from ordinary users, who tear their hair trying to
figure out under which GUI rock the needed control is hidden. Or even
which control or controls cause the current annoying misbehaviour. And
Microsoft has a different theory of rocks than Apple, so if you have
spent too little time with such a GUI, it will all be so very
frustrating.
It's more a function of Apple having an incorrect theory that rocks
don't matter. When you know what is causing the problem, but the Mac UI
won't let you fix it 'cause Apple doesn't think anyone needs to adjust
that "rock" then there is a fundamental problem with the UI.
This isn't a problem I've encountered. Could you give some specific
examples?
Apple's usual strategy is the make the technical controls invisible to
the ordinary user, but one can do damn pretty much anything from a
terminal window. Or, if one enables it, the root console.
On every attempt to actually accomplish anything on a Mac (usually
trying to help a Mac user who couldn't figure it out either) I have
consistently found that the language, structure and in many cases simply
the existence of proper configuration options was a significant issue on
Macs. These were not simple UI differences.
Um. I have no idea why you have such problems, but theorize that it's
simple lack of sufficient experience with the Mac GUI.
Experience with the Mac UI doesn't explain their mangling of the english
language or their complete mislabeling of some configuration items (I
seem to recall them calling encryption keys "passwords"), or having
certain settings that are part of a standard completely missing.
In other words, Apple jargon differs from Microsoft jargon? This is
sort of like complaining that Americans don't use quite the same words
as Brits.
I do however have several friends that use Macs to
varying degrees and all have had plenty of problems. One friend is a
teacher who uses both Macs and PCs extensively and reports that the
Macs
crash at least as often as the PCs.
With students messing with them? In education, that has been the
prime
problem. Schools have always liked Macs because the teachers could
keep
them running without needing an IT guy.
This teacher *is* an IT guy and reports no difference in the frequency
of crashes between the PCs and Macs.
My point was that most teachers are *not* IT guys, and are happy that
way.
That's nice, however it has nothing to do with the fact that an IT
literate teacher with significant experience with both Macs and PCs
reported no difference in the rate of crashes between the two.
It that teacher isn't the only one mucking with the computers, one would
see just this pattern. In fact, students mucking things up is *the*
problem that even university IT shops live with, never mind grade school.
Another friend uses Macs almost
exclusively and in 5 years and like three Macs she had a ratio of
about
20:1 to the Windoze problems I had during that time. I did not see
any
decrease in the frequency of problems with the switch to OSX either.
My experience was and is the exact opposite.
Which only goes to show that the stability of either system is most
dependent on the operator, not the OS.
I don't know what she was doing, but clearly you are far more the IT
guru than she. An aggressive or merely clumsy user with admin privilege
can make themselves lots of trouble.
Indeed that is what I've concluded. The myth that Macs are more stable
than PCs is simply that, a myth. I've also noticed that many Mac users
seem to under report the number of system issues, somehow not counting
the need to reinstall an application to get it to work properly as a
system problem.
So we return to the mainstream problem, that Windows systems are far
more fragile in practice than Macs, once one removes the effects of
clumsy meddling. PCs isolated from the world actually work tolerably
well these days, once correctly set up, but how many people want to be
isolated, or to spend their time managing multiple anti-virus and
anti-spyware programs?
MacOS was total crap up until Apple finally
realized they lacked the expertise to write an OS and put their
UI over
someone else's Unix core. Now instead of being a crappy UI on top
of
a crappy OS, it's a crappy UI on top of a so-so OS.
Don't mistake me for a Windoze bigot either, ...
Could have fooled me. Listen to yourself, listen to the music.
How do you figure that? Anyone with any technical knowledge knows
that
the pre OSX versions of MacOS were hopelessly deficient in many
areas,
particularly the lack of memory management. OSX fixed many of the
core
problems, but the UI that I can't stand (I hated the UI on the first
Lisa as well) remains. If I wanted an alternative to Windoze it
certainly would not be Mac as there is simply no advantage whatsoever
to
MacOS over Linux or another Unix variant.
I submit that your answer above proves my point in spades. Listen to
the tone of voice, and parse the implicit assumptions.
Huh? Hardly. I find the Windoze UI vastly more tolerable than the Mac
UI, largely because I can customize the Windoze UI sufficiently to
eliminate the most annoying parts. This does not in any way indicate
that I am a Windoze fan or bigot, simply that I hate the Mac UI. My OS
preference is VMS, however there is a bit of a shortage of affordable
applications for the things I do.
