Speculation about a possible spin-off of Sun’s hardware business by Oracle may not be so wide of the mark after all. The reason: Oracle’s initial acquisition interest only extended to some of Sun’s software assets.

Also, HP spent months doing due diligence over a possible bid of its own but backed out at the last minute without ever submitting an offer, so there’s a chance of it emerging as a buyer should Oracle opt to shed any parts of Sun in future.

Those are some of the things to be gleaned from Sun’s detailed disclosure to the SEC today of the events leading up to its deal with Oracle (HP’s identity is not disclosed – it is mentioned in the filing only as “Party B”. But one person who had a ringside seat to the action has told us that HP was actively engaged for months, and that despite widespread speculation, Cisco never showed any serious interest.)

The last-minute nature of Oracle’s bid for Sun in mid-April has received plenty of attention. The sheer speed with which it moved on the final weekend left IBM looking very flat-footed.

But what’s clear now is that this was not Larry Ellison’s preferred route. Back in mid-March he proposed buying what Sun describes as “certain of our software assets,” along with taking a minority equity stake. That idea was rejected.

When he finally bought the whole company, Mr Ellison gave two justifications. One was that he wanted to get his hands on Java, Solaris and some of Sun’s other software. The other was that owning the hardware division as well would leave Oracle better placed, as a vertically integrated company, to deal with the way many enterprise customers will want to buy their technology in future.

Based on the new information today, the “vertical integration” argument sounds increasingly opportunistic. Of course, Mr Ellison has since said he is “definitely not going to exit the hardware business,” but such declarations of intent often have a habit of changing as circumstances, and strategic objectives, shift.

As for HP: In the only explanation given for its lack of a firm offer, Sun says of the discussions late last year that “pursuing a transaction in the near term was not optimal for Party B at that time.” That isn’t surprising. Mark Hurd has a lot riding on successfully integrating the EDS acquisition. Attempting a second big consolidation play at the same time, in servers, would be a risk.

All of this suggests that, at some point, a deal between HP and Oracle for some of Sun’s assets could make sense. But for now, both companies are likely to keep their heads down as they focus on the immediate integration work at hand.

Copyright The Financial Times Limited 2018. All rights reserved.