Workplace dead, parts are still alive
by Volker Weber
Workplace Collaboration Services is a product that few people understand. With that I mean IBM people. Outside the company you will be hard pressed to find anybody who can explain it. And then, ask them to explain the technical differences to Workplace Services Express.
Now it looks like I will soon have to retire this t-shirt. A little birdie tells me that the next version will be called Cold Dead Fish Advanced Components, sorry, WebSphere Portal Advanced Components. And that is exactly what it is — stuff that runs within Websphere Portal. Let's hope it will run in the current version of Portal, which is about to go to 6.0.
Update: Another little birdie tells me, it might also be called Services instead of Components. Also watch out for magically disappearing products formerly sailing under the Workplace brand.
Another update: Surjit is no longer VP of Marketing at Lotus. You have to learn an (old) new name: Marjorie Tenzer, formerly Vice President WW Software Group Channel and SMB Marketing.
Comments
Vowe,
It's funny that you blogged this as I was thinking the same thing about the whole Workplace brand and what's it's future will be. Now that Hannover has been announced and hopefully will go into beta soon I think that most of the Notes and Domino world can now finally say good bye not only to your t-shirt but also to the most confusing product/branding initiative ever created by IBM.
Uh oh, I hope all the customers won't be as upset as the Sametime customers were when Sametime was renamed. Oh wait, Sametime had customers. This whole things seems to have been a very expensive experiment by IBM.
It pays to mention that the same "services" and "components" that represent the bulk of the value to come from the workplace experimentation and effort are also going to be surfaced as parts of Hannover and future Domino servers.
It looks to me like when they started working to strip down workplace to the core of its value add, it made sense to move those parts into existing product lines and let the rest go.
Frankly, I applaud the move.
Surjit's blog doesn't know this :-)
http://www-03.ibm.com/developerworks/blogs/page/surjitchana
hm. well maybe they'll bring back some form of Garnet now... ;-)
Bruce, his last posting is from February. But he has been seen alive in Karlsruhe at DNUG. Maybe he does not know, that he once had a blog.
Andrew, there are high expectations for Hannover, but success is not a given.
Volker - that depends on your definition of success. Hannover is a major revision change on an application with more than 120 million people. Even if the stuff being brought over from the workplace effort fails utterly (and its hard to imagine for a few things) it will still be an excellent Lotus Notes client.
More than 2 years ago, I made the statement that even if workplace fails completely, the rich client with Notes was going to result in good things for the Notes user base. I still hold that to be true.
--Andrew
Predictions are hard. I know you are always confident about yours.
I will reserve judgement until I have seen how "Hannover" works on existing hardware, especially if you are also running Sametime 7.5 and some other business applications.
Check out this article I wrote:
http://www-128.ibm.com/developerworks/workplace/library/wcs-architecture/
See if that helps...
bob
Bob, it's a very good 30 minutes explanation. I was thinking about a good five seconds explanation (This car lets you take people and stuff where there are no roads) or a subsecond explanation (Land Rover).
Volker -- It's as you say. I am confident in my predictions and in fact still have the presentation from London where I gave the first non-IBM overview of what was at that time called the Notes Plug in to the eclipse workplace rich client (and later as a whole became Hannover) in which this statement was made.
I'm by no means always right -- but with this particular path I've been out ahead saying things about the emperor's clothes that have been proved right fairly consistently.
The big Websphere push -- I predicted that websphere would not come close to supplanting Domino, even before the days when the word Domino disappeared from IBM's meeting rooms as a viable solution to anything for about two years. I predicted Portal would fail to capture significant market share, and I predicted that workplace's success would be all about Notes.
The key to this name change is what the formerly non-allied with IBM Bob Balaban used to call "Kremlin Watching". Its about studying the way IBM's internal changes are reflected in sometimes strange or subtle outward changes -- like who is standing closest to the Premier at the press conferences.
In this case, keep in mind that I'm not aware of any instances where anything can outwardly be said to have "failed" at IBM software. Websphere the server became WAS so that Websphere could become a brand, and as it still didn't sell to predicted volumes MQSeries was added to the brand. This legacy style IT product is a bread and butter addition and its volume completely masks the independent numbers of WAS.
Workplace went from a product to a family, and that family came to include Notes and Domino, along with Portal. Portal technologies started finding their way into Domino and Notes.
Finally, the "plug in" of Notes to the Eclipse framework was so rapid and so successful that Notes quickly came to drive the definition of features and functions that make up the Workplace framework. Its this framework that makes "Eclipse" into the IBM Workplace Client Technology.
