I've received quite a few thoughtful and interesting responses to my current Moore's Views & Reviews column: "Do Mac Web Journalists Have A Professional Obligation To Use The OS X Beta?" The blockquotes in colored text below are citations from my column for context.

From Daniel Knight:

Why I don't use OS X Beta.

First, I don't use OS X Beta because my hardware doesn't support it. A lot of readers are in the same situation, either because they have clones or pre-G3 Macs.

Second, I paid $29 to find out that OS X Beta wouldn't run on my SuperMac S900.

Third, I didn't dare install it on the internal drive of my computer at work. I have to be productive there. If OS X causes problems, it would be a disaster.

Fourth, I finally got my hands on a big external hard drive, installed Mac OS 9, updated that to 9.0.4, ran Software Update to bring that to the latest level, and installed OS X Beta on my G/400 at work. It's a very cool looking OS, but certainly doesn't have enough going for it to make me want to switch -- maybe when it goes beyond Beta.

Fifth, I played with OS X for maybe an hour a day for a week. I downloaded Fire so I could chat with friends on AOL Instant Messenger. I surfed with IE5. It never crashed. It looks gorgeous. But then one day it refused my password. Again and again I tried, using the same password and ID that had worked when I logged in maybe 20 minutes earlier. In the end, OS X refused to let me switch back to OS 9 on my G4.

So I shut down the computer (why no Restart in OS X?), turned off the external drive, and restarted the G4. Then I powered up the external drive and erased it. No more OS X for now.

OS X is promising. I owe it to myself to try it. But I owe it to my readers to use my Mac as a tool first, a test bed second. If a piece of software causes problems, you get rid of it. Sadly, that's what I did with OS X Beta. (Yes, I did send a "bug report" to Apple.)

Dan Knight, publisher
Low End Mac
http://lowendmac.com/index.shtml

From M:

Really great, summarizes many views, pro and con, in a very fair manner. As one whose posted the sort of entirely biased and one sided criticisms harping on Aqua's flaws that it is currently popular to dismiss as 'fear of change' etc, I really appreciated that.

Apparently many people are under the impression that 'beta' means 'immune to criticism.' And wanting Aqua to be better than it is and pointing out it's shortcomings, is considered anti-Apple propaganda and doomsaying.

What actually pisses me off is that most of the discussion of the public beta I've seen is so worthless. Aside from the one sentence love it or hate its, there's mostly the people who whine because the beta's not feature complete, the ones who think beta status excuses anything, and the ones arguing over whether the MacOS or OSX way of doing something is better as if those were the only options. Plus the ones who think because you'll be able to get feature A, or some simulation there of, by changing this preference, using that command-line hack, or installing some third party add-on that may come out there's no reason to complain. I don't think it should require that much work to get to what I consider a minimum level of usability and if it does then what did I pay Apple to do? I've also noticed that non-Mac reviews of Aqua tend to be more glowing, ironic since I think Aqua seems to be becoming the dumbed down, overly simplistic, and limiting GUI with lots of pretty pictures that some of them always accused the Mac of being. :)

OSX has a lot going for it and Aqua has its good points too, but -IMNHO- currently overall Aqua is not nearly as powerful or as easy to use as the Mac. There's no reason it couldn't or shouldn't be and as John Siracusa at Ars Technica pointed out many of its mistakes are things where a solution is obvious and/or already part of the current Mac interface.

-M

From Bill Chin:

I have a few comments on your recent column entitled, "Do Mac Web Journalists Have A Professional Obligation To Use The OS X Beta?"

Then there's the issue of fixing OS X when problems inevitably crop up.

There are several aspects of the above statement that I would like to address. However, a complete treatise would be enormous. I'd just like to point out a few things.

1) The lack of protected memory means that it is more common for files to get corrupted in Classic Mac OS, thus requiring the need for extensive troubleshooting.

2) The archaic patching system that goes on for extensions and the like requires more extensive troubleshooting after installing new software or hardware.

By eliminating the two above, the chances that your machine won't boot properly drops dramatically. Would you sacrifice the familiarity of the current system if you rarely had to troubleshoot the MacOS? I've been the administrator for many machines based on Mac OS X's predecessor, NEXTSTEP/OPENSTEP, and we rarely had a system troubleshooting issue. We had apps that crashed and we've had network issues, but the need to drop to a command line and mess with the system was very rare. Matter of fact, it was more common to apply operating system upgrades every 9-12 months than troubleshoot the client machines.

