Notes postmortem
Sine readers suggest reports of Notes’ death may be greatly exaggerated, but save the obituary just in case
LAST WEEK’S COLUMN (See ” Notes is Dead “) touched a deep nerve. I expected it would; I did not undertake the piece lightly, as some have suggested. Notes at its core is highly personal, and its proponents take it that way. I know — I’m one of them.
“With friends like these …” you say. Notes has many dear friends, 20,000 Lotus business partners being just the most visible among them. Here at InfoWorld some of us complain about the vagaries of Notes mail, but not about the seamless workflow applications that clear the way for vacation requests and glue together sales, research, and communications.
In the world before the Internet, Notes stood alone as a bulwark against the deluge of enterprise information. Arguably, Notes pre-invented the Web, albeit with proprietary protocols. Its linked infrastructure and separation of form from data extended the promise of HyperCard and Visual Basic to the enterprise.
The browser wars triggered a corollary war for Notes’ collaboration platform. Netscape acquired Collabra and stitched together a suite riding on top of a pastiche of Web, messaging, and newsgroup protocols. Microsoft tried to move away from Exchange’s MAPI (Messaging API) legacy toward the same LDAP, IMAP, NNTP, and SMTP protocol base.
Lotus, newly acquired by IBM, bootstrapped some Big Blue Web server code to bolt on the Domino extensions — a URL-addressable Web standards approach to the Notes object store. At the time, Lotus (and Microsoft) felt the core Notes value proposition lay in its replication technology. In retrospect, Lotus may have been protecting the wrong crown jewels.
The Notes single data store is the singular innovation: an unstructured database upon which mail, calendaring and scheduling, and line-of-business applications could be constructed, layered, and linked. Some ports were less successful than others; Lotus Organizer’s group calendaring code suffered in the migration to the Notes mail database template.
But Lotus developers had a powerful mechanism to deliver custom applications that could be rapidly prototyped and then rewritten into the Notes kernel. In the collaboration race, Lotus could move faster by bringing Web protocol access to the Notes store than Netscape could integrate a suite of separate applications. We all know how Microsoft handled Netscape.
With Exchange 2000, Microsoft cloned Domino — and its URL-addressable access to the data store. But that set off an internal scramble between the SQL Server team and the Exchange team for control of Bill Gates’ dream of a single data store across all servers and the operating system.
Ironically, the ultimate losers in this Mad, Mad World race were the two earliest movers: Exchange and Notes. Exchange 2000’s Web Storage System used XML, but in an early form that required revamping as SOAP and the Web services stack advanced through the standards process.
Notes was facing a much bigger enemy as it rushed to bootstrap XML’s second wave of the Internet. To be blunt, IBM is Notes’ greatest threat today, and Lotus business partners know it. Much was written about IBM’s acquisition of Lotus. Would Ray Ozzie stay? Would Big Blue’s button-down style mesh with the Lotus Cambridge culture? And so on. In fact, IBM did marshall the transition effectively on the political front, boosting Notes seats to the current 100 million level.
But they fractured their developer community by bungling the integration with WebSphere — tiptoeing around the LotuSphere developer conference with conflicting signals: support for the native Notes file format in the forthcoming R6 release but yanking useful Java extensions previewed in early betas. The end result: a brutal combination of disingenuous infighting and lackluster execution.
As developer Luke Kolin wrote on the Lotus Developer Domain Forum, “Notes is dead, because it doesn’t do anything [particularly] well anymore. It’s not a great e-mail client; everyone admits that. It’s a great rapid app-dev environment, but doesn’t have the extensibility that a pure Java environment would provide, and its own Java implementation is halfhearted and half-baked. From a Web perspective, you don’t get standards-compliant HTML or even basic JSP/session support.”
Meanwhile Microsoft has moved toward the .Net Framework, with its powerful constructs for simplifying and speeding development of Web services applications. This force of nature will bootstrap Web and peer services to deconstruct lines of business applications into fluid components that can be reconstructed and reused at a fraction of the cost of current integration and hybrid so-called cross-apps.
Of course, I’ll leave it to the father of Notes to underline the truth behind my “Notes is Dead” message. Responding to its predecessor “The Death of e-mail,” Ray Ozzie blogged: “Not, of course, that eMail is going away; of course it’s not. E-mail is the place where most conversations begin, and will remain forever a critical business tool … particularly with regard to acting as a filtering proxy that dispatches notifications to your wireless mobile devices, wherever you happen to be.”
Ozzie, like any responsible parent, is speaking to the concerns of Notes developers. Increasingly, Groove is providing the constituent pieces to make a rapid transition to the .Net infrastructure: harnessing Visual Studio as the development tool, using the Compact Framework to speed porting to wireless devices, adding an Edge server to allow componentization of Groove’s peer services, and most strategically, adding doclink infrastructure to allow porting of Notes application constructs to Groove and Office.Next.
Deep Throat said “follow the money.” Bill Gates said “feed the developers.” Notes didn’t fail. It succeeded, by defining the boundaries of the .Next architecture.
