Look before you leap when choosing between Microsoft and Netscape.
By Sally Atkins
Microsoft's Internet strategy is to challenge Netscape product-for-product. Some IS managers seem to be buying it. Should you? A look at the buying issues when considering Netscape versus Microsoft for enterprise site licenses and intranet infrastructure may help you to decide.
At stake is nothing less than the future of client/server computing. Microsoft and Netscape both hope to provide the back-office (server) tools for enterprise infrastructures. Netscape is based on open systems, including the OSF and X/Open models. Microsoft is coming from its own Windows-up strategy. Will it suffocate users or free us?
Microsoft hit the ground running with its Internet strategy in December and has made strides in marketing to the Fortune 500 since then. Netscape has no equivalent enterprise marketing machine. It has a network of VARs and consultants that get the job done, but nothing to compete with Microsoft's live-in account reps and internal consultants in big companies. What Netscape does have is tools that work, based on established Internet utilities, and the Internet itself as a marketing tool and distribution medium. But Microsoft has caught on to that, so it is not a lasting advantage.
Microsoft has announced plans for products and technologies intended to make it the dominant player in Web servers, browsers, access and security. Through its tried-and-true method of controlling APIs and extensions, it will attempt to dominate the Web software market. Yet, to further this ambition, the master of proprietary standards is doing an open systems tap dance. This is a holding pattern, waiting on vaporware, but there is reality underneath all the marketing bluster.
Microsoft's strategy, in a nutshell, is to win the mindshare of corporate developers and their managers and CIOs while conducting the software company equivalent of a barn raising for its Internet product lines. A prime example of this approach was revealed when Microsoft licensed Java on the same day as announcing own scripting language, based on Visual Basic.
The Real Message
Cruising the Microsoft Web page for CIOs for clues to how the Netscape-Microsoft buying challenge for IT managers will shape up, I found plenty. This is a battle to be waged in the trade papers and the stock market. It will match the scale and intensity of OS/2 versus Windows all over again. Only this time, Microsoft is in IBM's role, the rich uncle who is less favored to win.
The CIO page promises to help developers to Internet-enable desktop applications; there is a promise that they will "enable the evolution of the Internet," not trying to change the base standards of TCP/IP, HTML and HTTP. The strategy is to build hooks on top of what exists and try to steer the market going forward. This quotation from Microsoft says it all:
All the noise aside, the Internet does represent the greatest opportunity for software innovation since the introduction of the personal computer. By both embracing the existing standards and using appropriate extensions, you'll be able to write killer Internet applications.
While the marketing and stock price wars are interesting to watch, there are real decisions to make. The following criteria should play roles in the process.
Functionality. Decide what you need the suite of Internet tools to do for you. Requirements analysis, we used to call it. Authoring, content management and Web site maintenance are a few common needs. You might ask how well the product lines deliver an integrated solution to your needs today. If you are in no hurry, you might wait for Microsoft's Internet Studio, which should pick up where FrontPage leaves off in functionality for larger Web sites and more interactivity.
Culture. If your organization's skill base is in Unix, the Netscape product line makes sense. If it is a Microsoft culture, Netscape can still make sense as it runs on NT. However, inserting the Microsoft product lines into a Unix culture makes no sense to me. If your systems administration staff is trained in Unix, why go backward when the rest of the world is trying to catch up to where you are? The Microsoft tools are not compelling enough for that sort of wrenching change. If you are lucky enough to have a culture of coexistence, as more and more IS shops do, it becomes a question of interoperability and integration.
Architecture. Does the suite fit smoothly with your architecture? For instance, why would you run a 16-bit application on a 64-bit server? Why run a Microsoft product on a Unix server? These are questions I ask my clients as they lean toward installing Microsoft Internet products on their zippy new servers.
Size and scalability are major factors. Keep in mind that clustering is not something NT does well. And while scalability is less of a problem now with NT, Netscape server products scale far better than NT to a far wider .range of hardware platforms, due to their Unix ties. The upside of the Microsoft architecture is that it can be simple for all-Microsoft shops. If all you have is Microsoft, you have no integration issues.
The bottom line for now is do your homework. What is your organization's Internet applications architecture? What are its business goals for the Net? If you have none, don't build an architecture, and don't buy any tools. You are throwing good money after hot air.
If Netscape is to contend in the intranet market over the long haul, it must get off its bucks and put some skin in the corporate account game. Its partner, Andersen Consulting, doesn't have the influence it will take and is not loyal enough to Netscape. It will take Netscape marketing people with as much single-mindedness as the Microsoft people to win in the short run. In the long run, the products may stand on their own.
Sally Atkins is president of IST Consulting, an affiliate of NetSource, Inc., based in Boston. She can be reached at Sally@kins.com.