| |||||||||||||
The "White Label Office" release candidateThe "White Label Office" release candidatePosted Dec 23, 2011 0:24 UTC (Fri) by cesarb (subscriber, #6266)In reply to: The "White Label Office" release candidate by hamjudo Parent article: The "White Label Office" release candidate
> As a long time subscriber to lwn.net, I'd like to say that I am entertained by this.
I'd guess the correct popular expression would be "pass the popcorn".
But I am more annoyed than entertained by this situation. It is like watching a slow-motion trainwreck, where while you are rooting for it to narrowly avert the impending disaster, you cannot see how it could escape.
> how many people will care remains to be seen.
I can see at least two classes of people who would still care.
* Those who did not get the memo, and still think "www.openoffice.org" is the place to go for a free office suite.
There are probably more.
The question is, are there enough people in these classes to keep viable such a large project, especially when a more advanced yet very similar alternative is available?
(Log in to post comments)
The "White Label Office" release candidate Posted Dec 23, 2011 7:28 UTC (Fri) by rahulsundaram (subscriber, #21946) [Link]
I think you left out a third class
* IBM
The "White Label Office" release candidate Posted Dec 23, 2011 10:54 UTC (Fri) by mjw (subscriber, #16740) [Link]
I think people underestimate how big and pragmatic IBM is. IBM does everything at least three times. And they have no problem pulling a whole group of developers out of a project without any notice and/or just switching to a completely different code base. IMHO it is nice if you get the backing of some part of IBM, but it isn't something you can rely on if you want to make sure your project is sustainable.
Back in the day I was one of the founders of Harmony (a now defunct project that we started as a bridge between the GNU and Apache free java efforts [1]), IBM contributed to GNU Classpath and gcc/gcj (and still distributes and supports that code through the IBMS's AIX Toolbox product) and also based one of their java runtime VMs (JikesRVM) on GNU Classpath code, these are different groups within IBM though, that just happen to use and contribute to the same code base without much coordination it seems. But another part of IBM (the OTI group) also had written some core java library code, that group was very aggressively pushing everything towards an ASL license, even claiming they didn't understand and couldn't use any GPLed code [2]. When the Harmony effort failed because the community wasn't that interested in a non-copyleft core java library seeing that GNU Classpath and gcj was that much ahead and Sun eventually adopted the GNU Classpath license for their OpenJDK effort [3], they simply pulled out completely and let the project die (though Google did grab some of the [non-copyleft] core library code and used that in their android effort, but they never contributed anything back). Now IBM joined the OpenJDK effort and publishes all their contributions under the GPL (although this again seems a new group within IBM because none of their new contributions are based on either their GNU Classpath work, nor based on their Harmony corelib work). Then there is also IBM's work on Eclipse, which produces among others a java compiler as alternative to the one in OpenJDK (this time under yet another EPL license, which is sadly GPL incompatible and regarded as a category B license by Apache, but can be used as a standalone source to bytecode compiler). IBM also maintained another (now also defunct) java source to bytecode compiler project called jikes, written in c++ under a special case IBM Public License.
[1] https://gnu.wildebeest.org/blog/mjw/2005/05/08/harmony/
The "White Label Office" release candidate Posted Dec 23, 2011 11:37 UTC (Fri) by rahulsundaram (subscriber, #21946) [Link]
I very well understand IBM may not be pig headed about Apache Openoffice and might jump ship, cut their losses and figure out how to work with the winning project. IBM's response when Red Hat's LVM2 won over their choice of EVMS was classy
https://lwn.net/Articles/14816/
However they are currently pushing hard for Apache Openoffice and I would say there are the primary if not the only reason it even happened and hence not a factor that should be ignored at this point.
EVMS tools Posted Jan 1, 2012 10:59 UTC (Sun) by gvy (guest, #11981) [Link]
> IBM's response [...EVMS...] was classy
Hm, and I haven't heard of EVMS UI tools running on top of LVM ever since (mind you, I used EVMS since 2003 or so and I'm *still* actually using libevms wrapped for guile as a part of ALT Linux installer -- worked on by Sergey Bolshakov for the most part, btw).
Am I missing something or was it so much for the spoken intent?
The "White Label Office" release candidate Posted Dec 23, 2011 12:12 UTC (Fri) by boudewijn (subscriber, #14185) [Link]
"I think people underestimate how big and pragmatic IBM is. IBM does everything at least three times. And they have no problem pulling a whole group of developers out of a project without any notice and/or just switching to a completely different code base. IMHO it is nice if you get the backing of some part of IBM, but it isn't something you can rely on if you want to make sure your project is sustainable"
That's probably true for all big companies. Dirk Hohndel warned about that in his Desktop Summit keynote. And I'm quite convinced that he back then already knew that Intel would pull out of MeeGo and was just warning us about that. It sounds weird, a big company handing out hundreds of devices, asking people to start writing code for a platform that they had already binned. Just as weird as the Nokia pavillion at the 2011 Mobile World Congress which hadn't been adapted to the new Windows Phone party line yet.
In short, if you want your project (or small company) to survive, it pays to have lots of little baskets for your eggs, widely spread out.
The "White Label Office" release candidate Posted Dec 23, 2011 16:53 UTC (Fri) by rsidd (subscriber, #2582) [Link]
What is the disaster that you foresee?
If the disaster is to Apache OO and/or Team OOo, I guess the answer is "who cares"? If you foresee a disaster to the larger free software community, I disagree -- none of this will foresee LibreOffice, which is already more than a year ahead of the other projects.
The only hope for Apache OOo is to merge with LibreOffice. And it is Apache who should make that effort. Otherwise, the future is with LibreOffice, and it looks bright enough to me.
Already many people I know use Google Docs, so the old hold that Microsoft Office had is gone. And if Google Docs is "good enough", LibreOffice is much more than "good enough".
|
|||||||||||||