I suppose that the argument over whether Leopard is a major disappointment, a promising work-in-progress, or the ultimate operating system, won’t be settled right away. While our ongoing poll showed that about half of you have had excellent, trouble-free experiences, there are far too many question marks at this early stage.

On the other hand, I expect that, if Apple’s past history is the guide (and it should be), most of the lingering problems will be resolved in near-term maintenance updates.

Yes, I agree with many of you that Leopard left the starting gate much too early. This is an unfortunate part of the software business, where marketing considerations may trump common sense and the presence of annoying bugs in the new product. Here, for example, Apple had already delayed Leopard once, and the coders were under great pressure to get their work done in time for an October release. Imagine how the tech press would slam Apple if it announced yet another delay? Imagine the impact to their stock price?

Well, they made their deadline, of course. Some of you may wish they didn’t.

But if you go through the history of the Mac OS, right from the beginning, you will see system software that’s required updates from time to time, particularly in the weeks following a major upgrade.

I remember, for example, when the first Power Macs with PCI peripheral ports came out. I won’t get into the thoroughly evil RAM installation routine for the 8500 and 9500, which made the Mac mini look simple in comparison. Instead, I remember my accursed 9500 crashed so often I hardly got any work done for a few days. The original system version, a 7.5 variant, was soon updated and the worst of the ills were vanquished, more or less.

When you go through Mac OS X’s history, you’ll find a similar situation. A new release appears with great fanfare, followed just weeks later by the inevitable bug fix to repair the serious stuff, and then, over time, more updates to fix up most of the remaining defects.

How many of you recall installing Mac OS 10.3 Panther, only to discover that some FireWire 800-based drives stopped working. This particular problem could be blamed on Apple and the need for some drive makers to deliver firmware updates for certain FireWire controllers. Indeed, this particular problem was far more severe than most anything that ailed Leopard (except for the notorious Command-drag file move bug), because it would destroy or damage a drive’s partition map, which meant a trip to the drive recovery center to recover your data.

It actually infected me, and it happened at an awfully unfortunate time, because I had just sold one of my older Macs and I was busy setting it up for its new owner. However, the affected FireWire drive just had some old backup files that I didn’t need.

Yes, I installed the firmware update, and the inevitable 10.3 fixer-upper from Apple, but I learned my lesson all-too-well. From here on, all my desktop Macs would be purchased with — or retrofitted with — a second internal drive where possible. Two backups are better than one. Today, I have Time Machine handling my external FireWire drive, and SuperDuper! (only partly functional in Leopard) doing its thing for the second internal drive. By the way, the SuperDuper! issue relates to making a proper bootable clone, but the data is otherwise intact, and the Leopard-compatible version should be out soon.

Now do you think Tiger was the greatest version of Mac OS X ever? That’s the way the Leopard doomsayers want us to remember 10.4, but that may not be true. When Tiger first came out, the major changes to the underlying networking architecture made third-party VPN software — required for many businesses — non-functional or partly-functional. Don’t believe me? Well, check the history over at John Rizzo’s MacWindows cross-platform information site.

Indeed, those problems took an awfully long time to resolve. In the end, it required updates from both Apple and other companies to set things right, and some say that problems persisted until the very end. So some of you may indeed think of Panther as the best Mac OS, except for that FireWire 800 data loss bug, of course.

But even the ongoing network repairs didn’t make Tiger perfect for everyone. There are lots of troubling reports that the 10.4.10 update caused a positively huge amount of harm for some of you, if the comments posted here are any example — and I think they are. I suppose 10.4.11 helped close out Tiger with most of its reputation intact, though.

So I am not surprised to see Leopard beset by various and sundry issues, at least for some of you. What makes them seem all-the-more notable in the scheme of things is Apple’s incredible success in moving so many copies in such a short time.

Consider: If only a few percent of new Leopard users encountered problems of one sort or another, that’s a large number, and people who complain tend to be far more vocal about their difficulties than people for whom the Leopard upgrade went perfectly, or almost perfectly. And, please, don’t use our random polling about Leopard as anything more than an intellectual exercise.

That’s why, for example, troubleshooting sites tend to have a huge percentage of problem reports. This doesn’t mean they should be ignored. You just have to put the information in perspective, and make sure that the bugs posted are repeatable among a cross-section of users, for otherwise it may be just one of those things that affects but a few and that’s about the size of it.

So what are we to conclude about all this? In the scheme of things, I think Leopard is no worse than other Mac OS releases. In fact, for me and others, the experience has been far, far better.

I know that many of you are the proud owners of vintage Macs, dating back to the very first model, back in 1984. Of course, there’s not much they can do today with anything but the software and operating systems released when those computers were first built, or a few years thereafter.

In moving to Mac OS X, Apple did a lot of unthinkable things, such as embracing open source components, open standards and so on and so forth. That stands in sharp contrast to Microsoft, who considers open source a nasty insect that must be swatted out of existence. And, no, I want get into those alleged patents that Microsoft claims it holds that Unix operating systems and applications infringe on. That’s strikes me as just a power play, and rather a clumsy one.

