October 12, 2001
McAfee.com announced in its third quarter earnings statement that it has pro-forma results of $0.06 per share earnings on revenues of $16.2 million for the quarter, a jump of 13% from the previous quarter. The company also announced it had acquired 200,000 more new subscribers – boosting its total of unique subscribers to 1.22 million. Furthermore, McAfee declared that it was holding on to approximately 80% of its existing subscribers come renewal time. McAfee.com provides anti-virus and personal firewalls as subscriptions. Users do not purchase CD versions of the AV applications; instead the entire offering is delivered online, with updates delivered over the Internet.
While there is no doubt that computer security has become an even more critical concern over the past month, we do not believe that the events of September 11 account for McAfee.com’s strong – if not breakout – quarter. Instead, we see increasing evidence that the concept of software as a service is making it into the mainstream of the computing public. While there will always be some consumers who insist on having products delivered in a medium they can touch and hold, like a CD, it is increasingly clear that these folks will slip into the minority as time passes.
The ease and relative painlessness of maintaining one’s anti-virus application online is making more and more sense to more and more people. Anti-virus applications are pretty simple when you boil them down. The key is to minimize the time between the moment a virus is released into the wild and point at which the antidote is made available to the end user. If end users are too lazy or forgetful to update their AV products, they extend the time window and proportionately their risk of infection. McAfee.com is not the only AV company that offers updates online. All do, but McAfee.com has spent an extraordinary amount of time and money building out the infrastructure needed for a 100% online product delivery system. Apparently, the company has made that infrastructure robust and responsive enough to convince 80% of their customers to stick around. There is a lesson here. Any enterprise hoping to offer online delivery of applications is going to have to invest heavily in the infrastructure needed to support those customers, lest they see 80% defection – instead of retention – rates.
A few days after announcing that the company would post a five to seven cent per share quarterly loss, take a $500 million charge and lay off up to 4000 employees, Sun Microsystems announced a 37% price reduction in the cost of its iPlanet Web Server Enterprise Edition 6.0 from $1495 to $940 per processor. The move was regarded as an apparent attempt to wrest market share from Microsoft’s Internet Information Server (IIS) product. In an unrelated development, an article posted on Sun’s public Web site claimed that IBM’s new high-end p690 UNIX server (AKA Regatta) “is trying to do what Sun has been delivering for years,” and claimed that Sun’s own Sun Fire 6800 UNIX server performs at higher levels of efficiency than the Regatta.
One generally expects most any company to exhibit a fair bit of public contrition after announcing a half billion-dollar write-off and major layoffs, but Sun has always reveled in its self-elected role as high tech’s recalcitrant bad boy so a bit of puffery is hardly out of character. However, a company nearly two decades old is a bit long in the tooth to be play-acting as an enfant terrible, and the events of the past week bring two serious questions to mind that cannot be answered with mere bluster. First, are Sun’s losses and lay-offs due to the “well-documented macroeconomic factors” company CEO Scott McNealy prefers to blame or are deeper problems vexing the company? Second, do Sun’s aggressive and public swipes against IBM and Microsoft suggest that the company believes it is capable of taking on and winning against either or both competitors?
While the past year’s spiral into recession has caused a good deal of damage both inside and outside the high tech sector, we believe that the genesis of Sun’s problems lays a bit deeper than the general economic malaise. Over the past year, the competition in the UNIX server space has gotten fierce, with IBM, especially, gaining ground on Sun’s once apparently insurmountable market lead. Additionally, the company’s competitors have virtually all implemented strategies that embrace both Intel’s 64-bit Itanium chipset and open source Linux solutions, leaving Sun as the only vendor focused solely on proprietary business hardware and software products. Does this suggest that Sun is headed directly or even quickly down the tubes? Hardly, but it does leave the company standing squarely at odds against a substantial market shift within its own industry. So far as the company’s tactics toward IBM and Microsoft are concerned, the best Sun can realistically expect is to scrape a bit of cream off its rivals’ much broader and deeper product offerings. The currently weak demand for business hardware and Sun’s own softening sales suggest that the company’s aggressive yammering should be regarded as just what is is; so much vocal sturm und drang directed at an audience that appears to be turning to new sources for entertainment and edification.
