"The history of this kind of collaborative effort is bad. We have so many examples of failures," said analyst Rob Enderle, of Giga Information Group. "It's difficult to merge products that come from different code bases."
But when faced with the decision to either sell two versions of the same software or kill one off, Sun and AOL's choice to merge overlapping products is reasonable, said analyst Dave Kelly of the Hurwitz Group.
"If the alliance is able to demonstrate significant independence and effective leadership and the ability to deliver on the product plans, then they will have a good shot at being successful," he said.
The alliance will continue to sell Sun's recently updated NetDynamics application server and a new version of Netscape's Kiva application server as separate products through the end of the year--and merge them in early 2000.
Similarly, they will release new versions of both Netscape and Sun's software for corporate email, collaboration, and calendaring, and merge them in the first quarter next year.
The joint effort will keep Netscape's directory or "search engine" server, with features to be added from similar technology developed by Sun.
While the two companies do have duplicate versions of the same product, their software is complementary for the most part, Kelly said. Both Sun and Netscape have gone after the same market to provide companies with tools to build Web sites for e-commerce, internal communications, and external communications with business partners and customers.
Mark Tolliver, president of Sun's consumer and embedded systems unit and head of the alliance, believes the joint effort will be successful.
"We're starting with the leading product portfolio today," he said in an interview with CNET News.com. "It's not like we'll sit back and have a little science fair for a few years and figure out what to do. We have great products now."
Some analysts agree, saying Sun's access to Netscape's e-commerce software allows the company to compete against IBM, Hewlett-Packard, and others to provide a complete line of hardware and software products.
"I don't see a big conflict," Kelly said. "Clearly there's significant overlap with the application server, but they couldn't have made the alliance without it. It's a small problem to have because they get so much more. They now have a good shot at being a strong competitor to Microsoft, HP, and IBM."
Kelly said his clients need to choose an application server this year for their Internet needs, so Sun has to do a good job of marketing the two application servers, and explain to customers which one meets their needs and why.
For example, the Netscape application server can support more users, while NetDynamics has better development tools, he said.
"They have two strong products on their hands and they need to rationalize the products. The challenge for a customer is: 'If I don't have a preference, how do I pick one?'" he asked. "It's very much like Microsoft had with Windows 95 vs. Windows NT three or four years ago when there were lots of questions on what to pick."
Sun's Tolliver said the alliance will release a new version of Netscape's application server in May and will continue to support all existing customers as the company works to merge the two products.
Tolliver said Sun plans to combine the best features of each application server. And analyst Anne Thomas, of the Patricia Seybold Group, believes it's feasible. Based on a briefing she had with Sun executives, the company plans to only add Netscape's fast "load balancing" feature--the ability to distribute transactions evenly--to the NetDynamics application server.
That means more of the NetDynamics application server will survive than Netscape's, she said. Netscape's load balancing feature is fast because it uses a proprietary programming interface, so the new alliance has to figure out how to move that feature over to NetDynamics, which currently uses a standardized technology to perform the same function, she said.
Kelly added that Sun's decision to use Netscape's directory server and to merge the separate calendaring, messaging, and collaboration software makes sense. "I expect Sun will focus on adding performance and scalability using Sun's technology, that they will offer 24/7 high availability," he said.
Other analysts however, believe merging products will be difficult.
"Sometimes gluing together separate code bases simply doesn't work," said analyst Vernon Keenan of Keenan Vision in San Francisco. "The thing that comes to mind is the Informix and Illustra merger. That was a huge disaster."
Informix attempted to merge its relational database with Illustra's object database to better compete against Oracle and other database makers planning so-called universal servers, Keenan said, adding that "It never worked and that [hurt] the company."
Enderle agreed, citing IBM's efforts with Taligent as another example. "It's more typical to take one product and just run with it. Most merged products don't work out," he said. "What's worked in the past is one company takes the clear lead. That's how Microsoft's and Intel's projects happen. And in each case, one company owns the project and the other helps.
"It shows a lack of experience," Enderle added. "This is the kind of thing IBM has tried, merging their own products, and it's not successful. It's better off starting [from scratch] because code bases are so different."
Overall, most analysts believe the alliance will benefit Sun.
"You have to appreciate that this alliance represents a complete solution for enterprise software buyers," Keenan said. "The alliance is a decent, competitive threat against IBM. What's unique about IBM is they have the network, hardware, operating system, applications and servers to integrate it."
But success comes down to creating the merged products, Kelly said.
"With AOL's access to consumers, Netscape's infrastructure capabilities, and Sun's ISP and platform orientation, they are well positioned to be a leader," he said. "But it will come down to execution. I would watch for any changes in product ship dates or people leaving--changes in leadership or management--as an indicator of success."