You really don't hear it, do you? OK, I'll parse it a little:
We'll set the stage with such dispassionate, value neutral statements
like "MacOS was total crap until Apple finally realized they lacked the
expertise to write an OS" - It may be crap, but 25 million users rather
like it, and were known to say similarly unemotional things about DOS
and Windows.
How is an OS that had -no- memory management until the entire OS core
was scrapped and replaced with a Unix core not crap? Windows was
evolving memory management (which I consider to be a fundamental concept
for an OS) in the Win 3.1 days and had it working reasonably well long
before Apple gave up on their OS core.
Because neither original Windows (pre NT) nor original MacOS (pre 10)
had memory management.
Microsoft solved the problem by stealing VMS technology form DEC,
yielding the original NT core. Books were written about this deathmarch.
Apple solved the problem by buying NeXT (getting Jobs back as part of
the package deal). NeXT was built from scratch by Jobs after he was
tossed out of Apple some ten years earlier. The OS core of NeXT was BSD
UNIX, from which Jobs and company built a through object oriented
software platform. When NeXT came back, the solution was obvious -
replace the NeXT GUI with a Mac-like GUI, retaining the UNIX core from
NeXT.
And will end with "Anyone with any technical knowledge knows that...".
In other words, anyone who disagrees by definition cannot have any
technical knowledge. Aside from the implicit ad hominem attack, this
assumes the truth of the very thing to be demonstrated, and thus is
circular.
Huh? It doesn't take a lot of technical knowledge to realize the
significant failings of pre OSX MacOS. While people complained about
Windoze bloat and the need to throw more CPU and memory at Windows, the
Mac world somehow accepted the need to throw more memory at a Mac, not
because of bloat, but because there was no memory management.
I give up. The above response is not relevant to the issue raised. You
don't seem to be able to separate technical issues from emotional issues.
Ah. Now we come to the core. Keep your machine away from the internet,
and all is well. Well, Macs don't need to be protected against the web.
Huh? Where did you come up with that idea? Every one of my machines has
Internet access.
Downloading stuff (including napster) is very much a part of the net.
It shouldn't be possible for this to cause such problems, even if some
users are naive and some people out there are evil, because such people
have always existed, and always will.
That is perhaps one of the most absurd statements I have ever heard.
Nowhere else in life do people have such an absurd expectation that they
should be magically protected against their reckless actions.
Go walking through a dark alley in the bad part of town at night and you
will probably be mugged and nobody will say that shouldn't be possible.
Hop in your car and go careening down the road ignoring safety rules and
you're going to get in an accident and nobody will say that shouldn't be
possible.
Bad analogy. A better analogy would be to ask if you expect the Police
to keep bad people from coming to your town and mugging people, or
breaking into their homes.
To say that it shouldn't be possible for a users careless actions to
cause problems on a computer is utterly absurd. If you want that level
of "big brother" protection, then your Mac would simply prevent you from
downloading programs like Napster that had not been certified by Apple.
No way do I want some company, especially Microsoft but even Apple, from
deciding what software I can and cannot run.
Most users are experts at something other than the computers that they
must use as a tool. It's expecting far too much to think that this will
ever change, if for no other reason that by definition the clueless
vastly outnumber the priesthood. But the money from vast unwashed
clueless masses folds and spends pretty good, and keeps the priesthood
in beer.
The fact is that whether you are using a Mac or a PC, downloading (and
running) questionable junk such as Napster *will* cause problems and I
have seen plenty of examples of this on both platforms. The Mac provides
no more protection from these careless user actions than Windoze does.
Tell me, would using a VMS host to download stuff from even the worst
sites cause any danger? If not, why can't Windows do that too?
For whatever reason, Macs are immune to this stuff. You can argue that
it's mere insignificance that protects the Macs, but the fact remains
that Macs are immune.
Your belief that Macs don't need to be "protected" from
the 'net is also false.
Most Macs do not run with any virus protection whatsoever, and are none
the worse for it.
And that is simply a function of threat volume and statistics, not a
function of platform security. Because PCs running Windoze outnumber
Macs 20:1 the volume of viruses targeting Windoze outnumbers those
targeting Macs by an even larger ratio and the probability of a
particular Windoze user being hit by one of those viruses is
consequently higher.
Whatever. See above.
Do you not wear a seat belt in a Volvo because they are perceived as
safer? Do you not follow traffic safety rules because your Volvo will
protect you? You might get away that false sense of security for a while
just due to statistics, but you *will* get nailed eventually.
