A million gallons of ink and four gazillion pixels have been spilt detailing or pontificating on Oracle's acquisition of Sun Microsystems.
My colleague Gordon Haff, for example, has nicely laid out Oracle's plans for Sun (as best as can be known this soon). So I'll be brief...
Oracle's interest in taking Sun's assets forward is at the very high end of expectations. In the almost-a-year since it became clear that Sun was shopping itself around, first to IBM then to Oracle, we've heard the direst of predictions. "Oracle only wants Sun to kill MySQL" and "they'd be lucky if 25 percent of Sun remains after the deal closes" are just two examples. In fact, now that the deal has closed, Oracle has laid out plans to retain most of Sun's key assets and lines of business. Java, Solaris, Sparc, storage, OpenOffice--Oracle has said clearly that it will continue these. In most cases, it's promised to invest more than Sun was investing. Sure, that's not necessarily a huge challenge, given the rolling waves of defunding that encroached on many of these projects over the years that Sun struggled. But it's exactly the opposite of what many expected.
Co-opetition continues but with much more leverage. Oracle will continue to work with IBM, Hewlett-Packard, SAP, Dell, EMC, Red Hat, VMware, and even Microsoft. These companies have too many joint customers and too many billions of dollars of revenue at stake to truly "go their own way." They have to work together--but not every day, in every account. Oracle will increasingly offer an all--or mostly--Oracle portfolio to customers that are amenable to it. While partners remain hugely important to Oracle's business, they'd darn well better be providing what the customer perceives as significant value, or else Oracle will seek to edge them out and own their erstwhile share of mind, market, and wallet. This isn't unique to Oracle. IBM, Cisco Systems, and others are doing exactly the same thing. But we expect Oracle to play this angle particularly well and aggressively.
Enterprise vs. lightweight. Oracle is making a clear distinction between enterprise software (Oracle DBMS and Fusion middleware, for example) and lightweight software infrastructure more suited to Web development. Some of Sun's software assets and technologies (most notably, Java, and maybe some parts around identity) will be pulled into Oracle's enterprise camp. But most of Sun's middleware assets--including those like MySQL, GlassFish, and NetBeans that Sun regarded as its crown jewels--will be developed and sold as more lightweight, Web-focused tools.
Open vs. open source. Oracle's model of "open" isn't the same as Sun's. In particular, it doesn't include nearly as much open source. Oracle does some open source, and the Sun acquisition brings in many more open-source assets and contributions to which Oracle can now lay rightful claim. But Oracle utterly declines to bow to the "everything on an open-source foundation" religion that Sun adopted in recent years. Key parts of Oracle's stack--the Oracle DBMS, its Fusion middleware, etc.--are decidedly closed source and are almost certain to remain that way. Oracle's model is much closer to that of IBM: encourage open source philosophically, leverage it where possible, support it where it's pragmatic to do so, but make sure to collect a toll for core pieces of "the stack."
Discipline and value. Oracle will be enormously more disciplined about the investments it makes (and doesn't make). It will also be much more disciplined and aggressive about seeking payment for value delivered. No more "build it, and they will come" daydreams. Oracle is a business, and an aggressive one. It has plenty of resources to invest--something Sun lacked in recent years--and it means to invest in a lot of Sun's areas of operations. But the investments it makes--in MySQL, in Solaris, in whatever--are not necessarily those that Sun would have made, nor with quite the same goals. To Sun's barbarian tribes, Oracle is the relentless, conquering Roman Empire.
Oracle is no longer a software company. It's an Integrated IT company. IT is maturing. It's no longer about "best of breed" point-products, or owning just one "category." De facto, enterprise customers value integration above almost all other virtues; IT businesses respond by having most systematic overall portfolios. I recently discussed how traditional boundaries and definitions have gone by the wayside as the IT industry rapidly consolidates. The new Oracle is part and parcel of this trend. Oracle now has a full-spectrum offer: servers, storage, networking, software, and services.
The fact that Oracle has decided to do most of what Sun was doing, only better, is bad news for many other vendors. Its breadth now rivals that of HP and IBM. And that means Oracle is going to be a serious (or more serious) competitor on many fronts.