With the release of Leopard now days away, I think it is important to look at the reasons not addressed by Leopard that keep the Mac OS blocked from the corporate desktop.
Leopard will address the two most widely cited reasons why Mac isn't used by corporations: Interoperability and compatibility with Windows. We'll see if the product stands up to the rumors shortly, but indications are that Leopard will nearly seamlessly drop into Windows shops and will likely be vastly more popular with small and medium-sized businesses as a result. However, large enterprise shops will continue to have serious problems with three aspects: Single Source, Trust, and Stability.
Corporate and government shops have a requirement to competitively bid large capital equipment purchases, and this has increasingly worked against both mainframe and UNIX platforms. However, because there was no choice, variances were both possible and common. In addition, for much of the time both of those platforms existed, the requirement to not do single source was vastly weaker than it is post-Enron.
It is a variant of this problem that forced Intel to license x86 to AMD and create the competitor it is now trying to put out of business, and it goes to the heart of why Dell went to AMD and Sun recently went to Intel for PC solutions.
Although, initially, Apple could be competitively bid, once in place, the differences in the platform make realistic competitive bidding problematic, and it is much easier to bid common vendors against each other.
Through much of the '90s Apple had an enterprise sales and support group. Unfortunately poor financial performance during this time forced it to abandon this segment, and it left a number of very large companies high and dry when it did so. This was also largely true with large government, and both entities are very concerned that if they adopt an Apple platform that Apple might abandon them again, particularly given the uneven success with the iPod (After more than 20 years Apple has only 2.5 percent of the PC market; after less than 10 years, it has more than 88 percent of the MP3 market.)
There is a broad realization that were a company to make a commitment with Apple and then Apple were once again to abandon the segment, the key decision makers would be viewed as negligent given this past history. Trust is very hard to rebuild and while IBM, which had a similar problem in the early '90s, proved it could be done, it took it five years and it didn't have to overcome the perception that it is largely a consumer electronics company.
Microsoft clearly has trust issues as well, but IT is unlikely to switch one trust problem for another, particularly if they increase their personal risk by trusting a company which seems to be spending vastly more resources on MP3 players than on corporate initiatives. Stability
Corporations like to be kept in the loop with regard to what is coming, and they want assurances that platforms will remain stable for at least 12 months. Apple has historically been unwilling to share future plans and it can't assure stability without that disclosure.
The other vendors in this space, from hardware to software, conform to this requirement in order to win business even though it does create competitive risks for them as well.
The PC stability requirement goes to the need to reduce as much complexity in a large shop as possible. This is because these shops have thousands and often hundreds of thousands of PCs. Any variance, when you have this many of anything, can make the result almost impossible to support with a relatively small number of people.
Enterprise IT shops ask for and get up to 18 months of PC stability from the other PC vendors and have grown to require it. These firms are used to asking even their very largest vendor companies comply with their policies, and Apple's likely refusal to do so will probably prevent their being accepted as a viable corporate desktop vendor.
Apple Market Share Gains - Linux Impact
I do think Apple will gain share in business, but the vast majority of this will be in small to medium business and small to medium government, including parts of education which is once again showing interest in the platform. In many cases, this will be as an alternative to Linux which has not trialed well and also falls victim to a number of shortcomings on the desktop which make it vastly less attractive than the Mac OS for this use.
It will be interesting to compare market penetration on the desktop between the Mac OS and Linux, both of which are UNIX variants, against what is an unprecedented Windows weakness in the business segment. Linux should be stronger in business than Mac OS but, as mentioned in the Dell Linux piece that preceded this one, it lacks adequate revenue, profit and has too many different versions to be as successful as it otherwise might be. Linux's potential in the consumer market is tied to embedded use, where it has similar problems but appears to be more successful.
Apple's physical limitations, as the only hardware vendor selling the Mac OS, will probably be the major barrier for growth (50 percent per year is the typical limit for a hardware company) over the next few years, which may make much of this discussion pointless anyway because Apple is likely to reach physical limits, if it can execute, with its consumer and small business gains anyway.