I wear a seat belt because they demonstrably reduce injuries, regardless
of the make and model of car in question. It's still a car, and an
unbelted person will collide with something in a crash.
But how does this analogy apply to computers? In Macs, no such "seat
belt" is needed.
That has nothing to do with "isolationism", it has to do with product
quality. Do you purchase brake shoes for your car from some guy in a
dark alley? Would you expect them to be safe? Why would you expect any
different if you get your software from equally questionable sources?
I don't see the analogy. Are you claiming that Macs are bought only
with small unmarked bills from junkies in dark and fetid alleys? This
is quite the scoop - I always wondered about them.
Load garbage software (Napster et al) from suspect sources onto a
computer (Mac or Windows or any other OS for that matter) and you *will*
have problems just as surely as putting cheap counterfeit parts on your
car *will* cause problems. Even a top grade ultra secure OSs such as VMS
will have problems if a privileged user were to execute malicious code
on them.
Rehash. See above.
If you read the PC magazines (yes, PC magazines), you will see that they
consistently rate the reliability and quality of Macs at the top, with
Dell close behind. Apple beats all PC makers on quality of user
support. Consumer Reports backs these findings up.
Consumer Reports has -zero- credibility with me in -any- subject area.
As for reliability and quality, the PC magazines use some questionable
criteria and also exclude many PC lines from their comparisons. The same
goes for support as the comparisons typically exclude the "business"
lines from the PC manufacturers.
So, who do you believe?
Consumer Reports sends out a survey to their subscribers every year. In
this survey, the subscribers report their experience with all manner of
products, and this experience is summarized and reported.
Specifically, on page 35 of the December 2005 issue of Consumer Reports
appears the "Brand Repair History" (based on 85,000 desktop systems and
49,000 laptops) and "Tech Support" (based on 6,500 responses for
desktops and 4,200 responses for laptops).
There is no better publically available source of by-brand reliability
data. And the scores don't vary that much from year to year, as
reliability and tech support don't just happen, they are achieved if and
only if the company management thinks them important.
As for security problems, there are tens of thousands of viruses et al
for Windows, maybe ten for MacOS (none that still work), and
essentially zero for most flavors of UNIX.
There are many, many security problems that affect most flavors of
Unix.
Yes and no. While it's true that no commonly used OS can long resist
knowing attack by experts, some are far harder than others, and the
first-order question is resistance to automated attack.
Er, please qualify that with "consumer" OS as there are a number of "non
consumer" OSs that do just fine against all attacks. Try my favorite VMS
which can give you C2 qualified security "out of the box".
Um. It's true that most consumer OSs lack Orange-Book (DoD 5200.28-STD)
certification while most server platforms do have such certs, but why is
that important? Nor do I see the relevance of VMS in this discussion.
The relevance is that you indicated that no commonly used OS can resist
attacks by experts which is not true. There are many commonly used OSs
that withstand attacks quite well and VMS is one of them (although it is
unfortunately slowly becoming less common).
There is a big difference between "quite well" and perfection. No CAPP
certified system has failed to yield to knowing attack by experts,
although in some cases they had to work at it. Nor are 5200.28 B-level
machines completely safe. If you read DoD 5200.28-STD, you will see
that it does not promise that penetration will be impossible, it instead
implicitly promises that penetration will take a lot of specific OS
knowledge, time, and general expertise. The levels in 5200.28 basically
specify the level of penetration effort to be thrown at the system under
test.
Actually, the old DoD 5200.28 family of standards have been withdrawn by
the DoD, replaced by Common Criteria and DoD 5200.1 and 5200.2. The
formal equivalent to 5200.25 C2-Level is CAPP (Controlled Access
protection level) EAL (Evaluated Assurance Level) 3 or better.
Recent versions of Windows have CAPP EAL3 certs, as do two Linux
distributions. All the major UNIX platforms have EAL3 or EAL4. I
haven't checked MacOS, but I imagine that Apple will get or has gotten
the certs, just so they can sell to DoD. With a BSD base, it won't be
hard.
The fact that Apple had to scrap their entire OS core speaks volumes to
their software expertise. I can't see the DoD buying Macs for what
essentially is nothing more than a UI.
Microsoft scrapped their entire Windows 3.x OS core as well, in favor of
NT.
Apple would seek certification to be able to sell to DoD. If this will
work or not is quite another matter. But don't dismiss it out of hand:
There was a big flurry a few years ago when the US Army (at Ft. Monmouth
if I recall) dumped all their Windows webservers in favor of Mac
servers, mainly because the Army was tired of being hacked despite their
heroic efforts to keep the WinNT servers patched and running.