Microsoft announced that twenty-four partners would be deploying the .NET Alerts Service to their customers sometime this quarter. Among the partners are eBAY, Bank One Corp., E*Trade, Expedia, McAfee.com, MSN CarPoint, NASDAQ.com and VeriSign. These vendors will use the .NET Alerts Service to provide their customers with up-to-the-minute information about the availability of predetermined items, such as sought-after items on eBAY, virus alerts, stocks and the like. .NET Alerts Services is designed to deliver information to users by a variety of means, controlled by the user. Email or instant messaging can alert users on desktops, and alerts are available for mobile phones, PDAs and pagers.
While we are not ready to declare this announcement as evidence that .NET has attained the full here and now, this announcement provides enough for a fairly clear look into the future of computing. We believe what we call “Service Computing” is inevitable, and that the architecture that will allow for all the benefits entailed in Service Computing will be provided by the like of .NET or SunOne, to name two contenders in the Web-based service arena. While this announcement highlights the delivery of simple messages on a variety of devices in a variety of formats, it does provide a lens for a future view that view looks something like this; Applications will be delivered, assembled and used in an entirely different way than they are today. Instead of loading huge masses of “one size fits all” suites onto each and every computer in each and every department of an enterprise, users will instead be given a much more focused package of applications that are built of componentry deployed on a much more ad hoc basis. Just as the users of the .NET Alerts Services can configure how and where they receive messages, they will in effect be able to determine what applications and functionalities will be available to them on their working desktops or handhelds.
We do not believe that each and every employee within an enterprise will be given the rights or ability to build apps on the fly within this vision of Service Computing. However, the determination of what apps are built, what features they possess and who will have access to them will eventually live much closer to the desktop than the server room where such decisions currently reside with apparent impunity. In essence, Service Computing will give department level Line of Business managers the ability to provide their personnel with applications that are suitable to the work at hand, thereby shrinking the slope on learning curves, speeding adoption times, and generally giving the enterprise end user what they need most, whenever and wherever they need it.
By Charles King
Over the course of the past week, Microsoft has issued a trio of announcements that relax or alter specific registration and certification requirements for business users of Windows NT, Windows 2000 and Windows XP. The company extended the deadline from February 28, 2002 to July 31, 2002 for its Software Assurance plan, which will eliminate some discounts and raise software prices for some business customers. In addition, Microsoft stated that customers will not have to upgrade to Office XP to qualify for the program, as was previously required. Microsoft also softened its policy for Windows XP re-imaging fees for large enterprise customers, though it left the policy in place for small- to medium-sized businesses. Finally, the company announced that it would not decertify Microsoft-certified systems engineers (MCSEs) for Windows NT 4.0 and below whose organizations do not upgrade to Windows 2000 by December 31, 2001.
While they might appear trivial to many Windows users, these events provide an interesting counterpoint to the crescendo building behind Microsoft’s upcoming Windows XP launch. The first message, of course, is that Microsoft is a happy smiles kinda company that is always willing to bend over backward to accommodate the myriad and sundry requirements of its best buddy business customers. Time to key up the violin ensemble and Mormon Tabernacle Choir.
There were, however, a couple of other messages that came through the subtext of these announcements. Subtext message #1: If Microsoft thinks it can shove new, higher priced Office licensing requirements down business users’ throats at the same time the company demands that they use a newly launched OS, they’re smoking rope. Subtext message #2: If Microsoft thinks its biggest business customers intend to cater to the company’s technical whims or pay additional fees for re-imaging, an installation process that Microsoft has accommodated for years, the rope they’re smoking must have been dipped in weed killer. Subtext message #3: If Microsoft thinks it can simply abandon companies who paid good money for Windows NT just because they do not wish to upgrade to Windows XP, the company is obviously smoking their rope from both ends at the same time. Subtext message #4: If a company decides it is going to simultaneously introduce a radical new operating system and office productivity suite, and substantially change the way they charge customers for products, licensing and support staff certification, it should put out the rope its executives are obviously smoking and gauge said customer opinion on the subject before proceeding. The moral of the story: If you insist on smoking rope, don’t leave so much around that you can be hung with it.
The Sageza Group, Inc.
900 Veterans Blvd, Suite 500
Redwood City, CA 94063-1743
650·366·0700 fax 650·649·2302
Europe (London) 44·020·7900·2819