There are scenarios where the system will fail. Malfunctioning hardware or buggy kernel level software (which includes 3rd party device drivers) would be the prime culprit. Apple has to make sure there is a good recovery scenario for NetInfo issues. While you and many Mac gurus find troubleshooting Macs easy, or at least easier than various Windows or UNIX flavors, troubleshooting the Mac is still daunting for beginners. It is also frustrating for experienced system administrators, as the Classic MacOS lacks many basic troubleshooting tools (ie. a central system logging facility). Apple does need to address the recovery scenario, but being based on UNIX does not necessarily mean that the system is inherently more difficult to troubleshoot. The difference is that in the Classic MacOS, one usually troubleshoots with a blunt instrument. You nuke the preferences, boot off other media, re-install software, enable/disable extensions, etc. You could do that with Mac OS X also. Or, if you know the UNIX command line, you can attempt a finer grained approach, like editing the preference file(s). How often did you troubleshoot your Mac using Macsbug, ResEdit, or other developer tools? You're certainly not required to do so and you could stay out of the command line and still use the blunt approach. The complexity is there, you just typically don't have any access to it. With the CLI, you have an intermediate access underneath the facade, but you don't need to use it.

if I am going to be obliged to learn UNIX anyway to support OS X, and since I'm not exactly enthralled with the Aqua interface, would another UNIX, perhaps a Linux running Eazel's Nautilus GUI, be worth a serious look?

I would encourage any pontificator of computer knowledge to attempt to be as open minded as possible and take the time to look at the diversity that is out there. To not do so is a disservice to the public, just as a general journalist should take a very good dose of world history. Go. Take a look at SuSE Linux or LinuxPPC and the Eazel project. It doesn't cost much to do so. I bet you'll be back. :-)

One significant advantage SuSE Linux for Power PC has over OS X is that it supports legacy Macs and some Mac clones back to 601-based PCI machines, including my 200 MHz, 604e, UMAX SuperMac S-900. It kind of begs the question as to why, if Unix-based Linux can support legacy Macs, why Apple's own OS X does not. I suspect that Aqua is probably big part of the reason, which, if so, reinforces my contention that abandoning the Classic Mac OS interface completely was an ill-conceived decision.

While Aqua is part of the reason, your conclusion, in my opinion, is incorrect. First, Apple has limited engineering resources and have chosen, as a matter of expediency, to limit support. They chose as the cut-off the original iMac class machine (which includes older beige G3s). Apple is not expending the energy at supported clones for which they don't have the specifications or machines. As a result, they don't test their system software with older machines and sometimes they break things. This is evident in the Public Beta, as it doesn't run out of the box on some 604 class machines that could run the earlier Developer Preview 4. From the Darwin Developer's mailing list, it appears that recent changes to add power management didn't work right on 604 processors. A later Darwin kernel fixed that, and there are alternative kernels available from various sources on the 'net that restore compatibility. The reason why SuSE Linux supports so many more machines is because someone out there found the problems and fixed them. Since the CoreOS of Mac OS X is available through the Darwin project, 3rd party developers are free to introduce support for older machines. Apple may, or may not, roll these changes back into Mac OS X at their discretion. Here is a relevant exchange on the Darwin Developer's mailing list between Ryan Rempel and Wilfredo Sanchez (Open Source Engineering Lead @ Apple)

=======

Ryan Rempel: When people make changes to Darwin to provide support for the 7300 - 9600 series, will Apple commit those changes to the CVS repository (supposing that the changes don't break anything for others). If the answer to this question is "no", then I think that we need to start thinking about a way for those interested to set up an alternate Darwin distribution, which would integrate the various patches that have been made (and will be made) to support the 7300 - 9600. If the answer to this question is "yes," then we need to make sure that the patches we want are formally submitted in a way that Apple can evaluate.

Fair questions.

If you were to get these systems working again (assuming they are indeed busted), then we shouldn't have any problems integrating good patches to that end. It is possible, however, that we may choose to make support for certain platforms optional features, which can be enabled/disabled in the build configuration and that we might opt to disable older platforms when we build the kernel in Mac OS X (perhaps to save on wired memory, whatever, I don't know). That brings up an interesting question as to whether we should then enable them in a "standard" Darwin build, and that would depend on whether people want the standard builds to relate closely to the Mac OS X builds, or whether they should include the support and therefore diverge a bit. I think we can figure that out when we get there.

Anyway, I would think that if the code could be cleaned up to support older hardware we should be amenable to accepting that work into the code base.