As for the vulnerability of Macs, that is a false sense of security
simply based on the volume of attempted attacks. If the virus kiddies
decided there were enough Macs to be worth attacking on any scale that
sense of security would evaporate very quickly.
While it's true that Macs are less of a target because they are a
fraction of the market, it's also true that Macs are harder to
compromise, especially by script kiddies.
Simply because the folks who write the Windoze attack utilities for the
script kiddies aren't spending much time writing attack utilities
targeting Macs.
Rehash. See above.
A lot of this is due to the
the fact that the security base of MacOS is BSD UNIX, and a lot is due
to the fact that most dangerous things in MacOS are locked down by
default, and/or require an administrator password to access. Windows
has just started to implement this, with fanfare.
This is nothing recent to Windoze, it has been common for a long time to
"lock down" Windows in business environments so that the users are less
able to compromise the systems.
The problem has been that a fully locked down Windows system is close to
useless, as many Windows apps won't run as anything other than admin.
This has just now started to change, but will take years to achieve what
MacOS now has.
As I noted earlier, careless users will cause problems on both platforms
and unless the Mac prevents the user from loading the non certified junk
(Napster et al) than the Mac is not going to "protect" the user. I think
even the Mac users would have a fit if their beloved Macs gave them the
"I can't let you do that" response when they tried to load the latest
music piracy tools.
See above.
Simply put, viruses et al are practical problems only for Windows.
Because such malware spreads itself, the problem grows exponentially
and far faster than systems can be attacked manually.
See above. If Macs were more than a single digit percentage of the
computing world the issues would be vastly different. The perceived
security of a Mac is a function of their scarcity, not their security.
Not quite; see above. And below.
And as a class, Macs and UNIX boxes are far harder to manually
compromise than Windows anything, but none are totally secure. Nothing
that complex ever will be.
I can't recall the last time I heard of a VMS or Tandem or Stratus
system being compromised.
By your own logic, this must be only because with their miniscule market
share compared to Windows (and the Mac for that matter), they just were
not worth the trouble to break.
Hardly, since those three OSs control a sizable portion of the financial
world.
Ah. A new issue emerges.
So, we should use only these machines for web surfing. And if they can
achieve safety despite naive users, why can't Windows do the same?
If you want a secure OS, look at VMS or the Tandem and Stratus OSs.
Oh my, a blast from the past. True enough. VMS was my favorite
command-line OS of that era. If Ken Olsen had had a religious
conversion and had made VMS open in time, he might have killed UNIX in
the crib. But it didn't happen. So, now VMS has the security of the
dead.
Digital, Compaq and now HP have had no clue how to market VMS, indeed
Comapq and HP have no concept whatsoever of the "enterprise class" world
outside the PC realm.
I'd have to agree here. But it's also true that the whole idea of
"OpenVMS" arrived about ten years too late, so it's not clear that
Compaq and HP could do anything to reverse DEC's blunder.
I think Compaq could have if they had understood what they had and the
enterprise world. HP was hopelessly screwed up with Carley running
around hyping every low margin consumer toy while selling off or
otherwise destroying the diverse underpinnings of the company.
I suspect that you are right.
Tandem and Stratus are still around I think, but sell into a very
specific niche, where perfect hardware reliability is needed. These
were used in some air traffic control systems, but have a key
conceptual
flaw - the custom-built application software is the common cause of
failures, not hardware failures. So most ATC systems have total dual
or
triple redundancy, and the hardware is just another (minor) cause of
failure and subsequent switchover (within one tenth of a second
typically).
VMS is still around as well. You don't hear about it much (like Tandem
or Stratus) because it's in applications that don't get hyped, and it
also "just works".
Software is most often the cause of problems and it's only getting worse
as the software gets both more complex and more poorly engineered.
Yes, but reliability is still a problem even if well engineered. It
usually takes a few years of intense post-delivery bug fixing to achieve
reliability.
Indeed, and this is something that MacOS is just as subject to as
Windoze.
Yes, they are still computers. But Apple seems to push it much farther
than Microsoft, and Apple has better control of the Mac ecosystem than
MS has of the Windows ecosystem.
Because MacOS is only for the clueless, it cannot be that the lack
of
trouble on Macs is due to clued-in users. So there must be some
other,
simpler explanation.
I have not seen this purported lack of trouble on Macs. Every single
Mac
user I have known (dozens) has reported plenty of problems.
