IBM and Notes: fumbling the future, or Hobson's choice?

Forbes recently published an article by Dan Lyons entitled "More Bad News for Lotus Notes." The gist of the story was that, according to Gartner and IDC,  Microsoft has overtaken Lotus in email share. I'm sure there's a lot of room for error in these email surveys, and the email faithful will argue forever over a percent here and there.

Dan's been pounding on IBM/Lotus for a while. He's been entertaining, but his impact is now muted by repetition. I think he may be missing a larger point, by focusing on email.

My view -- perhaps one-sided -- is that the email share change has been going on for some time, but it's not the point. It's the symptom and not the disease. The disease is a shift in the collaboration platform (email, conferencing instant messaging, VoIP, portals, conferencing ... ) that customers are choosing. Email is simply the collaborative application that's been around long enough to measure well, and regularly.

I see a lot of customers and, in the last year, I've seen real growth in the number considering or executing a shift away from Notes. Clearly this is a self-selecting set: I don't see many who are doing nothing or doing the reverse. However, the increase in absolute numbers is striking. It's driven -- as the customers tell it -- by:

1) Pull: strong "2nd release" versions of Microsoft collaborative applications like Sharepoint, and Live Communications Server. These new products shifted the debate from Exchange vs Notes (never a real "apples-to-apples" comparison, and one Lotus found laughably easy to defeat) to Microsoft's collaborative platform vs Lotus' collaborative platform. This is a much more reasoned comparison.

2) Pull: IBM announcing its transition away from Notes to a new "Workplace" product line, based on Websphere and DB2. The things that made Notes unique -- it's approachable programming model and its replicated database with fine-grained security -- have no parallels in the world of Websphere and DB2. IBM has done a commendable job -- under the circumstances -- of committing support to Notes for some unspecified long term. They did the same thing for OS/2. But why put at risk this huge customer base? Is IBM simply fumbling the future of collaboration -- a market they once owned? The folks at IBM aren't fools: this is Hobson's choice for them. They have to consolidate around a few key technologies for database, app platform, etc. They can't keep two infrastructures going forever, as analysts and customers will always ask: now which one is your/our strategy? But a drastic change of platform -- with no forward compatibility for applications or skills -- is taking its toll on the Notes community -- IBM's only base of real end users.

However, migration isn't always easy and customers also have to deal with coexistence. Many Notes applications (including email) can move easily to alternatives like Websphere Portal or Sharepoint. However, the big issue for Notes customers is the small set of large and complex applications that are deeply tied to Notes. The Notes platform is powerful and unique (as well as idiosyncratic -- its biggest problem) and there is no easy path to anywhere for complex Notes applications.

Which brings me back to the Forbes article. In it, Gary Devendorf (an ex-Lotus employee now at Microsoft) was quoted as saying "We’re definitely targeting Notes developers."

Perhaps rescuing or helping convey the situation better than targeting.

Gary is answering a growing demand to help customers extend or capitalise on these Notes applications, and -- even more important -- the vital business data they have locked away. He's leading a session at the upcoming Microsoft Professional Developers Conference, and it's mentioned in his recent blog entry.

I wonder when Dan Lyons will write the real story?

Comments

[Jim Bernardo]

Cliff, your comments raise an interesting point. When we were all at Lotus, we certainly helped stoke the fires of a religious war between Notes and Exchange, and clearly IBM is still trying to do that (if you doubt that, go to "The Boss Loves Microsoft..." session at Lotusphere). Like you, I see lots of customers, and what I hear from them is that they're looking at Notes as they do any other technology, and their investments in it in the same way. That is, the technology itself, and the applications built on it, have lifecycles, and frankly, many are underwhelmed by what IBM is putting forward as a successor technology. So, in the same dispassionate way they evaluate virtually every other technology decision, they're evaluating their options around collaboration, and, to your point, many are seeing the IBM path as a Hobson's choice. Good news for Microsoft, and, frankly, sad for those of us who put so much of our lives into building Notes into what it once was. But, it's a technology stack, and like any other, as Steve Mills himself said, there comes a point where you have to make some fundamental changes! :-)

[Richard Schwartz]

Welcome to the blogosphere, Cliff! Overall, a well-balanced article, with just one sentence that I take serious issue with. My further thoughts are here:

http://smokey.rhs.com/web/blog/PowerOfTheSchwartz.nsf/d6plinks/RSCZ-6G26WC

-rhs

[Cliff Reeves]

Richard. First I have to commend you for having the fortitude to respond in this tiny typepad window. Jeez. It's as though I didn't want a reply :-)

Readers will have to go to your site to look for your reply. I hope they do. It's worth the trip.