======

Apple has two competing demands: completing Mac OS X, including new features vs. additional model support. Since each Macintosh model has its own quirks which require specific system software support, each model or at least class of machines requires modifications to Mac OS X. Also, Apple has truly made some horrible machines in the past which would function fine without a GUI or with a spartan GUI, but a state of the art vector based system is out of the question. Further, Linux for the PowerPC has been around for much longer and their support for newer Macs is poor compared to Mac OS X. Would it be more acceptable for Mac OS X to not support the dual processor G4, iBook, or a Umax S-900?

As for abandoning the Classic Mac OS interface, I don't believe Apple has done so. There is still a menu bar that goes across the top of the screen. There is still an equivalent to the Finder. As compared to how much could have changed, very little has changed. Scroll bars are still on the right (but they should be on the left), it is still click to focus, Get Info still exists, as do resource forks. I could go on. The Classic Mac OS interface was designed for a single tasking, single user, small (512x384) display and definitely not networked. Over time, these things have been bolted on, and while many users are currently comfortable with the frankenstein-like result, it is time for a change to lay the groundwork for the next decade. Mac OS X is a statement of Apple's direction, and not changing the UI would be a big mistake. As it is, I believe Apple changed too little for the sake of providing a gentler migration away from the Classic UI.

Bill Chin
M Dimension Technology

From John Martellaro:

Recently, I spent an entire week using Mac OS X at a computer show. Each day, I learned more and became more proficient. There was nothing I couldn't do - no road blocks or limitations. Now, when I use Mac OS 9, it seems old fashioned and awkward. --

John Martellaro

From Joe Walters:

Sir:

Read your article concerning your fear/reluctance to embrace OS X because of its "Command line interface."

As a retired Bell Labs geek I admit my bias, however I"m sure that it will take only a short time before you wonder how you ever got along without UNIX. The commands are terse, however, there are "rules" as to their name and once you understand the rules, the names drip off your tongue in a never ending flow i.e., grep did not come from an explosion in a font factory :-) It means: Global, Regular Expression, Print.

A regular expression is perhaps the most slippery thing to nail down because it kept changing as UNIX evolved. Some commands have a much more restricted view of just what is a RE than others.

A RE is a strong that is used to match text in a file for a match.

Kind Regards,
Joe Walters

From David:

The short answer is: 'No, just don't write about OS X if you don't use it.' The longer answer is: 'Where would we all be if no reputable Mac journalists used it?' And frankly, I enjoy reading what you say and would like to compare your experiences to mine.

Actually, I am amazed by some of the incredibly silly things I've read. In fact, since the release of the public beta, I've seen many articles online that remind me of the dreck you see on usenet news. But what really amazed me were two articles quite critical of OS X by people who admitted they had never used any build of OS X.

Personally, I am. (at least) in the cautious camp. I am extremely fond of the Classic Mac OS, which I have been using since System 6.0.1. The Mac Graphical User Interface is the overwhelmingly dominant reason why I choose to use Apple computers, and why I am sitting here right now writing this column. I know the Classic OS; I know what I can and cannot expect of it; and I have tailored my work habits over the past eight years around it. The interface suits my needs and tastes to a T. And almost equally important, I know how to fix it (most of the time) on the relatively rare occasions when the Mac OS (or more likely some piece of third-party software running on it) fouls up.

Darn good reasons to continue working with Mac Classic. I was lucky that the public beta was delivered right in the midst of a calm spot in my life so I installed it on an external hard drive and used it for a little over a month. Eventually I played and hacked it into oblivion and I removed it. That month was extremely exciting - and frustrating - and what I learned is this: I'll probably buy the release version and treat it exactly as I did the beta. Then as more software becomes available for it, I'll begin my serious transition to OS X. That is how I moved from OS 6 to OS 7. It took me about a year to move entirely to OS 7 and take 6 off my hard drive.

Having said that, I am also not a complete stick-in-the-mud, and I acknowledge that the Mac OS has some shortcomings, although I would argue that the latter are almost all in the underlying, and now antiquated, guts of system, and not in the GUI, which I consider the best of any personal computer user interfaces I have yet encountered.

Quite true. And yet, there are things that I now remember didn't make much sense and only slowly got used to. For example: Take a floppy or zip drive and mount it. Then drag a file or two from it onto the desktop. Now unmount the disk. I remember years ago being confused as hell about where my files went. Hadn't I copied them to the desktop? Now I accept that there is one desktop for each logical drive mounted, as well as one trash can for each logical drive. That is one issue that OS X is trying to deal with in a more intelligent fashion. You and I are comfortable with the older metaphor but if there is a better way, shouldn't Apple be encouraged to find it and shouldn't we embrace it?