You need a better grade of Mac users. By your own analysis, the
clueless make their own trouble; this will be platform independent.
Indeed, and this will likely be found in users who treat the machine as
a tool and not a toy. Since the Mac UI generally seems to appeal more to
the "creative" types vs. the "technical" types, presumably there are
plenty of writers and such that have Macs that are perfectly stable and
also devoid of questionable software. Presumably also if you gave these
same people the same applications on a PC and they treated it the same
(no questionable junk), they would likely see the same stability.
This is a bit self contradictory. Those flighty non-technical creative
types love the Mac but are clueless about IT, have no internet
discipline whatsoever, and yet they prosper.
I think the ones that "prosper" are the ones that keep work and personal
machines separate.
Should not be necessary, as mentioned above in multiple places.
Just think what stolid
uncreative technical types could do with such a tool.
Nothing, absolutely nothing, because the whole Mac concept is to prevent
anyone from doing anything technical.
Not so. The technical controls are at the terminal window (and root
console) level, and many controls are only at that level. One can argue
that this or that control should or should not be GUI accessible, and
I'm sure that there will be some migration in the coming years, but the
controls are there, but mostly kept away from naive users.
With the growth of Linux in the market, more commercial apps will
support
Linux, so this advantage is likely to erode over time.
The upcoming homogenization of the hardware market will help this a
lot.
The switch to OSX was one step towards Apple getting out of the
hardware
business which they have never been very good at. Now they have
announced they are abandoning IBM's antiquated CPUs.
The PowerPC architecture is hardly "antiquated", and is about twice as
fast per CPU clock cycle than Intel.
The PPC architecture has been around for quite some time and was a
rehash of a retired workstation processor. Neither the Intel x86 nor PPC
CPUs are remotely as fast / efficient as the (DEC/Compaq/HP) Alpha and
indeed that's why Intel stole much of the Alpha design for the Itanium
before eventually reaching a "settlement" over the theft.
Be careful about who you call a "rehash" (nice neutral word that):
Intel processors are by the same token an absolute hash, retaining
compatibility with every past version, with bits encrusted upon bits.
They had been until the infusion of stolen Alpha technology.
Not exactly. IBM invented the RISK processor architecture, and the
first RISC CPU was the IBM P801.
Nor is the Alpha really a RISC machine, as it had to execute the VMS
instruction set. I no longer recall the details of what was done in
silicon and what was done by the compilers, but the VMS instruction set
is pretty big and complex, even if it is nicely designed. My
recollection is that DEC implemented all the one and two address
instructions in Alpha silicon, and everything else was emulated.
Just like every ancient architecture. Did you ever write assembly on a
Univac 1100-series computer? They never threw anything away; every
instruction bit had three meanings, controlled by the current state of
the processor.
Nope, I don't go back that far. I did write assembler on VIC-20s and
II+s though. These days it's just the occasional stuff on something
small in the PIC family. I'm not a programmer and only write a little
assembler now and then to go with some hardware I built.
My first computer language was Basic, my second was Fortran, and my
third was Univac 1100 assembler. And embedded programmers almost
universally used assembly in those days, although C (not C++) is real
common nowdays.
The PowerPC is a clean new (in relative terms) design, with no
encrustations of prior architectures. That's why it's able to do twice
the computational work per clock cycle.
The stolen Alpha technology is bringing that to the Intel line.
Not exactly. See above.
And one part of the Alpha one would not wish to steal is the power
demand. I recall an effort to use Alpha chips in a line of MIL-SPEC
single board computers that failed because it was too hard to get that
much heat off the board and out of the cabinet. The comment of the day
was that we would have to cool the thing with boiling seawater.
The problem is that IBM is more interested in making large massively
multiprocessor servers the size of commercial refrigerators than little
desktop systems, and so IBM's direction increasingly deviated from what
Apple needed to win the CPU horsepower races.
More importantly Apple has been realizing that they need to get off
proprietary hardware which regardless of any technical merit, they can
never be economically competitive with. OSX was the first step towards
making their OS portable to a generic hardware platform. The
announcement of the switch to Intel CPUs was the next step. In the not
too distant future will be the announcement of MacOS for the PC,
followed later by the announcement of the end of proprietary Mac
hardware.
When the consumer is able to select a "generic" computer platform of the
size, scalability and fault tolerance for their application, and then
independently select from a dozen of so OSes depending on their
preferences, the consumer will be well served.
It's true that Apple is arranging things so they can take advantage of
the whole PC hardware ecosystem, but it does not follow that Apple will
allow the MacOS to run on generic PCs.
