451 CAOS Theory 
A blog for the enterprise open source community
Why Oracle’s donation of OpenOffice disappoints
Jay Lyman, June 2, 2011 @ 2:17 pm ETWhile Oracle deserves some praise for its donation of OpenOffice.org code to the Apache Foundation, it is disappointing again to see a legitimate open source market contender that has been marginalized by miscommunication and mismanagement of the project by a large vendor.
OpenOffice.org, warts and all, was probably the most significant competition for Microsoft Office for years and in many ways demonstrated the advantages of open source, helping usher in wider use of it, as well as greater usability. OO.o was in fact my reason for originally investigating and moving to open source software more than a decade ago. Regardless of past mismanagement of community and technology, that competitive factor has been diminished greatly since Oracle took ownership of OO.o. Now, after prompting a fork — as has been the case with a number of open source projects that fell to Oracle with its Sun acquisition (OpenSolaris-Indiana, OO.o-LibreOffice, Hudson-Jenkins), Oracle is again turning to a broader open source foundation to ‘free’ the project. It shouldn’t be surprising given our research into the balance of control and community, where we see a preference among both users and vendors for the ‘foundational’ approach that is typically less encumbered by real and perceived issues of control.
But by not making this move sooner, Oracle has again demonstrated that it does not appreciate or accept the broader community benefits of open source software. It ties open source investment and development directly to monetary value, meaning it is focused mainly on Linux and MySQL. Oracle should be commended for its honesty here, given its indication that it will contribute and support open source when it bolsters Oracle’s bottom line. However, the company is failing to tie its own success in open source with the success of the larger communities, which begs the question, is Oracle limiting the commercial opportunity for the open source projects on which it is focused by diminishing the community opportunity for projects it is leaving alone?
I might have more enthusiasm for OO.o as an Apache project, but I am somewhat skeptical for OO.o because of the current inclusion and use of LibreOffice in popular Linux distributions. This is how I came to use LibreOffice, and I’ve found it quite sufficient for my document, PDF, spreadsheet and other office suite needs. I would be glad to see a reunification of OO.o and LibreOffice and despite complex issues such as licensing, it is encouraging to see the leaders of LibreOffice and the Apache Foundation coming together toward a positive outcome.
Back to Oracle, the company again deserves credit for its positive and meaningful contributions to open source software, particularly MySQL and Linux, which would not have nearly the enterprise credibility it does without longtime, first-class treatment and support from Oracle. However, Oracle continues to demonstrate that despite how far open source has come in the enterprise, there are still large and powerful forces in the industry that do not fully understand open source software’s potential.
Comments (7) Categories: Software
Me thinks you heap too much praise on Oracle vis a vis FOSS. They have only had the Sun acquisition 18mos. In that time the only update of OO has come out and only 2 of MySQL both of which were bug releases. The heavy lifting for both projects happened about the time of the Sun acquisition of these tools. That’s almost 4 years now.
Oracle has shown an unbending lack of appreciation of the FOSS concept. Their unfailing near sexual desire for a quarterly ROI is not in keeping with the precepts that most FOSS projects work under. It also manifests itself in Oracle’s inability to keep the OO.o team together after the acquisition had been completed.
While Oracle has taken some serious missteps with regards to the various open source projects it has acquired via Sun your representation of the progress of MySQL under Oracle is wrong. Development of MySQL stalled badly at Sun and while many many MySQL developers have left Oracle version 5.5 delivered significant performance improvements, thanks largely to the InnoDB development work carried out within the company.
JohnMC, Matt: My personal observation of MySQL under various owners is:
– MySQL 5.1 was released during Sun ownership, much delayed. This release was already delayed from first marketing announcements before Sun ownership. Sun CTO (software) Bob Brewin personally oversaw the final months to actually get it released. So in essence, the problems with MySQL were created during MySQL AB and if anything, we should credit Sun for helping to fix the issues. (Same is true for other topics, like the closed source backup controversy: the truth is exactly opposite to how it was reported at the time.)
– MySQL 5.5 was released during Oracle ownership. This is the first MySQL release with good multi-core scalability, among other things. Almost all of that work was done under Sun, started immediately after Sun acquisition and Sun moved several of its senior high performance engineers to complement the MySQL AB team. (Remember that multi core scalability was more important to Sun early on, due to Sparc CMT architecture.) So credit for the great MySQL 5.5 release goes to Sun.
– So what about Oracle? I’d say Oracle has done more good than harm for MySQL. As a short summary Oracle has done poorly – and in some case been outright hostile – with interfacing with the MySQL community, but continued the good work of Sun when it comes to producing code. Here credit goes to Tomas Ulin who is just a great engineering manager (he ran MySQL Cluster team since Ericsson days) but also to Oracle for employing him and his team.
The point is that Oracle rightfully deserves credit for its positive contributions and participation in open source communities, and it also rightly deserves criticism for its failure to fully appreciate and leverage community — both of which you’ll find here.
JL
[...] Meanwhile Jay Lyman offered a perspective from The 451 Group, including disappointment that Oracle didn’t move quicker, [...]
[...] it can be to maintain open source software community projects, involvement and goodwill in Oracle’s difficulties with some of the technology and community it acquired with Sun [...]
[...] it can be to maintain open source software community projects, involvement and goodwill in Oracle’s difficulties with some of the technology and community it acquired with Sun [...]