Ergo, I am not intransigently resistant to the concept of switching to a different OS, but I need to be convinced that there is compelling reason to do so. My exposure to OS X, admittedly limited so far, has not convinced me in the slightest that it will make my work day more efficient, more convenient, or more pleasurable. The contrary in fact. Several of the features that I like best about the Classic Mac OS are missing from the OS X Aqua interface, and while Aqua is not utterly horrible, I have serious problems rationalizing the idea that switching to a tool with fewer features and more cumbersome operation than the one I am now using makes logical sense.

In my first few hours with OS X I felt as if Steve had stolen my Mac. Then I began to get used to it. Now, sitting here with OS 9, I miss some things - some of which I hated at first. Okay, I want all my drives and my trash can on the desktop. I was able to accomplish that. I hate the 'My Computer' metaphor. However, I love the new finder windows and their toolbar. Really, as much as I hated it before, I like it now. The inspector needs work. I despise the dock but I know DragThing will be ported over so I needn't worry about it. I very much want tabbed windows and system wide contextual menus and clippings.

So, I hear what you are saying, and yet...Our current OS has accreted over 15 years and I think Apple was very smart to NOT carry everything over. First, let's dump the stuff that can be done better or we don't need any longer. Second, if there are features that few people use, let them be handled by third party developers. Spend time and money on the important things. Third, not all the features are good no matter how used to them we might be. As wonderful as the Apple Menu has been for the last 9 years, I haven't been using it for nearly 5. Perhaps it is time to give it a rest.

Of course there is OS X's more robust, Unix-based core, which does offer significant improvements over the old, ancient, Mac OS , notably protected memory and preemptive multitasking -- the former almost in and of itself enough to convince me that putting up with Aqua and the Dock is worth the trouble -- IF -- and it's a big if, protected memory really will spell an end to crashes. My son's Lombard PowerBook seems to lock up as frequently or more often running OS X Public Beta then it does on OS 9. Real multitasking would be nice to, but for me it's not a killer feature

I only locked up twice and both involved Mail. I somehow managed to frag my system twice with Apple's Mail program. Once I stopped using that program I was stable as could be and when programs did crash I could re-run them. At one point I had my desktop running 24 hours straight for 13 days. That darn unlucky thirteenth day! Not even my trusty iBook goes more than a week without a restart.

Then there's the issue of fixing OS X when problems inevitably crop up. As I noted above, I can troubleshoot and repair classic Mac OS problems pretty well, but I am a command line ignoramus. I never owned a DOS or a pre-Mac Apple Computer, and programming beyond level of HyperCard, or perhaps AppleScript, is almost as exotic to me as nuclear physics. There are of course those who argue that the average end-user will never have to know that UNIX code is lurking there beneath the flashy Aqua interface, and while that is probably true for the folks who would seek tech support for even the Classic Mac OS, it is not true for me, living 150 mi. from the nearest Apple dealer, and 50 mi. from any sort of computer tech support facility.

If Apple has done its job right, dealing with OS X problems shouldn't be much of an issue. I have a Linux box that I use for playing and programming and running my home network and since I installed LinuxPPC on it nearly two years ago, I haven't done anything to it. Granted, it doesn't get the workout that my desktop does, but the way Linux works I don't expect to be troubleshooting it all that much.

One of the big advantages of a Classic Mac OS for me is that I COULD fix it myself. Now, while I'm not the village idiot, and I suspect that I can, if I have to, over time, learn enough UNIX expertise to deal with OS X problems, which will inevitably crop up, and while learning some UNIX might not be a bad thing, I chafe more than little at having it thrust upon me.

And isn't it nice to know that the Mac community has so many people willing to share expertise? Ask a problem involving NT or Windows and you are likely to be bombarded by idiots crowing over your ignorance and misfortune. I seldom see that from Mac users. In fact, about two years ago I asked about a problem I was having with AppleShare on usenet news and got into an email conversation with a fellow who eventually drove 90 minutes to come help me.

I think that there are an awful lot of Mac users out there who feel more less the same way, and I naturally feel comfortable advocating their concerns. There are plenty of voices on the Mac Web cheerleading (I think with genuine enthusiasm) the advent of OS X. I also think there is room for some voices speaking for those who intend to stick with the Classic OS that they know, and that is doing a good job for them, for at least a while yet.