I guess we'll have to wait and see, but my Apple predictions so far have
come true. I was predicting many years ago that Apple would eventually
have to face the fact that they did not have the expertise to write a
good OS core and would have to take their UI elsewhere and this has
happened. I was predicting years ago that Apple would have to abandon
their hardware platforms where they were typically behind and this has
happened in stages (first PPC now Intel).
OK. The market will tell, soon enough.
The common hardware platform will both drive down the hardware cost and
also let each OS stand on it's own merits independent of hardware
differences.
This deviation was particularly acute in laptops.
Also, as part of their "fit in but stand out" strategy, Apple wanted
Macs to be able to run Windows apps at full speed, rather than in
emulation at a fraction of full speed.
The need for emulation / Windoze support of course being a function of
market share. Few companies can afford to write Mac only software and
ignore 95% of the market.
Almost true. There are a few Mac only companies, but a common pattern
has been to develop their first product on the Mac (where the smaller
market means less competition and greater margins), and use the profits
from the Mac market to fund the launch into the much larger Windows
market.
True for the Mac-centric companies. For the Windoze-centric companies
it's (unfortunately for the Mac users) quite easy for them to stick with
the Windoze versions and ignore the small Mac base.
Yes and no. There are people switching. Like my kid sister. And
frustrated Windows users ask me from time to time, but it's more often
an initial choice than a later choice.
The PowerPC architecture (from both IBM and Motorola) basically rules
the military and industrial embedded realtime markets, with something
like 70% market share.
Not sure where you got that figure, I follow the embedded world to some
extent and I see very few PPCs. In fact, a flip through the Dec '05
Circuit Cellar magazine revealed -0- references to PPC.
Um. Circuit Cellar is for hobbiests, not the military industrial
complex.
Hardly. Circuit Cellar is for the embedded engineering world, but uses a
format with the hobby projects of those embedded engineers to highlight
a lot of the new stuff and keep the magazine interesting. As a useless
side note, I've provided some of the props and ideas for the covers in
recent years.
A lot of embedded projects are for microwave ovens and the like. These
are very small systems, and typically use very slow CPUs (because they
don't need anything faster). The Military Industrial Complex is solving
a very different set of problems.
If you look through magazines like Embedded Systems
Programming, you'll get a far different picture. For instance, the bulk
of the VMEbus SBCs (single-board computers) sold are made by Motorola and
use the PowerPC processor. The runner-up is Intel processors, and DOS
isn't dead.
Been quite a while since I've looked at those magazines.
OK. Some libraries have them.
The Intel architecture is actually older than the PowerPC architecture,
by many years, so by longevity alone, Intel is antiquated. So what
exactly do you mean by "antiquated"?
Antiquated in large part means weighed down by "compatibility barnacles"
which limit the ability to adopt significant architectural changes. This
problem has affected both the Intel x86 and the IBM PPC lines.
Yes, the Intel is very much encrusted by backward compatibility. The
PPC is not yet encrusted, but give them time.
With the hardware abstraction trends the backward compatibility
barnacles should slowly evaporate across all the processor lines.
Um. Hardware abstraction layers are another form of barnacle, and kill
performance. We live with the performance cost for practical reasons,
but there is nonetheless a cost. Look into the history of microkernels
in operating-system design.
In the near future you will simply select a generic hardware platform
from the vendor of your choice and in the size / expandability /
fault
tolerance for your application, and then select your favorite OS to
run
on it from a field of dozens of variants that all run on the same
hardware platform.
For MacOS, it won't happen soon, as Apple makes far too much money on
hardware. Probably one will be able to run Windows on Mac intel
hardware, but will not be able to run MacOS on generic intel PCs.
I predict that MacOS will be available to run on generic PC hardware
within another 2 or 3 years. One of Apple's big problems is that the have
to make large profits on the Mac hardware since they sell so little of
it compared to the PC world. This causes them to either have to price
the product too high relative to the competition and try to hype reasons
it's worth the extra money, or to try to compromise to cut manufacturing
cost and risk reliability problems. We've seen examples of both paths
from Apple.
It won't happen anytime soon. This has been suggested for years, and
Steve Jobs (a founder and the current CEO of Apple) always says that
allowing MacOS to run on generic PC hardware would put Apple out of
business. I see no reason not to take him at his word.
Given Apple's various product duds and reversals of concepts like open
architecture to closed architecture and back to semi-open architecture,
I see no reason they won't eventually decide to exit the hardware arena.
You mean like the iPod?