Now, while I realize some of you will disagree with me, my daily Mac computing experience is far more reliable than it used to be. In those days, I was lucky if my Mac didn’t crash a couple of times a day. Sure, some of you never encountered those problems, but I’ve always been deeply immersed in desktop publishing, graphics and other content creation software. These are the applications that have always stretched the limits of the system and the hardware, and, as a consequence, generate the most instances of system instability. That is, aside from the overzealous installation of system add-ons.

But in moving to a world-class operating system architecture, Apple decided to ditch many of the concepts and components from the Classic Mac OS. To a large extent, that was a good thing, as the aging Mac OS of old was about to crumble in its last days. But there were also good ideas that never survived the transition, and that’s unfortunate.

In past articles, I’ve talked about the ultra-configurable Apple menu, and the need for third-party solutions to provide solace under Mac OS X. But there’s yet another important component of the Classic Mac OS that disappeared, although there is sort of a replacement, if you can call it that.

I’m talking about desk accessories, those tiny applications that provided all sorts of added functions, such as a calculator. They existed side-by-side with your regular applications, and were called upon simply by choosing them from the Apple menu.

The replacement, Dashboard, is an interesting concept, lifted from a shareware application, Konfabulator, which also used widgets to provide various degrees of functionality. But Dashboard, while workable, remains controversial, simply because it exists on its own layer, as the Finder and application windows are grayed out.

Sure, this condition can be dispensed with courtesy of  third-party software, but Apple hasn’t seen the need to change Dashboard’s basic functionality, although they made it easier for you to make widgets. Starting with Leopard, if you want to build one from scratch, there’s the Dashboard SDK, or you can make a Web clip from Safari 3.

But were desk accessories so broken that they had to be replaced with something that really isn’t quite as good or as convenient? I wonder.

There is also the venerable Scrapbook desk accessory of old, a little repository of clip art and other stuff that I’m sure many of you found awfully useful over the years. I didn’t use it all that much, but one of my long-time clients, a publisher of marketing books, came to depend upon it heavily. He was also rightly disappointed that the feature disappeared in Mac OS X and was never restored.

Indeed, there are third-party solutions that can do the job. But why not a fancy 21st century scrapbook utility direct from Apple?

Oh, and one more thing: My friend David Biedny reminds me that, years ago, there was a nifty little system extension that did nothing more than drop down the menu when you moused over a menu bar label. No clicking needed until you actually selected a command. This may not seem so big a deal until you consider how many mouse clicks you saved over a period of months and years, not to mention wear and tear on your wrist.

Is there a Mac OS X equivalent? Anyone know? And if not, why hasn’t anyone tried to realize this super-simple concept?

When I wrote a commentary about whether or not Leopard was successful, predictably the comments were all over the place. Some of you had experiences that match mine, with no problems or just issues so minor they weren’t worth serious attention. But others voiced endless frustration, going so far as to zero your Mac’s hard drive in other to attempt to make Leopard “take” without protest.

To be sure, I’m rightly concerned that the experiences appear to be all over the place. While I grant that the vast majority of Leopard users have had great experiences, there are far too many to ignore. But how do we get to the bottom of this dilemma?

Now it’s fair to say that many of the Leopard-related issues are related to conflicts with third-party software. This is something that will no doubt be resolved over time as more and more products are updated for 10.5 compatibility. Indeed, that’s common with new system releases, and it doesn’t mean Apple got Leopard out the door prematurely, although I think it’s clear they did.

Another potential source of misery is the installation process. The promise of installing with a few mouse clips, and having your Mac emerge intact with your new system ready to roll can be a unrealized dream for many of you. I’m sure a basic upgrade works fine in many cases, but all-too-often, the process fails.

So why should this be? Is there something wrong with Apple’s upgrade technique that yields inconsistent results?

Not necessarily, because the installer script has to anticipate a huge number of possible system configurations while at the same time placing (or replacing) hundreds of thousands of files on your Mac. This is nothing to take lightly, even though it works smoothly most of the time.

In cases where you’ve directly modified the system with some enhancement — maybe something you installed years ago and just forgot about — the files the installer expects to be there have been changed or, worse, removed. Now, in theory, the installer ought to stop in its tracks in such a situation with a warning message, but if it’s just an added kernel extension that’s wreaking havoc, that’s something likely to be overlooked.

Of course, Apple’s solution in such situations is to use Archive & Install, which has two options, one of which is to preserve your Users folder and settings, so there will be minimal work on your part after Leopard is installed. The most draconian step is erase the drive as part of the setup process, and that’s what I did on one of my Macs, only because it had undergone constant use and abuse for nearly 18 months, and it was a good time to start over.

Yet despite all these methods and considerations, there are still times where the Leopard installation continues to fail in some fashion. So what has gone wrong? Well, assuming there’s no drive directory damage, and certainly you should check it before proceeding, where do you go next?

