This article may contain URLs that were valid when originally published, but now link to sites or pages that no longer exist. To maintain the flow of the article, we've left these URLs in the text, but disabled the links.


June 1999

Microsoft Systems Journal Homepage

Editor's Note

Any self-respecting developer has seen all of the Phantom Menace trailers at least twice—once in the theater, having spent the night on line, and again on the net (http://www.starwars.com). For those among us who have been orbiting Coruscant for the past year, the next Star Wars installment is actually Episode I; it tells the story of Luke Skywalker's father, Anakin. The production values of this prequel must be amazing—perhaps even rivaling Windows 2000. But after the fifth viewing, we may hear about the technical flaws that slipped past George Lucas' eyes.

    We should talk. In the May 1999 issue, our very own Obi-Don Kenobi Box fell victim to the Dark Side of our editorial process. A few figure Naboo-boos slipped past our defenses. Didn't you catch them? Oh well, we're coming clean anyway. In Don's article " Windows 2000 Brings Significant Refinements to the COM(+) Programming Model," two figures contained errors. In Figure 11, The ThreadingModel Decoder Ring, the Main STA Thread member of the left column (headed "CoCreateInstance called from") should fall into the same "New Object Activated in" category as Aux STA Thread and TNA (STA Thread). In Figure 14, Transaction Streams, the transaction stream containing Contexts B and C should be labeled Transaction Stream Y and only Context B should be the root context. Only one context per transaction (the first context in the transaction stream) may be the root context. Naturally both of these figures have been corrected on our Web site (http://www.microsoft.com/msj/0599/complusprog/complusprog.htm).

    Also last month, Don used the Force to coin a new word for the English language—"Marketecture." He attributed marketecture to the Name-That-COM contest held almost annually in Redmond. Hmm, could we have used it to apply to Windows DNA? If you're lacking assistance from C-3PO, information on Windows Distributed interNet Applications architecture can be found not only in this very issue, but at http://msdn.microsoft.com/library/backgrnd/html/dnasolutions.htm, and at TechEd, http://events.microsoft.com/events/teched.

    We first reported on Windows DNA in the December 1997 issue. Since then Windows DNA has matured. It encompasses as many technologies and tools as an X-Wing fighter, but basically it's composed of three layers: presentation, business logic, and data. The presentation layer defines the entire range of clients in this three-tiered architecture. This layer includes the Win32 clients which we've all known and loved. COM-component-based, DHTML, and scripting clients are followed by the simplest and thinnest HTML clients. COM+, MSMQ, and IIS define the business logic layer. Hmm. Where have we been reading about these technologies? ADO, OLE DB, and XML—the foundation for the data layer—provide access to many data stores including mainframe, directory, RDBMS, email, and the file systems. DNA analysis may have yet to prove that Al Gore is the father of the Internet, but we feel that Windows DNA is now quite a bit more than the marketecture it was last year.

    You can continue to rely on MSJ and our Rebel alliance publication, Microsoft Internet Developer, to keep you well ahead of the marketecture and armed with the hottest goods this side of Tatooine. Your other ally had best be MSDN (http://msdn.microsoft.com), which has recently received a major overhaul and merged with the former empire, Site Builder Network. We love the hyperdrive search capabilities and configurable UI. Congrats to the new MSDN!

J.F.


From the June 1999 issue of Microsoft Systems Journal.