Anyway, let's wait and see.
Mac hardware is far less trouble to assemble and configure,
That's because it is largely non-assembleable and non-configurable. You
get saddled with a generic box, you have few choices for options and you
have to pay for included items you may never use.
Macs are about as configurable as Dell PCs, right down to configuring
and ordering from the Apple website. If you like, go to
http://www.apple.com/ and click on the Store tab. You can walk
through the entire chose and configure and price process without having
to register or provide a credit card. (What's in the stores is a
fraction of the configurations available from Apple.)
And exactly how much of that is *user* configurable?
I don't get your point. The Apple store cannot tell if you are a
unwashed user or an administrator, so long as your money spends good.
All products are packaged in some manner, so you always get more than
you absolutely wanted or needed. I guess I don't see your point.
In the PC world (regardless of OS), I have far greater flexibility to
configure a hardware platform to my exact needs.
If this is simply another way to observe that the Windows ecosystem is a
factor larger than the Mac ecosystem, I agree. But the question was
adequacy for a purpose, not ecosystem size per se.
and is far
more reliable than most PCs
I've not seen any hard data showing any greater hardware reliability for
a Mac vs. PC. All computer hardware these days is far more reliable than
any of the software that runs on it.
Look into Consumer Reports and also the PC (not Mac) magazines. The Mac
magazines also say this, but what else would they say - they must be
True Believers.
As noted, Consumer Reports is *not* a credible source for anything. Also
as noted, the magazine comparisons tend to leave out entire lines of PC
hardware making them inaccurate. And again, all hardware is pretty damn
reliable these days so a quality PC and a quality Mac should have little
difference in hardware reliability.
See above. Better yet, get the December 2005 issue and look, so we can
debate from the same page.
, largely because in Macs there is a single
design agent, Apple, ensuring that it all fits together and meets
minimum standards of design and implementation quality.
... and incompatibility with the rest of the computing world.
If that's another way of saying that Macs are not Windows, OK. But both
are able to perform the same tasks, just like there are many brands of
truck, but they all use the public roads.
Yes and no, Macs could do the same tasks as PCs, but in some cases they
lack the available options (both hardware and software) to do so. It
also too Apple quite a while to join that "public road" and abandon
their proprietary networks and busses (SCSI being the only notable
exception).
And Firewire. And ethernet.
Standards, quality and compatibility were issues in the PC world more
than a decade ago. These days quality and interoperability are quite
high. Only on the most complex systems do you run into any configuration
issues and that is infrequent and in areas where Macs simply aren't
applicable anyway.
Well, the PCs have gotten far better it's true, but the Macs were always
there.
Um, there have been plenty of problems with Macs along the way as well.
In absolute terms perhaps, but Macs are far less trouble in total.
And Windows interoperates well only with Windows. There have been a
number of court cases on this issue, and Microsoft is slowly yielding
ground.
That's not interoperability, that's openness to third party software.
Interoperability is working with established standards, something that
Macs have been loathe to do.
Um. Have you been following Microsoft's tangles with the EU antitrust
regulators? They are proposing fines of a million dollars a day.
This is a major reason that people have been willing to pay somewhat
more for Apple hardware. It's simply less trouble.
That's the myth, not the reality. These days very few problems on either
platform are a result of hardware problems. Come to think of it, my Mac
friend did have a 17" Powerbook replaced under warranty when it failed
after about 3 months use. I don't have details on what actually failed,
but I know the machine was not physically abused.
Even the best of laptops have about a 10% failure rate, according to
Consumer Reports, so one can always find someone with a dead laptop.
I think Apple is the only one who had melting laptops though.
True, but lots of PC laptops get too hot to have on one's lap. In any
event, Consumer Reports shows that Apple desktops are by far more
reliable than any PC, but that Apple laptops are in the middle of the PC
range (which isn't that wide).
My kid sister just switched from Windows to MacOS. She is a Graphic
Artist, and that field is dominated by Macs, but her first husband was a
self-described DOS Bigot. Anyway, she got the big FedX box just after
Christmas, and called to tell me how easy it was to get set up and
running. (Her current husband is not a computer guy at all.) It took
all of an hour.
Macs have been loosing some ground in parts of the graphic world, due to
a number of factors. Since most of the same software is available for
Windoze and works just as well there, and the fact that Windoze is still
a business essential and less expensive per-seat than Macs has led some
companies to ditch Macs in favor of PCs / Windoze (CCI did this).
Macs are something like 60-70% of the Graphics Arts world.
