The wait for the long-awaited Yukon and Whidbey is going to be longer still, eWEEK.com has learned.
Microsoft Corp. Director of Product Management for SQL Server Tom Rizzo confirmed that Microsoft expects to ship both Yukon—Microsofts code name for the next major update of its SQL Server database—and Whidbey—the coming update of Visual Studio—in the first half of 2005. In the meantime, a third beta has been added to the current beta schedule of Yukon, with 15 beta customers from across all major vertical industries signing up to run Yukon Beta 3 live in production settings before giving the thumbs up for Microsoft to make the product generally available.
Rizzo also confirmed that rumors about the final names for the products, gleaned from leaked screenshots, are correct: The final name for Yukon is SQL Server 2005, and the final name for Whidbey is Visual Studio 2005. According to Rizzo, both products are on the same timeframe for shipping for a key reason: Namely, Microsoft wants to release the best of its developer tools with the best of its database technology “to really change the industry,” he said. “If you look at Oracle [Corp.] and IBM and other competitors in the open-source space, they dont have releases where new and innovative tools are released with a new and innovative database. Customers want that: the next generation of tools that exploit the next generation of database technology.”
This is the latest in a series of major delays for Yukon, a build that Microsoft has been constructing for years. The companys first references to the update were in October 2000. Microsoft, of Redmond, Wash., originally cited spring 2004 as the ship date. Blaming .Net integration issues, Microsoft in June 2003 pushed the ship date until the second half of 2004.
These major delays have been sprinkled with minor delays: Yukon Beta 1 was originally promised for the first half of 2003 but didnt come out until October 2003. And so far, say testers, theres no sign yet of Beta 2.
Cause of the delay
One Yukon alpha and beta tester who requested anonymity said that, although Microsoft hadnt explicitly told him about the delay, it was easy enough to read the tea leaves by considering the scheduling of beta versions.
“If you—as I did—can put 2 and 2 together, you can see that 04 is not happening at all,” said the tester. He noted that there has been no word about Beta 2, which was supposed to be out this summer. In fact, there could be another “pre-beta 2” build in the offing. “That means Beta 2 must be a way off,” he said. He offered no theories on why the beta is late but added that, as yet, he hasnt heard “a peep” about Beta 2.
Sources said that the delay is attributable to evolving requirements, as security, integration with .Net, and competition with RDBMS rivals IBM and Oracle have all pushed the focus of Yukon development hither and thither.
“Weve called [Yukon] a montage release,” said Betsy Burton, an analyst for Gartner Inc., in San Jose, Calif. “Its being pulled in different directions by different groups within Microsoft. Senior management had a vision for where they saw Microsoft going; developers had a vision; [and] as .Net evolved, that began to influence SQL Server.” Then the Slammer worm struck some 14 months ago, Burton noted, and played yet more havoc with Yukons release date, as the development focus shifted to security. Microsoft CEO Bill Gates Secure Computing initiative also got handed down in the middle of Yukons development cycle, causing all new development to come to a halt in order to align Yukon with the initiatives dictates.
Because of these shifting requirements, Gartner, amongst many observers, had already foreseen Yukons general availability being postponed until the first half of 2005, Burton said. “Based on the evolving requirements for this product, and based on the current timing for Yukon, and based on where the product is today in terms of its beta cycle, we dont believe Yukon will be delivered until the first half of 2005,” she said.
Microsoft says no wish
-washiness involved”> Microsofts Rizzo disagreed with the premise that Yukon development has been wishy-washy, however. “If you look at the focus of Yukon over the years, weve always had the same tenets: enterprise data management, including security, reliability, manageability and availability. Another thing weve focused on is developer productivity. Integration with the .Net framework, with XML, with Web services support. And finally, business intelligence. Were taking business intelligence to the next level. Were providing an integrated, comprehensive BI platform that delivers information customers need at the right time to make the right decisions. Weve talked about those three areas of focus for the past five years.”
There are both positive and negative aspects of these delays. Users would, of course, prefer that vendors ship rock-solid products, regardless of how long that takes. “In the case of Yukon, Im confident that SQL Server customers would rather wait another year and get a rock-solid release, than [they would like to] potentially get a release tomorrow that isnt as well-tested and debugged,” wrote Steve Foote, a consultant with Enswers Inc., in Cambridge, Mass., in an e-mail conversation.
Besides, customers dont want to spend any more of their time installing major new releases of database software than is absolutely necessary, Foote pointed out. “Its disruptive of the business,” he wrote. “E.g., downtime for installation/upgrading, development modifications to take advantages of new database features, etc. Furthermore, as is the case with every new major release of any vendors software, there are always going to be bugs … usually because the major new release is rushed to market.”
