Douglas Heintzman of IBM wrote,
” Apache also promotes a licensing regime that lends itself to innovation and participation of well-resourced organizations and has a higher comfort level with many corporate customers.
The question of licensing models is for some a key issue that merits examination. The LibreOffice community works with a copyleft regime. This is partly because, when that community forked the code from the OpenOffice project, it was the only licensing regime available to them, and partly because the core of that community comes from Linux vendors that are very comfortable with, and have had prior success with, copyleft licensing. This does not mean that they have failed with permissive licenses; it just means they have more experience, and a greater comfort level, with copyleft regimes.
Copyleft licenses rely on a viral mechanism to enforce disclosure of code modifications. Basically, if you are benefiting from the code, you contractually must disclosure any modifications or enhancements you make.”
Compare that with an article written earlier (2003) by the same authour:
“There is a general consensus that for the most part these licenses do not represent a barrier to the integration of OSS into business solutions and will likely have little, if any, impact on the success of OSS one way or the other.”
Clearly Heintzman does not get FLOSS. The GPL, for instance is a licence, not a contract, so one it not “contractually obliged to do anything”. One is permitted to copy by a licence from the creators under the conditions laid out by the GPL. OpenOffice.org ships under a mixture of licences for different parts of the code, reflecting its long history and huge number of contributors.
He never does get around to explaining why IBM chose Apache/ASL licensing except to state that IBM chose it. He certainly does not explain why IBM went with the code contributed to Apache instead of the code forked to LibreOffice and the greater numbers of contributors if they were interested in “community”. OpenOffice.org has yet to make an ASL release while LibreOffice is chugging away making release after release and doing well while OpenOffice.org is still under code review years later.
Is IBM too proud to admit that it made a mistake? It’s not too late to undo some things. So far, it looks as if IBM and Oracle’s move has come to nothing. OpenOffice.org is still not a full-fledged Apache project. They are still fiddling away trying to create the website. The latest status report includes:
“Most Important To Address:
1. Completion of the IP-review portions of the incubation checklist
2. A Successful Podling Release
3. Increase size and diversity of the active development community”
They are now doing nightly builds for developers and they have some committers but the whole project seems to have overwhelmed Apache. It seems like a high price to pay for stripping out copyleft code.
” We believe that all category-x copyleft code has been removed.
Development focus has shifted to reparing our initial Polding Release (release 3.4)”
Meanwhile, LibreOffice has made 7 releases, tuned up lots of code, added lovely features and has a huge base of contributors, nearly 400, (and more details on code clean-up)… Judge the licence like a healthy, growing tree, by the fruit of it.
I recommend LibreOffice over OpenOffice.org any day. It was a good ride for OpenOffice.org but SUN really didn’t set it free and Oracle and IBM have wrecked the rails.
UPDATE Here’s another article about progress or lack of it at Apache: Update on Apache OpenOffice
Yes, “Removal of Copyleft” is prominent. Yes, it looks like some version of the Munich Migration… It will get somewhere eventually but there does not seem to be any advantage over the energetic advancement of LibreOffice. Is it a case of cutting off the nose of OpenOffice.org in order to spite the face of it?