At a previous job we had one advertising / PR / graphics person who used
a Mac. It was a pain for us to support since it was the only Mac in an
office of several hundred PCs. We eventually got a decent PC, loaded it
with all the same applications that she used on the Mac and put it on
her desk next to the Mac with the hopes that she would give it a try and
eventually switch to it.
The end result is that after a few months with no extra prodding or
support other than instruction on how to access the same work files from
either platform, this user switched to the PC. Obviously she knew that's
what we were hoping for, but she did indicate that she found that some
tasks were easier on the PC and none were more difficult.
The social pressure must have been immense.
I was the sole holdout at work for many years, long enough to have
managed to miss all the Windows 3.x and NT dramas my coworkers told me
about over lunch. It was a lot of fun, at least for me. After a while,
someone would notice that I had been uncommonly quiet, and would ask me
a question. I would reply with some variant of "I have a Mac, and it
just works, so I have nothing to report". Getting Windows to support
CD/ROM drives was a leading cause of wasted weekends.
Yep. I'll probably get one of those $700 Dell boxes. Already got the
hardware firewall.
An old Optiplex GX100 (P3/733) runs Mach3 just fine under W2K on a
machine that will do 60IPM or so. Nice and cheap used as well.
Yes, but my heart is set on a Dell. For one thing, I want one company
to yell at.
The old Optiplex GX100 *is* a Dell. Just one I got used from a corporate
surplus source. Since most large companies cycle PCs out every three
years to coincide with the 3 yr warranty that is standard on most
manufacturers "business" lines, there is a steady stream of good cheap
PCs that are plenty capable for all but the most intensive applications.
OK.
It will be interesting to see what happens in the market when Macs can
run all these Windows-only apps at full speed, so there is some real
competition between platforms.
I don't think that will cause any real competition. What it will mostly
do is remove a handicap from those who prefer the Mac UI. I don't think
there are any significant numbers of people wanting to migrate to a Mac
but being held back by a lack of apps. Those wishing to migrate away
from Windows are more likely to explore free options like Linux that
will run on their existing hardware.
It will certainly remove that handicap. But it will also expose lots of
people to the Mac, and comparisons will be made.
Comparisons will be interesting as I see the Mac in its current OSX
form as nothing more than another UI shell available for a standard Unix
base.
The standard retort is that by the same token there is no difference
between a Porche and a Chevy; they are both cars.
Microsoft itself does not agree that lack of applications is what
prevents migration away from Windows. This came out in spades in the
antitrust case, where they were caught doing all manner of illegal
things to preserve this barrier. It's all in the opinion handed down by
the Federal Appeals Court.
Any successful, dominant company is going to be attacked and for some of
the most bogus reasons. Yes Microsquish has done a few things that were
wrong, but they have also been bashed for doing things that they have
every right to do as far as I'm concerned.
More than a few things. It's all there in the court rulings.
How can you possibly justify forcing Microsquish to include competitors
products with their distributions? Is GM required to include Ford
products with the cars they sell just because some users may prefer to
put a Ford dashboard in their GM car? That's about on par with some of
the stuff pushed on Microsquish.
Any company that achieves ~90% market share in an important industry
will find its freedom of action curtailed. The classic example is AT&T,
which achieved a similar market share by methods that are now illegal,
but were common back then. The solution was to turn telephones into a
regulated utility. This is probably one of Microsoft's biggest fears,
but their current conduct makes such an outcome more and more likely.
Datapoint: At its peak, IBM had only a 70% market share.
Anyway, when barriers are removed, migration happens. Some will go to
Linux (if they like that unpolished an environment), and some will go to
MacOS (polished exterior, real UNIX available below).
Yep, Apple might reach 8% market share while Windows drops to 67% and
Linux/Unix rises to 25%.
Truth is, Apple (and Mac users) would be perfectly happy with ~10%
market share.
I think we are mixing unlike things here. The desire for independence
and freedom from lock-in exists regardless of the skill of the
programmer, especially as the programmer becomes experienced (and has
been screwed when something he depended upon is made unavailable).
Freedom from lock-in and abuse by marketing-driven companies is its own
good.
The pseudo-programmers I reference are not concerned with such things,
they exist to glue purchased MS code libraries into horrendous "business
apps" for just long enough for them to migrate into the "management"
world.
OK.
I've seen the headaches the poor "real" programmers who come along later
have trying to fix and maintain these horrible apps. It's not pretty and
makes me happy I'm on the systems end of things.
True enough, but it will never change.
Joe Gwinn