What about the SA
program?”> The downside of delays with major new releases is that they skewer customers long-term, strategic decisions, Burton said. “Microsofts big challenge is resetting expectations of partners, who are trying to make strategic decisions today based on the date of the Yukon release,” she said.
Another problem Microsoft will have to face is responding to customers who bought into Software Assurance agreements with the primary goal of receiving upgrades in return for their money, Burton pointed out. “When the Software Assurance agreements first came out, the reason people bought them was to get the upgrades,” she said. “Our advice for people who bought [into a Software Assurance agreement] for upgrades is that they need to negotiate. We do believe that a certain percentage [of SQL Server customers] will not renew their SA agreements because there are no upgrades.”
Microsofts Rizzo defended the SA program by pointing out that it encompasses far more than software upgrades. “You have to go back to the beginning of the Software Assurance program and what the programs all about,” he said. “Its about building and providing a long-term, ongoing relationship with our customers and partners. And it is more than just product upgrades. We dont view it in this tunnel-vision way. Its about providing real business value that helps reduce costs for customers and has value-added services with things like support, training, deployment tools, employee discounts on software and extended warrantees.”
Still, as noted by Chris Alliegro, an analyst for Directions on Microsoft, in Redmond, Wash., anybody who bought into a three-year upgrade-rights program should be getting nervous as the time nears when these programs will lapse. Another sore spot for users: What will Microsoft do with regard to continued support of SQL Server 7 and SQL Server 2000? Mainstream support for both is scheduled to end on Dec. 31, 2005, Alliegro noted. “That doesnt give you a lot of time to upgrade [to Yukon],” he said.
Extended support for earlier
versions?”> Rizzo said Microsoft is now looking into the issue of software lifecycles, although no news was immediately forthcoming regarding extended support for either product.
In the meantime, various high-level features that would have made Yukon a compelling competitor to enterprise-class databases—i.e., IBM DB2 and Oracle—have been cut. These include, according to Burton, clustering, hash partitioning, the use of Yukon as a unifying storage engine with Microsoft applications and file system support within the database management system.
Are these high-end features important to the small and midsize businesses that are Microsofts sweet spot in the market? Not necessarily. Kevin Kline, president of the PASS (Professional Association for SQL Server) user group and director of technology for SQL Server at Quest Software Inc., in Nashville, Tenn., said that specific performance-related features like hash clustering can provide a gain, but in the relative prioritization of features, they dont necessarily come out as all that important to the SQL Server users he works with. “Microsoft might offer [such features] to match what a competitor has, but in the end, you can probably match the performance you can get from that competitor with other features,” he said.
But according to Enswers Foote, the most important items on the Yukon wish list arent such enterprise-level features. Rather, theyre run-of-the-mill tweaks that Microsoft could have added to SQL Server with incremental functionality updates over the course of the past few years. For example, Foote wrote, “the simple act of modifying users in a SQL Server instance that already has several thousand users defined is in fact quite bothersome. Enterprise Manager (from 2000) does not provide the ability to search the database users to see what names already exist—you have to scroll endlessly in order to find a name that you are looking for.”
Another example Foote cited is the adding of privileges to a role or user. If a SQL Server user has thousands of database objects, such as stored procedures, tables, views, etc., for which he or she administers access privileges, the user has no sorting or searching ability in Enterprise Manager. “Youve got to scroll endlessly to find just the stored procedure that you are looking for to add a respective privilege to the user or role you are working with,” Enswers wrote.
The list goes on, but perhaps the most frustrating aspect of the situation is that such enhancements would not require a major new release of SQL Server. All of these examples and more could be implemented in existing products with existing class libraries that Microsoft already provides, Foote said. “SQL Server customers would be a lot more likely to renew their maintenance contracts if they had been receiving even incremental updates over the past several years,” he wrote.
The failure to pump out incremental updates could also hurt Microsoft in the long run, Burton said, as enterprises back off on plans to push SQL Server into broader usage in enterprise applications. “Theyll still use SQL Server,” she said. “Theyll just not continue to push it further up into the enterprise.”
Wait before you pass judgment, though, said Microsofts Rizzo. “Its still too early to say whats in and whats out,” he said. “We have a second beta coming out. We expect a lot of feedback on what they find useful and whats not useful. In the end, requirements may have changed for customers, and technology may have changed. We expect Beta 2 to really solidify what we do in terms of features and functionality.”
Mary Jo Foley of Microsoft Watch contributed to this story.
Be sure to add our eWEEK.com Windows news feed to your RSS newsreader or My Yahoo page: