-
Created on Tuesday, 06 April 2010 18:49
-
Written by Chris Heller
I was having some beers at Pleasaton's (in)famous Hopyard last night and the topic of IBM and PeopleSoft came up. Specifically in the context of the announcement at the last PeopleSoft Connect conference about how we (PeopleSoft) were going to start heavily using the IBM WebSphere technology stack. Since I also had a regular blog reader come up and harass me at the table about a lack of blog entries this year, I figured that this would be a good one to write-up. Besides, it's been awhile since we've shared any good behind-the-scenes dirt :-)
So, what happened?
If you'll recall, at the time of the PeopleSoft Connect conference in Sept. 2004, PeopleSoft had been battling Oracle for about 15 months. Aside from dealing with Oracle, PeopleSoft internally had a major issue going on trying to figure out what to do architecture-wise. SAP had shipped it's new
NetWeaver platform earlier in the year and were making lots of noise about it. PeopleSoft sales people and industry analysts were all asking "What is the PeopleSoft equivalent of the
refrigerator slide?"
It was a classic example of something that was an issue in the sales cycle, but not an issue that PeopleSoft customers really cared about. Sales is always an issue to a public company, but when a single quarter of bad sales could mean that your company gets acquired by one of your biggest rivals, then you really don't want to miss your numbers, so the lack of a "next generation" application platform was a big issue.
PeopleTools 9 / PeopleTools X
We've written before about the ill-fated attempts at coming up with a
PeopleTools 9, so we won't go into that too much here, other than to point out that by this point (mid-2004) that the name PeopleTools 9 was jinxed enough internally that the last attempt at coming out with it was called PeopleTools X instead.
At the same time PeopleTools X was being scoped and worked on, there was another effort under way to work a deal with IBM. What IBM wanted was for PeopleSoft to completely switch over to WebSphere as a development/runtime platform. This would go way beyond the way that PeopleTools ran on top of BEA WebLogic and Tuxedo; instead it would be moving everything over to WebSphere. As Craig Conway was quoted in the press release that came out at the Connect conference:
"This is the most ambitious, most aggressive announcement from IBM and PeopleSoft," Conway said. "Maybe the most ambitions from the enterprise applications business ever."
Definitely ambitious and aggressive, but the term that we used internally in the PeopleTools group for it was something that rhymed with buster duck.
The problem wasn't so much with WebSphere itself, it was with the idea that IBM could somehow come up with a hundred-plus engineers that were WebSphere experts, match them up with developers in the PeopleTools team, and the combined team could come up with something that would work for customers in a fairly short timeframe (while also continuing to put out additional PeopleTools maintenance releases along the way of course). PeopleSoft didn't have the bandwidth to make this happen, and it didn't seem like IBM would be able to steer it to successful completion either.
An Alternate Ending
So, what would have happened if Larry Ellison had given up on buying PeopleSoft at that point?
For starters, PeopleSoft and IBM would have had to finish negotiating the contract for this ambitious partnership. That's right, there was never an actual contract in place for all of this. I think that this came as a surprise to Oracle when they completed the PeopleSoft acquisition to discover that they didn't actually have to do anything to unwind the deal with the IBM :-)
Then, after a signing a contract, there would have been a lot of work to do. This would have been the equivalent of what Oracle is doing with Fusion, but with more challenges to making it work. Here's a few of the key problems that would have needed to be overcome:
- No common ownership. It's hard enough to get large groups of people to work together when they work for the same company, but with two separate companies it is much, much harder. Even within Oracle, progress on Fusion applications was slow until Thomas Kurian took over Fusion applications in addition to Fusion middleware.
- No customer buy-in. PeopleSoft customers weren't asking for a conversion to WebSphere, so the fact that doing that could have helped PeopleSoft stay independent wouldn't have meant much to them, especially since the cost of moving to whatever a "PeopleSoft built on WebSphere" would have been significant.
- No executive buy-in. This is related to the previous point, but it's worth calling out separately. If Oracle had walked away and the deal with IBM had gone through, and PeopleSoft customers got put through the wringer as part of WebSphere move, all of the PeopleSoft project teams would be put in the awkward position of explaining to their management why these additional costs and headaches were happening. Essentially they would need to "sell" the partnership internally to their own management team. That's not a fun conversation to have.
Looking Backward
Looking back on the whole thing 5+ years later and it's interesting to see how things played out.
Oracle ended up being the ones to deliver PeopleTools 9, except that they took it one step further and made it backwards-compatible with existing applications and called it PeopleTools 8.50.
Oracle also realized that really nailing a next generation application platform takes some time. They have definitely taken a few hits on this, since some of the
early estimates of when Fusion applications would ship were too optimistic, but I think that most PeopleSoft customers are fairly pleased with how things have turned out.
IBM probably wishes that they had actually signed a contract with PeopleSoft to make WebSphere the strategic platform (although I don't think you'd get them to
say that now). Although Oracle continues to support WebSphere as a web server platform for PeopleSoft applications, we don't see anyone that is not already an IBM shop starting to use WebSphere for PeopleSoft.
Labels: 2010, History, IBM, Oracle, PeopleSoft
2010 | history | ibm | oracle | peoplesoft
This is great stuff. I'm ex-PSFT, worked on Fusion Middleware for 5 years...and your "refrigerator" comment was spot-on.
What's it gonna take you to blog again. More trips to the Hopyard?
Ah, the Hopyard. :) Brings back memories of testing PeopleSoft 5.1 out in Pleasanton. Good times ...