Well, there’s one other Mac troubleshooting issue that is not addressed as often as it should be, and that’s the state of your RAM. Now this is a factor easily overlooked. It’s possible you bought your Mac with the memory already installed, and never paid attention to it before. Or maybe you did add memory, but things worked just fine up till now.

So why should this be a matter of concern?

As you might remember when Mac OS X Panther first came out (that’s 10.3 in case you forgot), many of you discovered the system was far more sensitive to RAM issues that didn’t affect you previously. Certainly, if data is corrupted in memory, it’s as bad as being corrupted on a drive. When your system loads into RAM, particularly if it’s defective in some fashion, all bets are off.

So how do you know it’s really the memory that’s responsible? Well, yes, you can say everything worked all right until now, but have you occasionally had unexpected system crashes that are inconsistent and don’t seem to center on a specific program? That may also indicate a RAM-related issue.

All recent Macs come with a Mac Hardware Test included as a separate disc or integrated with the one that contains your original operating system. Regardless, check your Mac’s documentation on how to use it, for it will test your Mac’s hardware with a reasonable degree of thoroughness. You can also do RAM tests with MicroMat’s TechTool Pro.

Either way, you should use the extended RAM tests to exercise more of your system’s memory in different ways and see if any errors are reported.

If the problems still occur, and the hardware passes all tests you give it, try, if possible, to remove extra RAM and see if the issues remain. If you have a Mac mini, where removal of RAM is a hideous process not recommended for regular people, or just one RAM stick, hand it over to your dealer to see if they can determine what might be wrong.

No, I’m not an apologist for Apple. But I know of situations where bad memory — or some other hardware component — did all sorts of screwy things, and this is definitely something you should consider if your Leopard installation went badly and no other solution seems to work.

After long months of anticipation, Leopard went on sale in late October, and I’m sure many of you succumbed to the lust for eye-candy and the promise of over 300 nifty new features and placed your orders early on. I know I did.

Others waited for hours for the big bash at an Apple Store, as if some famous entertainer was going to be present. All this for a personal computer operating system? Well, in 1995, Microsoft got a similar reception to the introduction of Windows 95. How times have changed!

But has Leopard truly realized its potential? Did Apple deliver on its promises, or deliver a buggy product that a few tech pundits are even comparing to the dismal Windows Vista rollout?

The problem here is that there’s a huge disconnect between reality and some of those published claims. At the heart of this is the fact that, like all new operating system releases, Leopard was plagued with some annoying bugs when it left the starting gate. True to form, Apple probably released the product a little early in order to meet the deadline of shipping by the end of October.

But most of the serious problems were fixed within weeks with a fast 10.5.1 update, yet the perception some so-called journalists want to convey is that those fixes never existed, or just weren’t very important.

Among the complaints that were supposedly so serious: Failing Wi-Fi connections, application crashes and, of course, that infamous bug where you could lose a file while moving it to another drive or partition using the seldom-used Command-drag process.

But, as I said, that’s a moot point, although I grant some of you are still having problems that, one hopes, will soon be solved by Apple or a third-party developer in the very near future.

What is a fact is that Leopard had a terrific reception out of the starting gate, garnering some two million sales the very first weekend. I don’t presume to have the figures since then, but I think it would be fair to say that there are probably going to be another million or two Leopard users by the end of the quarter, even if most of those sales involve copies preloaded onto new Macs. My impressions of Leopard’s ongoing success are apparently confirmed by a MacNN report about the results of a recent NDP retail study.

With such a fast influx of new Leopard users, you have to expect some ongoing teething pains, but to suggest that Leopard is somehow a failure strains logic. Lest you forget, previous versions of Mac OS X also had their share of troubles in the initial releases. Tiger, for example, had problems with VPN and cross-platform networking, and it took quite a while to sort things out. However, I don’t recall a spate of public pronouncements proclaiming Tiger as some sort of abject failure. In fact, Tiger is considered a great product, and it’ll take a lot for Leopard to attain that status.

So why the difference?

Well, the Apple of 2007 is quite different from the Apple of 2005, when Tiger came out. Although Apple was doing quite well with iPod sales and all, and had garnered great reviews for Macs, sales for the latter had yet to really soar.

Today, Macs are ascendant, and growing fast. There are estimates that some 30% of Americans planning to buy new computers in the next three months intend to select a Mac. This is one extraordinary development with a product line that, only a few years back, had sales of but a few percent even in the U.S., Apple’s largest single market.

With the iPhone and iPod also growing rapidly, Apple can’t do a single thing without attracting attention just about anywhere in the developed world. That is a double-edged sword, because it means that the vultures will be looking hard for bad news to create headlines, even if they have to ignore a few facts.

Now if you have had bad experiences with Leopard, I suppose you might be inclined to believe it has been a failure on an individual level. But when I see it on Top 20 lists among the misses for 2007, I just believe that’s going way too far.