You assert that IBM offers significant upward compatibilty for Notes skills and aplications.

Upward compat and skills transfer are two different animals and shame on me for mixing them.

Applications (upward compat): they'll run in the future Workplace environment. However, don't change them!. Don't look for enhancements to the existing environment.

People have made the point that there are new are new Workplace apis available to Notes developers. True enough, but not the point. What;s the right way to to write new apps? It's not a blend; it's a choice.

Skills transfer: there is no Lotuscript or Notes object model in Workplace. Let's stop faking this one.

[Richard Schwartz]

The tiny edit window is actually not as hard for me to deal with as the tiny font that the post and responses are rendering with. :-)

Re: don't change them... From 6 to 6.5, and from 6.5 to 7 there have been relatively few enhancements to Notes programmability. Certainly far fewer than from 5 to 6 or from 3 to 4, but not out of line with the level of change between 4. and 4.5. Any Notes programmer can point to a laundry list of enhancements that we would like to have seen years ago, but have managed to do without anyhow. It's a very mature programming model at this point, and IMHO it can do just fine without a lot of enhancements in most areas. What it needs is two things: enhancements that open up more flexibility in UI -- and that's exactly what we're going to get from integration of Notes with Workplace Rich Client Technology, and we need the programmability to continue keeping up with emerging technology trends, and I fully expect IBM to do this. We're seeing web services now, and who knows what is to come later?...

I give on the "what's the right way?..." issue. About a year ago I compared Workplace to PL/1, offering too many ways to accomplish the same thing. OTOH, look at all the choices that .NET offers in terms of languages, in terms of an ASPX model versus a Windows app model; and layer all the various third party options and choices between ActiveX, Flash or AJAX for rich UI interactions. And at an even higher level, there are additional choices to make about which Microsoft products to incorporate into a solution: Exchange, Sharepoint, Sharepoint Portal, BizTalk, InfoPath, SQL Server, Office... (And where does OneNote fit in, by the way? It looks to me like it has a lot of very cool possibilities, but as near as I can tell, it's mostly being treated as an island unto itself rather than as another piece of the big picture.) Anyhow, I can just as easily argue that Workplace is hardly unique with respect to this "What's the right way?..." question.

Finally, regarding skills transfer... Who's kidding who here? The only reason VB.NET exists, IMHO, is to make .NET more palateable for pointy-haired managed who see that the name is still "VB" so training their programmers in it must be simple -- not nearly as hard as teaching them a whole new language; but it's the framework, not the language that is the real issue, and there's virtually no difference between training a VB6 programmer on VB.NET versus training them on C#.NET. IBM is at least being up front about the fact that developers are learning a new language. Sure, IBM's got some work to do, especially on the issue of providing a Notes-like object model for non-Domino-based Workplace apps. (Deliberate choice of phrases here, btw, to emphasize that Notes and Domino are *part* of Workplace, so anyone who wants/needs the Notes object model now can still use it and still be falling under the Workplace umbrella, but I digress...). This is something I've been talking about since I first saw Workplace and compared it to a third-party J2EE-based product that included a near clone of the Notes object model that simply blew away what IBM was showing. It's clear, however, to anyone who knows who is actually working on the Workplace Designer and what they have worked on in the past, that a Notes-like object model is going to be there in Workplace. It's not there yet, but neither is there an overwhelming need to have it today.

[cliffreeves]

So we are agreed on the "two programming models: Notes and Workplace" point. However, you seem confident that "... a Notes-like object model is going to be there in Workplace ..."

It would have been a good idea for IBM to to make that a design point and say so clearly.

The advantages of that approach are so obvious that if it were going to happen, IBM would have said so by now.

The damage is done now.

The remaining point is skills transfer. That's based on two facts:
1) there's no Lotuscript in Workplace
2) the difference in programming and object models inhibit skills transfer.

Thanks for your thoughtful reply.