However, this raises some subsidiary issues. When Windows 95 arrived, I spoke with quite a few PC people who vowed that they were going to stick with DOS forever, regardless of the newfangled OS with all the pretty pictures. I think that some people really did like DOS, had developed their work habits around its idiosyncrasies, and found that it did a good job for them. Nevertheless, the notion of die-hard DOS fanaticism now seems quaint, notwithstanding the fact that command lines are actually making a strong comeback with Linux, and probably there are a few individuals out there still using DOS on their old 286s and 386s.

You raise a few good points. Windows 3 was a major triumph for Microsoft and never repeated. Windows 95 and Windows 98 were utter failures as upgrade software. The Win95 platform grew as a result of new computer sales, not upgrades and the same was true for Win98 and I expect WinMe too. And from my early Mac consulting days, I saw the same thing with OS 6 to 7. 8 and 9 seem to have been more successful. I wouldn't be surprised if OS X grows more from new computer sales than upgrades during the first 18 months. Then as the new OS X applications start being delivered I expect more upgrades.

The Classic Mac OS, IMHO, is a far superior operating system to DOS, and will continue to do a fine job for many years for users who are inclined to be content with it. However, I suspect that in 2005, stubborn refusal to move beyond Mac OS 9.1 or 9.5 or whatever the final version of Mac OS Classic is called, will seem as mildly eccentric as using a DOS on an old PC does in 2000.

You might be surprised to discover how much DOS is still being used. For many companies, the cost of moving their databases from DOS applications to Windows is so prohibitive, or it would disrupt business to such a degree that they are still chugging along.

For better or worse, Captain Jobs has set the course for the good ship Apple's future on the compass of OS X, and if we're going to remain Macophiles, the reality is that we're eventually going to have to adapt to OS X, and whatever OS numbers follow it, whether we like it or not.

Forget for the moment that Linux is a bear to install (but getting better with each new release) with a few notable exceptions, the quality of Linux software doesn't match that of the Mac. Nor does the quantity. Actually, quality isn't the right word. Linux programs are of high quality but they lack polish. And the same is true of all the various GUI interfaces. As yet no one has stepped up to the plate to create a user interface guideline that everyone in the linux/unix world can agree too. It is too geeky and there is too much independence. This is both Linux's strength and its weakness.

This would not even necessarily involve abandoning the Mac platform. For example, SuSE Linux has just released their version 7 OS for Power PC, which the company claims is user-friendly enough for Linux newbies to install.

SuSE is also working closely with the Eazel project, which, incidentally, was conceived and is spearheaded by a blue ribbon group of veteran ex-Apple programmers including Andy Hertzfeld, who wrote the code for the original Mac OS toolbox. Eazel is developing a user-friendly GUI for Linux.

I'm looking forward to seeing this, but having a friendly GUI is only part of the problem. The Windows GUI isn't all that bad, but the software developers don't follow the Microsoft guidelines, including MS itself!

One significant advantage SuSE Linux for Power PC has over OS X is that it supports legacy Macs and some Mac clones back to 601-based PCI machines, including my 200 MHz, 604e, UMAX SuperMac S-900. It kind of begs the question as to why, if Unix-based Linux can support legacy Macs, why Apple's own OS X does not. I suspect that Aqua is probably big part of the reason, which, if so, reinforces my contention that abandoning the Classic Mac OS interface completely was an ill-conceived decision.

There is support and support. There is a good deal of difference between SuSE saying that their version of Linux should run on a 7200 and Apple saying it. Apple is saying that the 7200 user will have the same Mac experience as a G4 user, not in terms of speed but in terms of plug and play etc. We've already seen that OS X works on lots of unsupported computers but for Apple to certify that will mean something entirely different.

While these musings may seem like heresy to some, if I'm thinking them, I don't doubt that many others are as well , particularly people annoyed at having their high-end, less than three-year-old, Macs orphaned by the new Mac OS.

Of course, the Mac experience is more than the Mac OS and Apple hardware. It is partly a state of mind and a community of people who share that state of mind. It includes the astonishingly rich selection of really great shareware applications and utilities for the Mac that I document every day in the Applelinks Shareware Beat feature. Happily, it seems that Mac shareware developers are rising to the challenge of OS X, with an avalanche of innovative new applications, some of which even replace those missing features in the Aqua GUI.

What I've experienced thus far leads me to believe that Apple will pull it off. Come on in and join us, the water is fine.

cheers

David

Charles W. Moore

.