Aside from this Eclipse based framework with its wiring and integration definitions now being defined largely by the needs of the Lotus Notes products, the other real key marketable power to come out of all the workplace effort is the Activity Explorer. The limitation on AE however, is that if you require Portal to make it work, you're talking about too large an investment in hardware, time, and money. That means nobody can try applying it to smaller problems. It can only be used for problems that justify spending over a hundred thousand dollars. How do you solve that? You do two things. You try to "collapse the stack" required to run it within workplace by paring down workplace to manageable hardware and software requirement in terms of cost, power, and complexity; at the same time you see if you can pull out the AE technology itself and put it into less expensive platforms. Those two steps really force you to look closely at what exactly makes AE powerful, and what is just stack. Do that, and soon you find you have something you can embed in Domino and Notes.
Finally you get to the crux of the situation. Workplace resulted, ultimately, in a couple of core technologies --- the Rich Client framework on Eclipse and the Activity Explorer -- that seem to have market legs. Both can be well leveraged in the Lotus Notes and Domino product line which have proven to be harder to duplicate than some at IBM seemed to have thought. In the end, Workplace as a distinct product and replacement for Domino dies a horrible slow death in a classic case of "Second System Syndrome". (see: http://www.thenorth.com/apblog4.nsf/Threaded/4E042BEC04D4A27285257186000B466F )
Andrew, with respect to your "I predicted Portal would fail to capture significant market share", WebSphere Portal is #1 in the portal market. The Gartner Magic Quadrant report (May 2006) has IBM ranked #1 in Ability To Execute and Completeness of Vision.
Alan -- what portal market? It's less than 10% of the market Gartner predicted would exist when they made that prediction 5 years ago. Portal -- which IBM claimed would "over time" overtake Domino in use is almost invisible to 90% of the IT market.
Andrew -- that's because another name for "portal" is "bottleneck."
Portal... is almost invisible to 90% of the IT market.Ouch! Really? Who are you quoting? ;o)
I see a lot of portal-oriented technology here, there and everywhere, but admittedly that's just in pretty large organisations...
I can well believe that portal technology is all but ignored in most SMBs -- hence your comments Andrew -- but as a business I wouldn't discount portals. They're so complex and fiddly, they must make a fair amount of money for the big vendors and their respective business partners ;o)
Do you mean "a very expensive exercise in showing independence"? ;-)
ROTFLMAO... that was a great quote.
"I will reserve judgement until I have seen how "Hannover" works on existing hardware, especially if you are also running Sametime 7.5 and some other business applications."
My take from this comment is that you expect Hannover to be a resource hog and that people will have trouble running it as a result. But large corporations will probably roll out Hannover about the same time that they roll out Vista. Based on the realworld hardware requirements that we've been reading about for Vista, I'm guessing that in comparison, running Hannover on a Vista PC won't be a big deal.
Is this a reasonable guess? And what do you base your "Hannover on existing hardware" concerns on?
My problem with Portals as a market is, and always has been something like this:
A 'Portal' is itself by definition a connecting passageway -- a doorway, a gate mechanism, or a translation from one thing to another. As such, its only value derives from its ability to bring things together that otherwise could not be easily navigated -- things that don't communicate well with each other. In other words, its an expense incurred to make up for a failure in strategy or design. That failure could be cosmetic, complexity (multiple passwords), or integration (language, voltage, storage location, or other technical differences). In a word, its "Overhead".
The portals sold by IBM and its rivals take "Overhead" to a new level, with project easily exceeding one hundred thousand dollars before you even start talking about large scale systems. As such, they are useful only in cases were a project that large is itself just a small part of the overall reason for being, and thus is a justifiable overhead.
They are then, by definition, limited to the largest customers who place a high enough value on the integration of many large and complex systems into a cosmetically cleaner, and complexity reduced interface that they can justify an expense equal to the annual gross of the average SMB without itself actually accomplishing any new technical solution to an existing problem.
It is thus extremely self limiting. Sure, IBM and CSC like Portal. It spurs sales across the hardware, software, and services business with big engagements over long periods. Nearly all "Portal" projects require direct intervention of one kind or another by the services organizations of the companies who make the Portal products.
I would never, in good conscience, recommend such a thing to a client.
Andrew has a point. The big steps forward are the activity explorer and the Eclipse RCP framework (supercharged by the managed client extensions). The activity explorer has been spotted running of a regular J2EE stack rather than portlets. Since you get a WAS licence with a Domino7 you could be ready for activities today. The Notes8 client will add offline activities.
Digging through the marketing confusion we have some nice building blocks. It will be interesting how they are combined and marketed:
- Websphere Everyplace Deployment (this is the foundation of the managed client, the management part, it's Eclipse RCP on stereoids)
- Notes/Domino with it's RAD capabilities
- Workplace designer: a study how to have an alternative RAD designer (with nicer XML and JavaScript support)
- Eclipse/Rational IDE: were all the dev tools will come together
- DB/2 Viper: Native XML storage capabilities nicely fit with Domino's DXL (that should be feature complete in R8).
Now it's time for marketing/pricing to sort that out. And to pick up vowe's idea about the unmanaged client.
My 2c
:-) stw
